Touch

touch
TOUCH
AND
MASSAGE
in Early
Child
Development
EDITED BY
Tiffany Field, PhD

Touch Research Institutes
University of Miami School of Medicine
L I B R A RY O F C O N G R E S S C O N T RO L N U M B E R : 2 0 0 4 1 0 6 1 2 3
Main entry under title:
Touch and Massage in Early Child Development
Johnson & Johnson Pediatric Institute, L.L.C.
ISBN 0-931562-30-9
©Johnson & Johnson Pediatric Institute, L.L.C. 2004
Printed in the United States of America. All rights reserved. Except as permitted under
the Copyright Act of 1976, no part of this publication may be reproduced or distributed
in any form or by any means or stored in a database or retrieval system without prior
written permission of the publisher.
The opinions and data presented by the authors are their own and are not necessarily
those of Johnson & Johnson Pediatric Institute, L.L.C. or the editors. Accurate referencing
is the responsibility of the writer and editor.
TOUCH
AND
MASSAGE
in Early
Child
Development
EDITED BY
Tiffany Field, PhD

Published by: Johnson & Johnson Pediatric Institute, L.L.C.

TA B L E O F C O N T E N T S
C O N T R I B U T O R S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv
P R E F A C E . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
I N T R O D U C T I O N : T H E I M P O R TA N C E O F T O U C H . . . . . . . . . . . . . . . . ix
Tiffany Field, PhD
S E C T I O N I . M AT E R N A L T O U C H A N D T O U C H P E R C E P T I O N . . . . . 1
Chapter 1: Fetal Responses to Foot and Hand Massage of Pregnant Women . . . . . . . . 3

Miguel Diego, MA; Tiffany Field, PhD; Maria Hernandez-Reif, PhD
Chapter 2: Touch Perception in Neonates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

Maria Hernandez-Reif, PhD; Tiffany Field, PhD; Miguel Diego, MA
Chapter 3: Maternal Touch Effects on Infant Behavior . . . . . . . . . . . . . . . . . . . . . . . . . 39

Hiselgis Perez, PhD; Jacob L. Gewirtz, PhD
Chapter 4: Touching During Mother-Infant Interactions . . . . . . . . . . . . . . . . . . . . . . . . 49

Dale M. Stack, PhD
Chapter 5: Differences Across Cultural Groups in Mothers’ Nonverbal

Teaching Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
Mary Martini, PhD
S E C T I O N I I . T H E R A P E U T I C A P P L I C AT I O N S O F T O U C H
I N P R E G N A N C Y, L A B O R A N D P O S T B I R T H . . . . . . . . . . . . . . . . . . . . . . . . 97
Chapter 6: Pregnancy, Labor and Infant Massage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99

Tiffany Field, PhD; Maria Hernandez-Reif, PhD; Miguel Diego, MA
Chapter 7: Skin-to-Skin Care for Breastfeeding Difficulties Postbirth . . . . . . . . . . . . 115

Gene Cranston Anderson, PhD, FAAN; Sheau-Huey Chiu, PhD, RN, PNP;
Barbara Morrison, PhD, FNP, CNM; Maria Burkhammer, RN, CDONA;
Susan Ludington-Hoe, CNM, PhD, FAAN
S E C T I O N I I I . T O U C H I N P R E M AT U R E I N F A N T S . . . . . . . . . . . . . . . . 137
Chapter 8: Tactile Stimulation of Neonatal Intensive Care Unit

Preterm Infants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
Lynda Law Harrison, RN, PhD, FAAN
Chapter 9: Preterm and Full-term Infant Massage in China . . . . . . . . . . . . . . . . . . . . . 163

Zhu Jianxing, MD, MSc
Chapter 10: Optimizing Growth and Bone Mass in Premature Infants:

Are Diet and Physical Movement the Answers? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
Laurie J. Moyer-Mileur, PhD, RD, CCD
S E C T I O N I V. E F F E C T S O F M A S S A G E O N S L E E P,
R E L A X AT I O N A N D W E L L - B E I N G . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
Chapter 11: Massage Therapy and Sleep-Wake Rhythms in the Neonate . . . . . . . . . 183
Sari Goldstein Ferber, PhD
Chapter 12: Massage, Relaxation and Well-Being: A Possible Role for

Oxytocin as an Integrative Principle? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
Kerstin Uvnäs-Moberg, MD
S E C T I O N V. M A S S A G E T H E R A P Y F O R O R P H A N S A N D
P E D I AT R I C P R O B L E M S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
Chapter 13: Massage Therapy With Preterm Infants and Children

With Chronic Illnesses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
Herminia L. Cifra, MD, MSc, MHA, FPPS; Melanie N. Sancho, MD, FPPS
Chapter 14: Massage Therapy Effects on Illness Symptoms in Infants

Living in Ecuadorian Orphanages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227
Vonda K. Jump, PhD
Chapter 15: Massage Therapy for Pediatric Problems . . . . . . . . . . . . . . . . . . . . . . . . . 239

Tiffany Field, PhD; Maria Hernandez-Reif, PhD; Miguel Diego, MA
CONTRIBUTORS
Gene Cranston Anderson, PhD, FAAN Herminia L. Cifra, MD, MSc,

Edward J. and Louise Mellen MHA, FPPS
Professor of Nursing Professor
Frances Payne Bolton School of Nursing University of the Philippines
Case Western Reserve University College of Medicine
10900 Euclid Avenue Head, Pediatric Intensive Care Unit
Cleveland, Ohio 44106, USA Philippine Children’s Medical Center
Quezon Avenue
Quezon City, Philippines
Maria Burkhammer, RN, CDONA
Project Coordinator Miguel Diego, MA
Frances Payne Bolton School of Nursing
Research Associate
Case Western Reserve University
10900 Euclid Avenue
Cleveland, Ohio 44106, USA
P.O. Box 016820
1601 N.W. 12th Avenue
Sheau-Huey Chiu, PhD, RN, PNP Miami, Florida 33101, USA
Assistant Professor
College of Nursing Sari Goldstein Ferber, PhD
University of Akron
Clinical Psychologist
Mary Gladwin Hall
Lecturer at Nursing Department
209 Carroll Street
Faculty of Social Welfare and
Akron, Ohio 44325, USA
Health Studies
University of Haifa
Mount Carmel
Haifa 31905 Israel
Sackler School of Medicine
Tel Aviv University
Tel Aviv, Israel
Contributors v
Tiffany Field, PhD Vonda K. Jump, PhD

Director Senior Research Associate
Touch Research Institutes Early Intervention Research Institute
University of Miami School of Medicine Utah State University
P.O. Box 016820 Logan, Utah 84322-6580, USA
Miami, Florida 33101, USA
Susan Ludington-Hoe, CNM,
PhD, FAAN
Jacob L. Gewirtz, PhD Carl W. and Margaret Davis Walter
Professor Professor of Pediatric Nursing
Department of Psychology Frances Payne Bolton School of Nursing
Florida International University Case Western Reserve University
PC 543 10900 Euclid Avenue
Miami, Florida 33199, USA Cleveland, Ohio 44106, USA
Lynda Law Harrison, RN, PhD, FAAN Mary Martini, PhD

Professor Associate Professor
University of Alabama School of Department of Family and Consumer
Nursing Sciences
University of Alabama at Birmingham University of Hawaii at Manoa
NB415 Miller Annex
1530 3rd Avenue South 2515 Campus Road
Birmingham, Alabama 35294-1210, Honolulu, Hawaii 96822, USA
USA
Barbara Morrison, PhD, FNP, CNM

Maria Hernandez-Reif, PhD Assistant Professor
Research Associate Professor Frances Payne Bolton School of Nursing
Touch Research Institutes Case Western Reserve University
University of Miami School of Medicine 10900 Euclid Avenue
P.O. Box 016820 Cleveland, Ohio 44106, USA
Miami, Florida 33101, USA
vi Contributors
Laurie J. Moyer-Mileur, PhD, Kerstin Uvnäs-Moberg, MD

RD, CCD Professor of Physiology
Research Associate Professor Swedish University of Agricultural
Pediatrics Director Science
Center for Pediatric Nutrition Research Department of Animal Physiology
University of Utah Karolinska Institute, SE-171
Salt Lake City, Utah 84112, USA Stockholm, 17177 Sweden
Hiselgis Perez, PhD Zhu Jianxing, MD, MSc

Statistical Research Coordinator Chairman, Department of Pediatrics
Florida International University Chief, NICU
Planning and Institutional Effectiveness Vice Chairman, Pediatric Academic
PC 543 Committee
Miami, Florida 33199, USA Xinhua Hospital
Shanghai Second Medical University
227 Chong Qing Nan Road
Melanie N. Sancho, MD, FPPS
Shanghai 200025 China
Pediatric Intensive Care Unit
Philippine Children’s Medical Center
Quezon Avenue
Quezon City, Philippines
Dale M. Stack, PhD

Associate Professor
Department of Psychology/Centre for
Research in Human Development
Concordia University
The Montreal Children’s Hospital
7141 Sherbrooke St. West (PY-170)
Montreal, Quebec H4B 1R6, Canada
PREFACE
The Johnson & Johnson Pediatric Institute, L.L.C., is proud to sponsor this
groundbreaking compilation of studies on the power of touch in childhood growth
and development.
Astounding advances in health and medicine have improved the lives of millions
of children in the last century. Vaccines prevented childhood diseases; antibiotics
cured those who became ill; sophisticated new technologies and gleaming
incubators helped physicians keep tinier and younger premature infants alive.
In the midst of this exciting progress, however, the importance of old-fashioned

touch—holding, massaging, rocking and soothing children—was often dismissed,
forgotten or even negated. Touching was not sterile. It was not scientific. It was not
modern. However, a handful of parents, clinicians and researchers continued their
commitment to this more traditional therapy. Scientific research has validated their
commitment as touch has now been shown to have numerous therapeutic benefits.
Today, a growing body of research from around the world shows that human touch
is a powerful contributor to infants’ ability to thrive and grow. Rigorous scientific
studies show that
• Even a few minutes a day of touching measurably increases a

newborn’s health and vigor, from body weight to bone mass
• During skin-to-skin contact, a mother’s chest naturally adjusts to

provide exactly the warmth and comfort that her baby needs
• Touch and massage significantly reduce infant crying and distress
• Infant massage under accepted guidelines can increase weight gain

in premature infants
and much more.

viii Preface
Each of the studies included in this book establishes a powerful case for touch.
In bringing this wide-ranging research together, the Johnson & Johnson Pediatric
Institute’s goal is to remind us that in this high technology world, there is still
room for low technology therapies that can increase the health and happiness
of infants and families around the world.

INTRODUCTION
The Impo r t a nc e o f To u c h
The act of touch fulfills the basic human need to feel safe, comfortable and loved.
Touch is also an intrinsic factor in child development, but despite touch’s importance,
in recent years “touching” has been tabooed in the American school system because
of fears of sexual and physical abuse. Elementary and high school teachers have
been mandated not to touch children, and even children as young as preschoolers
cannot be touched in many private nursery schools because of potential accusations
of sexual abuse following increased publicity about abuse in schools.1 Although
the rates of sexual and physical abuse have climbed steadily despite this mandate,
teachers uphold this rule from fear of lawsuits by parents and losing their jobs. In
a study we conducted on the amount of teacher touching in preschool classrooms,
we observed significant decreases in teachers touching nursery school children as
the children became older (from infant to toddler to preschool age) including less
holding, hand-holding, stroking and carrying, and resulted in increased aggressive
behavior.2 Being concerned about the extremely low levels of touching, we then
provided demonstrations and instructions to the teachers in an attempt to increase
the levels of touching. Following this intervention, the teachers increased their level
of touching (including holding, hugging, kissing, hand-holding and caregiving
touch), although touching was still relatively infrequent following this intervention.
Others have tried to increase touching in nursery schools by having the children
give each other back rubs.
The implications of limited touch for children involve significant effects on their
growth, development and emotional well-being. Extreme cases can be seen in the
Romanian orphanages where children have achieved half their expected height due
to extreme forms of touch deprivation. Their cognitive and emotional development
has also been significantly delayed by the lack of physical stimulation. Other data
suggest that touch deprivation in early development may contribute to violence
in adults.
x Introduction
In a cross-cultural study, cultures in which there was more physical affection toward
infants had lower rates of adult physical violence and vice versa.3 The amount of
touching that occurs in different cultures is highly variable. For example, touching
behavior between couples was observed in cafes in several countries.4 Couples
were observed sitting at tables in cafes for 30-minute periods, and the amount
of touching between them was recorded. Among the highest touch cultures was
France (110 times per 30 minutes) and among the lowest was the United States
(2 times per 30 minutes). The high touch cultures had relatively low rates of
violence while the low touch cultures, in turn, had extremely high rates of youth
and adult violence. In a 1994 report by the Centers for Disease Control and
Prevention, the homicide rate per 100,000 population was 1 in France and 22 in
the United States.5
The cafe data suggest that French families may provide more touch stimulation for
their children, and conversely, American families provide less. Studies we conducted
on Paris and Miami playgrounds suggested that these cultures differed in the
touching of preschoolers by parents, in peer touching among preschoolers and in
peer touching in young adolescents.6,7 For the Paris parents, touching occurred 43%
of the time as opposed to 11% of the time for Miami parents. Affectionate touch
between the preschool children occurred 23% of the time on the Paris playground
and only 3% of the time on the Miami playground. In contrast, aggressive touch
occurred 37% of the time on the Miami playgrounds and only 1% of the time on
the Paris playgrounds.
Because infants and toddlers are touched so infrequently in nursery school, we have
begun instructing teachers on how to massage infants and toddlers. In our studies,
we have noted that infants/toddlers who are massaged before naptime at preschool
are less irritable, go to sleep faster and sleep more soundly. Thus, teachers are very
happy massaging them. Parents have also been noted to help reduce irritability and
enhance sleep by massaging their infants before bedtime. These are potential ways
to reintroduce touch into our society.
This volume highlights the importance of touch across early development. Diego
et al present research data on how the fetus responds to tactile stimulation, while
Hernandez-Reif et al discuss the data on newborns’ responses to tactile stimulation.
In a chapter by Perez and Gewirtz, studies are presented on how young infants
respond to different types of touch. Infants’ responses to touch during early
Introduction xi
interactions are addressed by Stack, and data on early interactions in different

cultures are presented by Martini.
In other chapters, Field covers therapeutic applications including pregnancy and

labor massage, Anderson et al address kangaroo care for neonates, Harrison reviews
tactile stimulation for preemies, Moyer-Mileur presents data on moving the limbs of
preemies, and Goldstein Ferber examines the effects of massage therapy on full-term
infants. These are followed by a chapter on oxytocin as a possible underlying
mechanism for massage effects by Uvnäs-Moberg and chapters on massage for
preemies and pediatric conditions by Cifra, massage for infant orphans by Jump,
and massage for several pediatric problems including attention and behavior
problems, pain, neuromuscular conditions, autoimmune disorders, and immune
dysfunction by Field et al.
The power of touch as a therapeutic modality is highlighted by the data from these
studies. Many of them also suggest the importance of training parents to massage
their infants for growth and development and their chronically ill children for
helping treat their medical conditions. Methods to provide adequate touch could
effectively be introduced in delivery units, children’s hospitals and schools. They are
easy to learn and have been noted to help not only infants and children but also
reduce the stress levels and stress hormones of the parents and teachers providing
the stimulation. Touching and being touched are comforting, growth-producing
and healing experiences that need to be supported by our medical and educational
communities. The data presented in this volume are compelling and will help
advance touch in our world.
Tiffany Field, PhD

Miami, Florida
xii Introduction
Ref ere n c e s
1. Mazur S, Pekor C. Can teachers touch children anymore: physical contact and its value in
child development. Young Children. 1985;40:10-12.
2. Field T, Harding J, Soliday B, Lasko D, Gonzalez N, Valdcon C. Touching in Infant, Toddler
and Preschool Nurseries. Early Child Development and Care. 1994;98:113-120.
3. Prescott JW. Affectional bonding for the prevention of violent behaviors: neurobiological,
psychological and religious/spiritual determinants. In: Hertzberg LJ, Ostrum GF, Field JR,
eds. Violent Behavior. Great Neck, NY: PMA Publishing; 1990:95-124.
4. Jourard SM. An exploratory study of body accessibility. British Journal of Social and Clinical
Psychology. 1966;5:221-231.
5. Centers for Disease Control and Prevention. National Center for Injury Prevention and
Control International Comparisons of Homocide Rates in Males 15-24 Years of Age, 1988-1991.
Atlanta, Ga: Centers for Disease Control and Prevention; 1994.
6. Field T. Preschoolers in America are touched less and are more aggressive than preschoolers
in France. Early Child Development and Care. 1999;151:11-17.
7. Field T. American adolescents touch each other less and are more aggressive toward their
peers as compared with French adolescents. Adolescence. 1999;34:753-758.
Touc h a n d Ma s s a g e i n Ea rl y Ch ild De ve lop me nt
SECTION I.
M AT E R N A L TO U C H A N D
TOUCH PERCEPTION
CHAPTER 1:
F E TA L R E S P O N S E S TO
FOOT AND HAND MASSAGE
OF PREGNANT WOMEN
Miguel Diego, MA
Tiffany Field, PhD
Maria Hernandez-Reif, PhD
Ab s t ra c t
The fetus responds to extrauterine stimuli depending on such factors as the nature
and intensity of the stimuli and gestational age. The study discussed in this chapter
assessed fetal responses to maternal foot and hand massages during ultrasound
examinations conducted midgestation (mean, 19.9 weeks). The study involved
80 women, divided equally into a foot-massage substudy and a hand-massage
substudy. Control subjects did not receive massages, but rather rested quietly during
the observation periods. Results indicated that foot massaging was associated with
a 16% increase in fetal activity. In contrast, hand massaging elicited only a 10%
rise in fetal activity, which did not significantly differ from the activity of the
control group. These results suggest that the hands may be less innervated than the
feet or that the foot- versus the hand-massage procedure may have triggered distinct
reflex arches and physiological responses. Further research is needed to explore the
mechanisms involved in the fetal response to maternal tactile stimulation.
4 Section I. Maternal Touch and Touch Perception
Introd uc t i o n
The fetus is often believed to develop in total isolation from the outside world.
Throughout gestation, the mother and fetus share a common environment, which
is affected continuously by the external world. Extrauterine light and sound cross
the maternal abdomen, uterus and embryonic membranes, thereby reaching the
amniotic fluid and the fetus.1,2 Furthermore, vibration applied to the mother’s
abdomen is transmitted mechanically across the intrauterine environment.1
Similarly, external tactile stimulation has been shown to produce changes in
the mother’s physiology3,4 and biochemistry5,6 that may affect the intrauterine
environment indirectly, even if such stimulation is not applied directly to the
mother’s abdomen.
Extrauterine stimulation has been used in the antepartum assessment of fetal well-
being.7,8 Vibratory stimulation has been used to test fetal hearing,9 and habituation
to vibratory stimulation has been used to test fetal function.7,10 Fetuses who respond
to vibratory stimulation have been reported to have better biophysical profiles.11,12
Fetal responses to stimulation depend largely on the properties of the stimuli and
the maturation state of the sensory system being tested. The fetus responds to
airborne auditory stimuli as early as 28 weeks gestation,13 which is only 4 weeks
after the development of the fetal auditory structures.14 Furthermore, the fetal
response depends largely on the intensity of the stimulus.15-17
Similarly, responses of the fetus to tactile stimulation also depend on the intensity
of the stimuli. Even though responses by the fetus to tactile stimuli have been noted
very early in gestation,18,19 fetal responses to vibratory stimulation are only apparent
after 26 weeks gestation.16,18,20 Taken together, these findings suggest that vibratory
stimulation applied to the mother’s abdomen may not be perceived because those
nerve endings have not yet been developed and myelinated.21 Alternatively, the
vibratory stimulation that is perceived by the fetus may be primarily auditory, and
not tactile in nature, and would be perceived only after sufficient development of
the fetal auditory system.
Other forms of external stimulation that may elicit fetal responses have also been
studied. Acupuncture delivered to the mother at acupoints SP-6 (located on the
Chapter 1: Fetal Responses to Foot and Hand Massage of Pregnant Women 5
tibia, above the tip of the medial malleolus) and LI-4 (located on the hand,
between the 1st and 2nd metacarpal bones) affect umbilical-artery blood flow to
the fetus during the third trimester.22 When delivered to the mother at accupoint
BL-67 (located on the fifth toe), acupuncture results in decreased fetal heart rate
(FHR)23 and increased movement (FM) in fetuses during the third trimester.23,24
The stimulation of specific acupuncture points is believed to activate specific
nervous system pathways.25
Massaging highly innervated areas, such as the feet and hands, may also stimulate
fetal activity. Midwives have reported anecdotally that massaging the Achilles tendon
of pregnant women during late gestation may induce contractions. Massaging the
feet and hands has been shown to reduce heart rate, respiration rate and systolic and
diastolic blood pressure,4,6 decrease levels of norepinephrine and epinephrine, “and
attenuate the response of cortisol to a stressful procedure.” These changes in the
mother’s physiology and biochemistry following massage therapy may contribute
to the increase in fetal activity. Data have also shown that massaging other highly
innervated areas of the mother’s body—nipples26 and breasts27—results in uterine
contractions during pregnancy.
Cur re n t St u d i e s
In this chapter, we review the studies we have done with several of our colleagues
on the effects of massaging the mother’s feet and hands on fetal activity.28 The
extremities were selected because the feet and hands are highly innervated and
because fetal responses to maternal foot acupuncture have been noted.23,24 These
studies were conducted midgestation, inasmuch as fetal responses to stimuli have
been noted as early as 14 weeks gestation,18,19 yet fetal activity responses to vibratory
stimuli occur only after 26 weeks gestation.16,18,20 Furthermore, the responses of
fetuses to acupuncture have been studied only after 33 weeks.23,24 As such, we and
our colleagues were interested in determining if foot and hand massage of the
pregnant woman would contribute to fetal activity. Finding increased fetal activity
would provide preliminary support for using massage as an assessment of fetal
responsivity prior to 26 weeks gestation.
METHODS
From a prenatal clinic at a large, urban, university hospital, 80 pregnant women in

midgestation (mean, 19.85 weeks) were recruited for this study.28 Pregnant women
were excluded if they met any of the following criteria: carrying more than one
fetus; smoked cigarettes or drank alcohol; carrying a fetus that had demonstrated
any abnormalities during an ultrasound; had a complicated pregnancy; or had been
prescribed any medications other than vitamins. The first 40 mothers, participants
in Study 1, were randomly assigned to a foot-massage or a control group. The
remaining 40 mothers, participants in Study 2, were randomly assigned to a
hand-massage or a control group.
The fetal activity assessment and stimulation procedures occurred immediately prior
to a standard clinical ultrasound examination. The pregnant women completed the
State/Trait Anxiety Inventory (STAI) immediately before and immediately following
the ultrasound. These women were then monitored by ultrasound for baseline fetal
activity while lying on an examination table on their left side for 3 minutes. This was
followed by a 3-minute massage or control period and a 3-minute poststimulation
observation period. All participants were assessed between 10:00 am and 2:00 pm
in the same ultrasound examination office. No significant differences were noted on
assessment times or days.
FETAL-ACTIVITY ASSESSMENTS
Fetal activity was assessed using interval recordings on a real-time ultrasound scanner
fitted with a single Doppler transducer applied to the mother’s abdomen.28 Every
effort was made by the technician to visualize the entire fetus by manipulating the
transducer and adjusting the zoom control (Figure 1). At 20 weeks gestation, the
majority of fetuses were visualized fully. If the fetus could not be visualized fully, the
technician focused on the head, torso and arms, as well as on the upper-leg regions
to assess leg movement. Every 3 seconds, the researcher was prompted by a tape-
recorded cue (heard through an earphone) to record the type of fetal activity that
was occurring. All behaviors occurring between tones were ignored. The 3-second
interval was chosen arbitrarily for its ability to generate an easy-to-determine total
([60 seconds/3 seconds = 20 observations] x 3 minutes = 60 observations). The
researcher recorded the following: a single limb movement, characterized by any
movement of any one extremity; multiple limb movements; exclusive of torso
movements (i.e. more than one extremity, including the head, moving at any one
time); gross body movements with or without single or multiple limb movements;
or no movement at all. The percentage of time the fetus engaged in each
movement category was calculated by dividing the total number of movements by
the total number of observation units. Total amount of movement was calculated
by subtracting “1” from the percentage time that the fetus did not show any
movement. Interrater reliability, calculated on 10% of the samples from 2 of the
observers, revealed adequate kappa values ranging from 0.82 to 1.00.
Figure 1. Ultrasound image of the fetus.
F O OT- M A S S A G E P RO C E D U R E
The 3-minute foot-massage procedure was performed by the therapist as follows:

stretching the feet for 5 seconds by holding the ankles with one hand and pulling
the tops of the feet toward the therapist with the other hand; squeezing the feet
intermittently for 20 seconds each, starting at the heels and moving toward the toes
by holding both feet with the thumbs on the top and the other fingers on the soles
of the feet; applying pressure with both thumbs to the arch of each foot (one foot
at a time), and then using the thumbs to rub between each toe using an up-and-
down motion (toward the web of the toe, and then toward the end of the toe) for
30 seconds on each foot; kneading the bottom of each foot for 20 seconds each,
and then the top of each foot for 20 seconds each, one foot at a time; flexing and
extending each foot, and then flexing and extending all toes on each foot, one foot
at a time for 3 seconds each; squeezing each foot for 5 seconds, moving from the
heel to the toes, one foot at a time; and, finger-stroking the entire top and bottom
of each foot, using all 5 fingers, one foot at a time, for 7 seconds.28
HAND-MASSAGE PROCEDURE
The 3-minute hand massage (Figure 2) was performed by the therapist as follows:
holding the wrist with one hand, and with the other hand pushing against the
fingers for 5 seconds each; with both hands, stroking the tops of the mother’s hands
briskly for 10 seconds each; with a thumb on the top of each hand and the other
Figure 2. Hand massage procedure.

fingers on the palms, compressing both hands intermittently, starting at the pinkies
and moving toward the thumbs for 10 seconds each; finger-stroking the entire back
of each hand, from the fingers to the wrists for 10 seconds each; using the thumbs
and forefingers to squeeze each finger on each hand for 20 seconds each; stretching
each finger by encircling each finger completely, gently pulling fingers away from
the hand, for 20 seconds each; applying pressure intermittently to one hand at a
time from the center to the edge of the bottom of the hand for 10 seconds each;
and, holding the wrist with the left hand, and with the right hand gently pulling
the hand away from the wrist for 5 seconds each.28
CONTROL-GROUP PROCEDURE
In the control procedure for the foot and hand massage the pregnant women were
asked to lie on their left side while resting quietly for 3 minutes.28
Results and Discussion
The mothers in the massage groups and in the control groups on average reported
feeling less anxiety following the massage or rest session.28 The reduced anxiety was
probably a function of the visit for ultrasound. Mothers in this study, each of
whom had normal, uncomplicated pregnancies, received immediate feedback from
the ultrasonographer on the health of their fetus. In an earlier study we had shown
that mothers who received feedback during ultrasound examinations reported less
anxiety than mothers who did not receive feedback.29 In addition, simply resting
on the left side for 12 minutes may lower anxiety levels.
Even though vibratory stimulation has failed to elicit fetal movement before 26
weeks gestation,16,18,20 the foot massage elicited fetal activity as early as 20 weeks
gestation.28 The fetuses of mothers who received a 3-minute foot massage showed
increased activity, whereas the fetuses of mothers in the control group did not
(Figure 3). The increased fetal activity is consistent with increased fetal movement
observed following foot acupuncture in older fetuses.23,24 Given that the feet are
highly innervated, stimulating the feet of a pregnant woman may trigger changes in
fetal activity via the activation of reflex arches, changes in maternal physiology and
other unknown mechanisms. These findings are consistent with research that show
that massaging highly innervated areas produces increased uterine contractions,26,27

as well as with studies that show that massaging the feet lowers HR, BP and
respiration.4 Changes in HR, HR variability and galvanic skin conductance have
been noted to affect fetal movement as early as 20 weeks gestation.31
Figure 3. Percent time fetal movement occurred in response to foot and hand massage.
Fetuses of mothers receiving a 3-minute foot massage showed a significant increase in
fetal movement.28
While the hand-massage procedure elicited a 10% increase in fetal movement,

compared with a 16% increase for the foot-massage procedure, this change in fetal
activity was not significantly different from that shown by a control group (Figure
3). These results are perplexing, as we had expected that fetuses of women who
received the hand-massage procedure would show a similar response to fetuses
of women who received a foot massage. Research into the effects of exercise by
pregnant women indicates that the responses of the fetus are dependent on the
intensity of the exercise.32,33 Although light exercise produces slight or no changes
in fetal response, moderate exercise produces increased FHR and fetal breathing,
while strenuous exercise produces the opposite effect. Perhaps the hands are less
innervated than the feet, which might explain why massaging the hands elicited
a less-intense response than massaging the feet. Alternatively, the foot-massage
procedure, versus the hand-massage procedure, may have triggered distinct reflex
arches and physiological responses. Using longer observation periods for the foot
and hand massages may have provided information on the onset and the duration of
the fetal responses to the foot-massage condition. Thus, the lesser effect of the hand
massage versus the foot massage may have been an artifact of the short observation
period. Further study may help elucidate the mechanisms whereby massaging the
feet of a pregnant woman results in increased fetal activity.
Conclusions
Additional research is needed to determine whether massage therapy given

to pregnant women elicits fetal activity at other gestational stages. Additional
assessments of fetal state, such as FHR, HR and variability, may enhance the
validity of these findings. Furthermore, assessing maternal physiological variables,
such as galvanic skin response and uterine contractions, in relation to fetal activity
across the different types of stimulation may help suggest possible underlying
mechanisms.
Ackn ow le d gm e nt s
We would like to thank the pregnant women who participated in these studies.
We would also like to thank Jean Pierre Lecanuet, John Dieter and Fawwaz Salman
for their help in designing the study and Julia Beutler, Shay Largie and Mercedes
Redzepi for their help in collecting the data. This research was supported by funding
provided by the Johnson & Johnson Pediatric Institute, L.L.C. and a National
Institute of Mental Health (NIMH) Senior Research Scientist Award (#MH00331)
and an NIMH Merit Award (#MH46586) to Tiffany Field.
Ref ere n c e s
1. Gerhardt KJ, Abrams RM. Fetal exposures to sound and vibroacoustic stimulation. Journal
of Perinatology. 2000;20(8, pt 2):S21-S30.
2. Kiuchi M, Nagata N, Ikeno S, Terakawa N. The relationship between the response to
external light stimulation and behavioral states in the human fetus: how it differs from
vibroacoustic stimulation. Early Human Development. 2000;58:153-165.
3. Delaney JP, Leong KS, Watkins A, Brodie D. The short-term effects of myofascial trigger
point massage therapy on cardiac autonomic tone in healthy subjects. Journal of Advanced
Nursing. 2002;37:364-371.
4. Hayes J, Cox C. Immediate effects of a five-minute foot massage on patients in critical care.
Intensive and Critical Care Nursing. 1999;15:77-82.
5. Field T. Maternal depression effects on infants and early interventions. Preventive Medicine.
1998;27:200-203.
6. Kim MS, Cho KS, Woo H, Kim JH. Effects of hand massage on anxiety in cataract surgery
using local anesthesia. Journal of Cataract and Refractive Surgery. 2001;27:884-890.
7. Devoe LD. Nonstress testing and contraction stress testing. Obstetrics and Gynecology Clinics
of North America. 1999;26:535-556.
8. Smith CV. Vibroacoustic stimulation. Clinical Obstetrics and Gynecology. 1995;38:68-77.
9. Ke X, Gu Z, Wu R. Vibratory acoustic stimulation test in fetal hearing monitor [in Chinese].
Chinese Journal of Otorhinolaryngology. 1995;30:264-266.
10. Kuhlman KA, Burns KA, Depp R, Sabbagha RE. Ultrasonic imaging of normal fetal response
to external vibratory acoustic stimulation. American Journal of Obstetrics and Gynecology.
1988;158:47-51.
11. Inglis SR, Druzin ML, Wagner WE, Kogut E. The use of vibroacoustic stimulation during
the abnormal or equivocal biophysical profile. Obstetrics and Gynecology. 1993;82:371-374.
12. Sarinoglu C, Dell J, Mercer BM, Sibai BM. Fetal startle response observed under
ultrasonography: a good predictor of a reassuring biophysical profile. Obstetrics and
Gynecology. 1996;88:599-602.
13. Querleu D, Renard X, Boutteville C, Crepin G. Hearing by the human fetus? Seminars
in Perinatology. 1989;13:403-420.
14. Pujol R, Lavigne-Rebillard M. Sensory and neural structures in the developing human
cochlea. International Journal of Pediatric Otorhinolaryngology. 1995; 32:S177-S182.
15. Kisilevsky BS, Muir DW. Human fetal and subsequent newborn responses to sound and
vibration. Infant Behavior and Development. 1991;14:1-26.
16. Kisilevsky BS, Muir DW, Low JA. Maturation of responses elicited by a vibroacoustic
stimulus in a group of high-risk fetuses. Maternal/Child Nursing Journal. 1990;19:239-250.
17. Lecanuet JP, Granier-Deferre C, Busnel MC. Fetal cardiac and motor responses to octave-
band noises as a function of central frequency, intensity and heart rate variability. Early
Human Development . 1988;18:81-93.
18. Hepper PG, Shahidullah BS. Development of fetal hearing. Archives of Disease in Childhood.
1994;71:F81-F87.
19. Hooker D. The Prenatal Origin of Behavior. Lawrence, Kan: University of Kansas Press; 1952.
20. Kisilevsky BS, Muir DW, Low JA. Maturation of human fetal responses to vibroacoustic
stimulation. Child Development. 1992;63:1497-1508.
21. Kandel ER, Schwartz JH, Jessel TM. Principles of Neural Science. 4th ed. Columbus, Ohio:
McGraw-Hill/Appleton & Lange; 1993.
22. Zeisler H, Eppel W, Husslein P, Bernaschek G, Deutinger J. Influence of acupuncture on
Doppler ultrasound in pregnant women. Ultrasound in Obstetrics & Gynecology. 2001;17:
229-232.
23. Neri I, Fazzio M, Menghini S, Volpe A, Facchinetti F. Non-stress test changes during
acupuncture plus moxibustion on BL67 point in breech presentation. Journal of the Society
for Gynecologic Investigation. 2002;9:158-162.
24. Cardini F, Weixin H. Moxibustion for correction of breech presentation: a randomized

controlled trial. JAMA: the Journal of the American Medical Association. 1998;280:1580-1584.
25. National Institutes of Health. Acupuncture. NIH Consensus Statement 1997. 15:1-34.
26. Christensson K, Nilsson BA, Stock S, Matthiesen AS, Uvnäs-Moberg K. Effect of nipple
stimulation on uterine activity and on plasma levels of oxytocin in full term, healthy,
pregnant women. Acta Obstetricia et Gynecologica Scandinavica. 1989;68:205-210.
27. Gantes M, Kirchhoff KT, Work BA Jr. Breast massage to obtain contraction stress test.
Nursing Research. 1985;34:338-341.
28. Diego MA, Dieter JN, Field T, et al. Fetal activity following vibratory stimulation of the
mother’s abdomen. Developmental Psychobiology. 2002;41:396-406.
29. Field T, Sandberg D, Quetel TA, Garcia R, Rosario M. Effects of ultrasound feedback
on pregnancy anxiety, fetal activity, and neonatal outcome. Ultrasound Feedback.
1985;66:525-528.
30. Nolte J. The Human Brain: An Introduction to Its Functional Anatomy. 4th ed. St Louis, Mo:
Mosby; 1999.
31. DiPietro JA, Irizarry R, Hawkins M, Costigan K. The psychophysiology of the maternal-fetal
relationship. Paper presented at: The International Conference on Infant Studies; July 17,
2000; Brighton, England.
32. Manders MA, Sonder GJ, Mulder EJ, Visser GH. The effects of maternal exercise on fetal
heart rate and movement patterns. Early Human Development. 1997;48:237-247.
33. Marsal K, Lofgren O, Gennser G. Fetal breathing movements and maternal exercise. Acta
Obstetricia et Gynecologica Scandinavica. 1979;58:197-201.
CHAPTER 2:
TOUCH PERCEPTION
I N N E O N AT E S

Tiffany Field, PhD
Miguel Diego, MA
Ab s t ra c t
Sucking and grasping behaviors occur in utero. This chapter presents data suggesting
that such behaviors exhibited by newborns are more than just reflexive responses.
Studies involving the handling and mouthing of objects by neonates indicate that
newborns clutch or grasp objects presented to them and suck on objects that provide
no nutritional value. Newborns, it has been determined, distinguish between
objects, perceiving differences in hardness, texture, weight, shape and temperature.
Thus, it appears that infants explore objects actively with their hands and mouths,
using behaviors similar to those exhibited by adults, attempting to perceive the
properties of specific objects. Future studies should explore further the touch and
perception abilities of neonates.
Touch Pe rc e pt io n i n Ne o na t e s
The newborn baby invariably responds to touch. Stroking the newborn’s face often
leads to head turning. Mouth opening and closing (or “rooting”) may follow, and if
a finger or nipple touches the newborn’s lips, typical responding includes opening
the mouth, closing the lips around the object and sucking. These behaviors exhibited
by the neonate are observed routinely during feedings and in the course of neonatal
assessments, such as with the Brazelton Neonatal Behavioral Assessment Scale.1
Newborns also respond when their hands are touched. Placing a finger in the palm
of the newborn results in the fingers of the neonate closing or grasping. In a study
by Nagy and Molnar,2 decreased heart rate was observed following grasping, which
was interpreted as neonatal calming. However, heart-rate decelerations have also
been used as measures of “attending” in the very young infant.3,4 Therefore, perhaps
grasping serves multiple functions, as does sucking, the goals of which can be for
feeding, for calming (sucking on a pacifier or finger) or for exploring objects.
To date, these responses to touch by the neonate have been defined in the literature
as “reflexes,” or involuntary responses to stimulation,5,6 implying little or no
intentional behavior on the part of the newborn. Interestingly, examinations by
ultrasound have revealed grasping of the umbilical cord by the fetus at 18 weeks
gestation,7 and thumb-sucking at as early as 15 weeks gestation,8 suggesting that
sucking and grasping are not new behaviors for the neonate, or that the neonate
comes into the world with sucking and grasping experiences.
In this chapter, we question the traditional view that grasping and sucking responses
to touch are merely reflexive responses performed by the newborn. For example, if
sucking is reflexive then why does the neonate stop sucking when he or she is full
or suck more vigorously when hungry? That the responses are variable may suggest
that they are not completely reflexive. Perhaps reflexes exhibited by the neonate are
the precursors of voluntary, intentional “touching back.” Or, perhaps stimulating
the palm or mouth of the neonate elicits the respective reflexive response initially,
but the prolonged behaviors that follow are more “voluntary” by the neonate.
Another view that we question is that grasping and other limited motoric abilities,
such as lack of finger movement, impede newborns’ explorations of properties of
objects with their hands.9,10 A counterview—and the thesis of this chapter—is that
Chapter 2: Touch Perception in Neonates 17
neonates may explore many properties of objects through grasping and mouthing
because of one or more of the following:
• These behaviors are exercised in utero, beginning during the second

trimester of pregnancy, and therefore control of these responses by
newborns may be underestimated at this time.
• Skin or sense receptors are in place and are functional prenatally, so

that the processing of touch stimulation should be possible postnatally,
if not earlier.
• The nerve pathways in the hand and mouth are well-developed, lending
further support to the view that these sensory parts of the newborn are
capable of exploring and picking up information regarding touch.
In this chapter, data are presented regarding the differential grasping and sucking
responses of the neonate to varying object properties. Data are also presented that
document neonatal habituation (becoming less attentive to a familiar object) and
dishabituation (becoming more attentive to a novel object placed in their hands),
as well as behaviors that appear to be precursors of active touching or exploring
by the newborn.
The Touch “Organs”
The largest fetal “organ,” and the first to develop, is the sensory-rich skin, or the
sense of touch. By the 30th week of fetal development (10 weeks before birth),
neurological structures and pathways are developed sufficiently for the fetus to
perceive pain.11 Studies indicate that during painful or stressful procedures, such as
during intrauterine needling or surgery, fetuses and preterm infants exhibit elevated
levels of stress hormones and increased activity, which some interpret as pain
responses.12,13 However, because pain is a subjective experience, and because of
the accepted view that responses to touch by fetuses and neonates are “reflexive
responding,” the perception of pain by fetuses continues to be debated.14,15
Studies done in humans and animals also reveal the responding of neonates to heat
and to the skin being touched with fine-hair stimuli. (For a review, see Fitzgerald
and Jennings.16) Interestingly, these studies53,54 found that the younger the offspring,
the lower the threshold for responding and the more exaggerated the neonate’s
response is to stimulation, suggesting that the newborn is in a more excitable, or
perhaps more vigilant state initially than is the adult.16 Also noteworthy are the
facts that there are approximately 17,000 skin receptors in each hand17 and that the
hands and mouth have the highest concentrations of sensory receptors compared
with other skin areas, further supporting the view that the neonate is capable of
“feeling” in these areas.18 Taken together, the hypothesis that neonates “feel” or
perceive when their hands or mouths are stimulated appears more plausible than
that their responses to touch or pain are reflexive or nonvoluntary.
Touch Perception
The exploration or manipulation of objects with fingers, hands or mouth is defined

as active touch, which is one mechanism believed to underlie perception or how
information about the environment is learned or detected.19 Because the manual
dexterity of newborns is not well-developed, some researchers of infancy believe
that newborns are incapable of feeling and distinguishing object properties with
their hands. Specifically, neonates are not expected to perceive texture, shape and
weight, because, based on adult studies (see Klatzky and Lederman, 199355 for a
review of adult studies), the perception of these properties requires active fingering,
handling and/or lifting of objects, which are not behaviors that are observed readily
in the newborn infant.9
When information is detected through the activity of the hands, fingers and/or
mouth, it is referred to as haptic (relating to or based on the sense of touch)
perception. Unlike the other sensory systems, the haptic system is unique in that
it can be used to explore and to alter the environment19: That is, the hands and
mouth can not only feel the properties of an object, but they can change them,
whereas the eyes can only see objects and the ears can only hear sounds, for
examples. Active touching assumes purposeful exploration and involves input from
skin receptors and kinesthetics (movement of muscles and joints),20 such as lifting
and holding, squeezing, finger tracing, prodding or poking. Finger tracing and
prodding are activities not typically performed by neonates.
Traditionally, the behavioral repertoire of the neonate has been described as

limited to holding or grasping and perhaps opening and closing of the fingers.9
In observations that we made with our colleague Largie, after placing an object in
the hands of 1- to 2-day-old newborns, we documented that some neonates turned
their hands or wrists (back-and-forth hand movements) and lifted their hands to
their faces or mouths.21 Other haptic behaviors reportedly exhibited by neonates
include squeezing18,22 and slow finger stroking of objects.23 Although limited or
crude in terms of the information that can be provided to the neonate, these
grasping behaviors are likely to aid the infant in abstracting information about
objects: For example, temperature (cold or warm) can be detected through simple
skin contact with the object, such as grasping with the palm or pressing the object
to the lips. Hand squeezing and sucking could reveal the substance of the object
(whether it is rigid or soft), as well as its texture (rough or smooth) and shape.
Turning the hand or wrist, as well as lifting the hand holding the object to the
face or mouth area, could facilitate perception of the weight of the object.
What follows is a detailed review of the research of these behaviors exhibited by

newborns in relation to objects: Mouthing studies are reviewed first, followed
by hand studies.
MOUTHING STUDIES
The infant’s mouth and tongue comprise a highly specialized system used to ingest
nutrients, express emotions and promote self-comfort, such as sucking on a breast,
pacifier, finger or thumb. A MEDLINE® search covering the last 35 years revealed
more than 1200 studies published on these topics in relation to infant sucking.
However, few studies were found regarding the mouthing of objects by infants for
the purposes of exploring, perceiving or learning about properties of objects (eg,
texture, hardness, temperature).
Fewer than one dozen studies emerged when the same literature search was
conducted for the perception of objects by newborns through mouthing or handling
activities. (However, more than 100 entries were found when the keyword included
“infants” rather than “neonates.”) We review the scant literature on neonatal haptic
perception in what follows immediately and in the next section.
One of the earliest studies of newborn “active oral touching,” conducted by Lipsitt
and Kaye (1965),24 revealed that newborns altered their sucking rate as a function
of the type of stimulus. Newborns were given a rubber nipple, a rubber tube or a
combination of rubber nipples and tubes on which to suck over consecutive trials.
Overall, the newborns sucked the nipples longer than they did the tubes, which
suggested that they preferred the nipples and, therefore, that they must have
perceived the nipples and tubes as different object properties. Almost 20 years later
(1983), using polygraph recordings, Rochat examined the sucking responses of
newborns and 1- and 4-month-old infants presented nipples that varied in shape
(elongated versus round) and material (rubber versus a brass component attached
to the surface of the rubber).25 His findings included increased scanning with the
infants’ tongues and lips and decreased sucking from the newborn period to
4 months of age, which suggested developmental or age differences in mouthing
(or exploratory) behaviors. Additionally, the newborns differentiated 2 nipple types
(round and round with a hole), as indicated by their sucking preferences, and the
1-month-old infants detected the material changes. These findings support the view
that from early in development infant sucking and mouthing behaviors serve
exploratory functions and facilitate the detection of object properties.
In a follow-up study (1987), Rochat reported that 2- to 3-day-old newborns

applied more sucking pressure to soft objects than they did to hard objects.18
However, with their hands, they squeezed hard objects harder than they did soft
objects. In the same study, similar mouthing and handling behaviors were observed
for 2- and 3-month-old infants, although, interestingly, the squeezing responses
of these older infants were of lesser duration and frequency than those of the
newborns. Rochat concluded that neonatal mouthing and grasping are voluntary
behaviors, not just reflexive activities. In other words, if the younger infants in
these studies were merely responding to stimulation, and not attending to the
objects’ hardness or softness, then sucking and grasping behaviors should have been
comparable in duration and pressure, irrespective of the stimulus type.
Studies of young infants’ cross-modal (touch-to-sight) abilities revealed that

1-month-old infants recognized object properties visually that they had previously
mouthed on the basis of texture information, such as “nubby” versus “smooth,”26,27
or substance, such as “hard” versus “soft.”28 Within hours after birth, neonates
have also been noted to match the shape of an object previously mouthed with
its visual shape.29
Newborns’ mouthing responses to objects that varied in texture30 or temperature31

have also been studied. Infants of depressed mothers were also assessed. These
infants are of particular interest because they are reported to be less engaged in
exploring objects.32,33 Infants who show less exploratory behavior tend to have
cognitive deficits later in life, including lower IQs.34,35,50
In the “texture” study, middle socioeconomic status pregnant women (mean age,
29 years) were assessed for depression until 12 depressed and 12 nondepressed
mothers agreed to participate.30 After delivery, the mothers were contacted and a
laboratory visit was scheduled. The newborns averaged 12 days of age at testing
(7 males/5 females in each group), and all were born full-term and medically
uncomplicated. Held by one of 4 “blinded” experimenters on his or her lap, the
neonates were presented with nubby- and smooth-textured fingertips to suck. The
“nubby” stimulus, a sterile rubber fingertip (Swingline® Rubber Fingertips) of the
kind worn by bankers and others for counting bills or turning pages, was covered
with .10 cm nubby protrusions spaced every .20 cm. The “smooth” stimulus was
simply the nubby fingertip turned inside out, which revealed a smooth, even
surface like that of a rubber glove. Each experimenter wore the nubby fingertip and
the smooth fingertip on the middle finger and index finger of his or her hand—
counterbalanced, so that sometimes the nubby fingertip was on the experimenter’s
middle finger and sometimes on the index finger.
The smooth and nubby fingertips were presented to the neonates for mouthing one
at a time, in an alternating fashion across 6 trials, so that each infant had 3 chances
to explore the texture of each fingertip orally. Each trial started with the newborn’s
first suck of the fingertip and ended when the infant released the sucking grip. The
amount of time (in seconds) spent sucking the smooth and nubby fingertips was
recorded and analyzed.
• Interestingly, the infants of depressed and nondepressed mothers sucked

the smooth-textured fingertips significantly more than they sucked the
nubby-textured fingertips, suggesting that they detected the different
textures because they preferred the smooth (perhaps more familiar)
surface. However, across the 6 trials, the neonates of the depressed
mothers sucked about 50% as long as those of the nondepressed
mothers. Moreover, the newborns of the depressed mothers spent
33% less time sucking the nubby fingertip and approximately 60%
less time sucking the smooth fingertip compared with the newborns of
nondepressed mothers (Figure 1). These findings suggest the following:
• From sucking, neonates discriminated a smooth texture from a nubby

texture.
• Neonates preferred to suck smooth textures, which might have resulted

from a familiarity preference, such as sucking on a bottle nipple, pacifier
or finger. It was unlikely that any of these infants had explored a nubby
fingertip previously.
• Newborns of depressed mothers were less inclined to suck textures, as

evidenced by their lesser sucking, in particular, on the more-novel nubby
texture.
Figure 1. Newborns of depressed mothers sucked the “nubby” and “smooth” objects for
shorter periods than did the neonates of nondepressed mothers.30
In summary, by approximately 12 days of life, infants not only perceived differences

in texture through sucking, but they also controlled the amount of time they spent
sucking objects presented to them, supporting the view that sucking is a voluntary
behavior very much under the control of the very young infant. Also, it was
apparent that neonatal behaviors may have been influenced indirectly by maternal
depression via the mothers’ altered biochemistry during pregnancy. (See Field,31 for
a review of the effects of maternal depression on the newborn.)
In an effort to assess the newborns’ perceptions and discriminations of temperature,

the second study (M.H-R., T.F., M.D. and S. Largie, unpublished data, 2003)
examined neonates sucking a “cold-condition” nipple versus a “warm-condition”
nipple. In this study, women who had just delivered normal full-term infants
were visited in the neonatal unit of a university hospital, and those who consented
were assessed for symptoms of depression. Twenty-two depressed mothers and 20
nondepressed mothers and their full-term newborns (mean age, 1.5 days old) were
enrolled. All 42 neonates were presented with 8 alternating trials of a smooth bottle
nipple kept in refrigerated water (“cold condition”) and an identical nipple kept in
room-temperature water (“warm condition”). The nipples were replaced throughout
the study to maintain their temperature. To prevent the infants from sucking in
air, the experimenters pushed their fingers through the nipples and presented the
nipples on the fingers for the infants to suck. The testing occurred with the neonate
in an awake and alert state while lying on his or her back in a nursery bassinet. The
“blinded” experimenters first touched the newborns’ mouths with the nipples and
then inserted the nipples into the infants’ mouths. The number of sucks, which
served as the dependent measure, was recorded.
Interestingly, unlike in the “texture alone” study,30 in this temperature study the
infants of depressed mothers sucked more frequently (mean, 117 sucks) than
the neonates of the nondepressed mothers (mean, 79 sucks). Also interesting was
that the duration of sucking was affected by the temperature of the nipples on the
first trial. When the cold nipples were presented first, infants sucked longer on all
subsequent presentations, regardless of the mothers’ depression scores. Why a cold
stimulus led to greater sucking is unclear: Perhaps sucking a cold nipple was more
stimulating or soothing, much like sucking a sweetened pacifier reduces crying in
newborns undergoing painful procedures.36 Nevertheless, this question warrants
further investigation.
Although the duration of sucking increased for all presentations, if the first trial was
done with a cold nipple, neither group showed a preference for sucking one nipple
over the other. This outcome was puzzling and suggested one of two conclusions:
Either the newborns did not discriminate the cold nipple from the warm nipple,
or perhaps they did perceive the temperature difference, but simply did not prefer
sucking one longer than the other. Also interesting was that mothers who suffered
from more severe depressions, as indicated by higher depression scores on the
Center for Epidemiological Studies of Depression (CES-D) scale,37 had infants
who sucked the nipples for longer durations (Figure 2). The possible reason: The
neonates of depressed mothers were sucking for self-comforting purposes, and not
for exploratory motives, in that both the cold and warm nipples were of a smooth
texture. As is known from at least 3 studies of neonates, infants prefer to suck on
smooth textures and nipple shapes.18,24,30 At birth, infants of depressed mothers are
reported to be more irritable38,39 and have higher cortisol (“stress-hormone”) levels
than do infants of nondepressed mothers.31,40 Thus, the “sucking-for-comforting”
hypothesis seems plausible.
Figure 2. The sucking responses of neonates to cold- and warm-temperature nipples

based on the severity of mothers’ depression scores.(M.H-R., T.F., M.D. and S. Largie, unpublished data, 2003)
CES-D = Center for Epidemiological Studies of Depression scale.
The mothers in this study were also assessed on the Behavioral Inhibition Scale/
Behavioral Activation Scale (BIS/BAS).41 Commonly, individuals who exhibit
behavioral “inhibition” are characterized as anxiety-prone, and they often withdraw
from novelty. Contrastingly, those who display behavioral “activation” are more
impulsive and inclined to approach or engage in activities.42,43 Differing biochemical
profiles have also been reported for infants of intrusive mothers versus infants
of withdrawn mothers, with the intrusive group showing higher dopamine and
serotonin levels than the withdrawn group, and the withdrawn group showing
higher cortisol levels than the intrusive group.44 However, to date, no studies of
perception have been conducted with neonates of inhibited or withdrawn mothers.
We hypothesized that, if the mothers’ biochemistry influences neonatal behavior,
different sucking responses would be evident for infants of intrusive mothers versus
withdrawn mothers. (M.H-R., T.F., M.D. and S. Largie, unpublished data, 2003)
Of the 42 mothers enrolled in this study, 12 met BIS/BAS criteria for behavioral
inhibition (withdrawn) and 12 met criteria for behavioral activation (intrusive).
Ten of these 24 women were deemed nondepressed (6 intrusive; 4 withdrawn):
Hence, this measure examined the influence of inhibited mothers’ versus withdrawn
mothers’ effects on newborn sucking only. Analyses of the neonates of the “intrusive”
mothers revealed that these infants sucked the cold nipple more than the warm
Figure 3. Frequency of sucking on cold and warm bottle nipples by neonates of

depressed-intrusive mothers versus depressed-withdrawn mothers.(M.H-R., T.F., M.D. and S. Largie,
unpublished data, 2003)
nipple (Figure 3), which suggested that this subgroup perceived the cold nipple
as different from the warm nipple. One explanation for this interesting finding
is that neonates of intrusive mothers may be more reactive to stimulation and/or
are more stimulus-seeking because of their higher dopamine levels, although the
infants’ biochemistry profiles were not measured for this study.
If the infants of the intrusive mothers found the cold nipple more stimulating,
this would support the “stimulus-seeking” hypothesis and explain the findings.
However, because these are post hoc analyses based on small sample sizes, replication
studies will be required before conclusions can be drawn about the interactions
with objects by neonates of intrusive mothers. In any event, the fact that differences
were found in sucking preferences between these 2 groups suggests the need for
studying the perceptual abilities of neonates who might be predisposed to varying
biochemical profiles or inherited maternal traits.
HAND STUDIES
Findings gleaned from studies of adults form the bases for studies of infants’ hand
explorations and perceptions. The literature on adults reveals that they use explicit
finger and hand movements to detect specific aspects of objects.20,45,46 For example,
adults rub their fingers repeatedly on objects’ surfaces when exploring their
textures.20,45,46 However, to detect hardness of a substance, they press or squeeze
objects.20,45,46 For the perception of temperature, adults simply make contact with
the surface of the object, such as by placing a hand on the object’s surface.20,45,46
To detect general information about an object’s shape or size, they enclose or mold
their fingers around the object, but to detect precise shape, they trace the object’s
edges with their fingers.20,45,46 They determine an object’s weight by displacing or
lifting the object with one of their hands.20,45,46
Based on these studies of adults,20,45,46 as well as on our knowledge of infants’ limited

fine-motor movement control, Bushnell and Boudreau established a timetable for
when infants might detect the properties of specific objects.9 In particular, these
investigators stated that infants under 4 months of age should perceive temperature
and size information because the hand behaviors required to detect these stimuli
(holding or grasping) are within young infants’ behavioral repertoires. However,
detection of an object’s texture, weight and hardness (although hardness may be
detected from grasping or squeezing) is not expected to occur until between the
fourth and ninth months of life, as this is when infants are readily observed
scratching, rubbing, squeezing and waving objects. Importantly, at approximately
9 or 10 months of age, infants master the process of “sitting,” and thus are able to
support themselves while holding objects,47,48 which would allow them to hold an
object with one hand while the fingers of the other hand trace the object’s edges for
perception of exact shape.9 These studies suggest that infants may perceive some of
the properties of objects using different finger and hand movements than those
used by adults.
PERCEPTION OF HARDNESS: Studies of the handling abilities of very young infants

provide both support and refute the timetable proposed by Bushnell and
Boudreau.9 In the study by Rochat, a soft or hard small tube connected to an
air-pressure transducer was placed in the right hands of newborns and 2-month-old
and 3-month-old infants, and the frequency of squeezing each tube (as measured
by a pressure transducer) was recorded.18 Regardless of age, the infants squeezed the
hard tube more often than the soft tube. However, these infants showed different
grasping patterns—squeezing and releasing the soft tube, in contrast to clutching
the hard tube.
These findings suggest that the perception of hardness by hand may occur as early
as during the newborn period, and that infants show “exploratory” behaviors similar
to those shown by adults when detecting information about substance. However,
the fact that infants modulated their hand activities depending on the object’s
hardness suggests that newborns possess control of their hand behaviors. Moreover,
these findings contradict the view that neonatal grasping is a reflex response.
PERCEPTION OF TOUCH: Borrowing the experimental procedure proposed by

Rochat,18 Molina and Jouen compared the hand activities of newborns when
handling smooth and rough textures.22 As in the study by Rochat,18 pressure
transducers were inserted into rubber tubes.22 One tube was covered with cloth
(“smooth texture”), while the other tube was covered with small plastic pearls
that had been sewn onto cloth (“rough texture”). The cloth- or the pearl-covered
tubes were placed in the newborns’ right hands (N = 22), and hand pressure was
recorded. The findings of this study included the infants exhibiting more hand
activity during the presentation of the “smooth” tube than during presentation of
the “rough” tube, as well as that they applied continuous high-frequency pressure
and discrete low-frequency pressure when holding the rough tube. As with the
study by Rochat,18 these findings suggest that neonates modulated their hand
behaviors according to the property of the object.22 These findings22 also suggest
that the perception of texture may be achieved earlier than the 4-month time frame
noted by Bushnell and Boudreau,9 and that the neonates might have detected the
textures via squeezing the tubes, unlike the adults who used rubbing of their fingers
across the object’s surface. Perhaps in the absence of fingering activities, newborns
grasp and squeeze objects to explore and perceive their properties.
PERCEPTION OF SHAPE: A study by Streri, Lhote and Dutilleul examined newborns’

discrimination of 2 shapes (rectangular versus triangular) by hand.23 The procedure
consisted of presenting one of the 2 objects to one of the newborn’s hands over
consecutive trials until the infant’s object-holding time declined. This procedure is
known as “habituation.” The 24 infants were habituated in equal numbers with the
shapes. The duration of holding (in seconds) was recorded over the period of trials,
and, as expected, the neonates either held the object progressively less over the trials
or showed an habituation response. Habituation suggests that the infant becomes
familiarized or perhaps “bored” with the same issue or object. Subsequent to this
habituation, the neonates were presented with the alternate- (or novel-) shaped
object for 2 test trials, and holding time again was measured. If infants noticed
the novel shape, it was expected that they would hold the object longer, or
“dishabituate.” As expected, infants held the novel shape longer (compared with
the last 2 trials during habituation), suggesting that their newborn hands, in fact,
processed “shape” information.
If the findings by Streri, Lhote and Dutilleul23 are conclusive, they refute the time
line for the detection of shape by infants, as well as suggest that neonates use
exploratory procedures that differ from those described for shape perception by
adults. You will recall that adults trace the edges of objects with their fingers to
perceive shape.20,45,46 This behavior is not evident in infants until late in the first
year of life.9 Streri, Lhote and Dutilleul23 also videotaped their neonates during
the procedure: Coding of these videotapes revealed the newborn infants bringing
the objects close to their faces or their other hand, squeezing the object and some
evidence of slow finger-stroking. Perhaps one or all of these behaviors facilitated
the perception of the objects’ shapes by the neonates. Additional trials with
neonates, which study their finger and hand activities with varying object shapes,
are necessary to determine the underlying mechanisms involved in shape perception
by newborns.
PERCEPTION OF TEMPERATURE: Yet another study examined the perception of

temperature by neonates.49 Because newborns grasp objects, and temperature
detection requires at least minimal touching of the object, we hypothesized that
neonates would be able to differentiate cold tubes from warm tubes placed in their
hands. The tubes were small (4 cm in length x 1 cm in diameter), of the same
weight (5 g), filled with cold water or warm water and changed frequently to
maintain consistent temperatures throughout the testing. Forty newborns, who
had a mean age of 27 hours, participated in this trial. Twenty of these neonates
were born to depressed mothers, but all were born full-term and medically
uncomplicated. Testing occurred in the neonate’s hospital bassinet, which was
inclined to a 45° angle. Infants were awake during testing and had shown grasping
when the experimenter placed his or her finger in the neonate’s hands. Following
the tactile habituation procedure established by Streri, Lhote and Dutilleul,23 these
neonates were presented with one of the 2 “temperature tubes” to hold by hand.
Half the newborns were assigned to hold the warm tube, while the other half
held the cold tube, until the duration of holding declined with repeated trials.
Subsequently, the neonates were given the alternate-temperature tube to hold for
Figure 4. A newborn of a non-depressed mother showing hand activity with an

object during the 1st minute of a temperature trial: 1) the neonate holds his hands
at mid-chest prior to start of trial, 2) the vial is placed in the newborn’s hand and
he raises the right hand that contains the vial, 3) he lowers the right hand with the
vial after a few seconds and 4) raises the right hand with the vial again.
(Photos taken by Maria Hernandez-Reif, PhD).
1) No object in hand 2) Right hand with 3) Right hand with 4) Right hand with
object raised object lowered object raised again
2 test trials: For example, after holding the warm tube over numerous trials and
showing little interest in continuing to hold it, the infant was presented with the
novel cold tube and the duration of holding it was recorded. During the procedure,
the experimenters also coded the infants’ hand activities as “active exploring” (hand
opening, closing, squeezing, turning or lifting hand to mouth or face area) or
“passive exploring” (exhibiting only one of the “active-exploring” hand behaviors,
or if they displayed any of the behaviors ≤50% of the trials, or infants who showed
no hand activities during the procedure).
As expected, the amount of time spent holding the same-temperature tube declined
over repeated trials, but then, when presented with the alternate-temperature tube,
both groups of neonates (of depressed mothers and of nondepressed mothers)
held the novel-temperature tube longer, suggesting that these newborns detected
the different temperatures presented. It is important to note, however, that the
newborns of the depressed mothers differed in several important respects from
the neonates of the nondepressed mothers: First, the infants of the depressed
mothers took twice as long to habituate, which suggests that they were slower
in processing the temperature information by hand. In the literature on visual
habituation, a longer time to habituate has been associated with less optimal
cognitive functioning.50 Secondly, rather than showing a decline in the amount
of time they held the same-temperature tube during the first few trials, the
neonates of depressed mothers showed an increase in this amount of time. Again,
in the visual habituation literature, this is referred to as “sensitization.”50
Sensitization has been linked to an inability to control arousal, and has been
reported for infants who take longer to habituate.51,52 Sensitization may occur
because the infant finds the object stimulating or because the infant is already
aroused or excited and becomes more excited when stimulated. Infants who exhibit
this behavior may prove more reactive to stimulation, have poor self-regulatory
capacities and/or have difficulty focusing or attending. Another observed difference
was that the newborns of depressed mothers showed less-active touch, indicating
that perhaps they were less interested in exploring objects with their hands. Less
handling or manipulation of objects by infants “at risk” has been correlated with
longer-term cognitive deficits.35
PERCEPTION OF WEIGHT: Weight perception has also been assessed in newborns

of depressed and nondepressed mothers.21 Because neonates have been reported to
lift their hand to their face and mouth23 and because studies in the adult literature
discuss hand lifting as the mechanism underlying weight perception,20,45,46 we

hypothesized that neonates would perceive an object’s weight change. However,
because infants of depressed mothers have proven less exploratory,32,33 we also
hypothesized that infants of only nondepressed mothers would perceive the
weight change.
As in the study on “Perception of temperature,”49 the handling of objects was

assessed by presenting the same object over consecutive trials until the holding
response by the neonate declined. Twenty-seven full-term neonates born to
depressed mothers, and 23 full-term newborns of nondepressed mothers participated
in this trial (mean age, 43 hours). The stimuli consisted of one tube stuffed with
cotton and weighing 2 g (lightweight) and another tube stuffed with pellets and
weighing 8 g (heavyweight). Over consecutive trials, half the infants were presented
with the same lightweight tubes, while the other half were presented with the same
heavyweight tubes. When the amount of time spent holding the tubes decreased,
the groups of newborns were tested with the now-novel (opposite) weight tubes.
The handling activities of the neonates were coded as described in the “Perception
of temperature” study described earlier.49
Figure 5. A newborn of a depressed mother showing little hand activity with an

object during the 1st minute of a temperature trial: 1) the neonate holds his hands
at mid-chest prior to start of trial, 2) the vial is placed in the newborn’s hand and he
raises the right hand that contains the vial, 3) the vial remains in the inactive hand of
the neonate for the entire trial until he becomes irritable, cries and later drops the vial
(Photos taken by Maria Hernandez-Reif, PhD).
1) No object in hand 2) Right hand with 3) Right hand remains raised

object raised but inactive
A surprising 85% of the neonates of depressed mothers were coded as showing no

hand activity. Therefore, it was not surprising that, as a group, the newborns of the
depressed mothers failed to notice the novel weights. Contrastingly, 78% of the
neonates of the nondepressed mothers showed hand activity. The latter group
exhibited weight discrimination, which was demonstrated by longer holding times
on the novel weights.
Su mm a r y
Newborns clutch or grasp objects that are placed in their hands and they suck
objects that offer no nutritional value. For decades, these neonatal behaviors have
been defined as “reflexes,” or as “involuntary responses” to stimulation.
The studies reviewed in this chapter, to the contrary, suggest that grasping and
sucking objects appear to be “voluntary behaviors” of the newborn that serve as
exploratory behaviors for detecting the properties of the objects. From studies of
sucking and mouthing objects, newborns displayed perception of objects’ hardness,
texture and shape. Newborns also perceived temperature by mouth, but this was not
evident for newborns of nondepressed mothers. Studies of grasping and handling
revealed that newborns detected all properties of the objects presented to them,
including temperature and weight perception by hand. The behaviors of these
newborns that apparently were used to perceive the objects included squeezing,
grasping or clutching, slow finger stroking, hand turning and hand lifting of objects
to the mouth or face. Taken together, these findings suggest that neonates explore
objects actively with their hands and mouth, often using behaviors similar to those
applied by adults to detect or perceive specific properties of objects. Moreover, in
the absence of sophisticated fingering abilities, squeezing alone may enable the
detection of gross aspects of objects, such as shape information.
The less optimal mouthing and handling behaviors of the newborns of depressed
mothers were also reviewed. Their perception might have been impaired by poor
finger and hand movements or inactivity. Toys for the newborn should be developed
to foster mouth, hand and finger activity, especially for infants “at risk” who may
show delayed perception.
Further study is needed on the sense of touch and the perception of touch by
neonates. Surprisingly, neonatal touch studies are rare, and touch perception
by neonates is still viewed as almost nonexistent. Touch is the most well-developed
sense organ at birth. Like the other modalities—smell, taste, vision and audition—
touch and touch perception need additional research.
We wish to thank the neonates and mothers who participated in our studies.
We also thank Shay Largie for her assistance with recruitment and data collection.
This research was supported by funds provided to the Touch Research Institutes by
the Johnson & Johnson Pediatric Institute, L.L.C., as well as by a National
Institute of Mental Health (NIMH) Senior Research Scientist Award (MH#00331)
and an NIMH Research Grant (#MH46586) awarded to Tiffany Field.
Ref ere n c e s
1. Brazelton TB, Nugent JK. Neonatal Behavioral Assessment Scale. 3rd ed. London, England:
Mac Keith Press; 1995.
2. Nagy E, Molnar P. Heart rate deceleration during the grasping reflex. European Journal
of Pediatrics. 1999;158:576-577.
3. Clifton RK, Meyers WJ. The heart-rate response of four-month-old infants to auditory
stimuli. Journal of Experimental Child Psychology. 1969;7:122-135.
4. Pomerleau-Malcuit A, Clifton RK. Neonatal heart-rate response to tactile, auditory, and
vestibular stimulation in different states. Child Development. 1973;44:485-496.
5. Illingworth RS. The Development of the Infant and Young Child: Normal and Abnormal.
3rd ed. London, England: E & S Livingstone; 1966.
6. Prechtl HFR. The Neurological Examination of the Full-Term Newborn Infant. 2nd ed.
London, England: Heinermann; 1977.
7. Petrikovsky BM, Kaplan GP. Fetal grasping of the umbilical cord causing variable fetal heart
rate decelerations. Journal of Clinical Ultrasound: JCU. 1993;21:642-644.
8. de Vries JI, Visser GH, Prechtl HFR. The emergence of fetal behavior: I. Qualitative aspects.
Early Human Development. 1982;7:301-322.
9. Bushnell E, Boudreau JP. Exploring and exploiting objects with the hands during infancy.
In: Connolly KJ, ed. The Psychobiology of the Hand. London, England: Cambridge University
Press; 1998:144-161.
10. Morange F, Bloch H. Lateralization of the approach movement and the prehension
movement in infants from 4 to 7 months. Early Development and Parenting. 1996;5:81-92.
11. Walco GA, Cassidy RC, Schechter NL. Pain, hurt, and harm. The ethics of pain control
in infants and children. The New England Journal of Medicine. 1994;331:541-544.
12. Anand KJ, Hickey PR. Pain and its effects in the human neonate and fetus. The New
England Journal of Medicine. 1987;317:1321-1329.
13. Giannakoulopoulos X, Sepulveda W, Kourtis P, Glover V, Fisk NM. Fetal plasma cortisol
and beta-endorphin response to intrauterine needling. Lancet. 1994;344:77-81.
14. Glover V, Fisk N. Do fetuses feel pain? We don’t know; better to err on the safe side from
mid-gestation. BMJ: (Clinical Research Ed.). 1996;313:796.
15. Lloyd-Thomas AR, Fitzgerald M. Do fetuses feel pain? Reflex responses do not necessarily
signify pain. BMJ: (Clinical Research Ed.). 1996;313:797-798.
16. Fitzgerald M, Jennings E. The postnatal development of spinal sensory processing. Proceedings
of the National Academy of Sciences of the United States of America. 1999;96:7719-7722.
17. Forssberg H. The neurophysiology of manual skill development. In: Connolly KJ, ed.
The Psychobiology of the Hand. London, England: Cambridge University Press; 1998:97-122.
18. Rochat R. Mouthing and grasping in neonates: evidence for the early detection of what hard
or soft substances afford for action. Infant Behavior and Development. 1987;10:435-449.
19. Gibson JJ. Observations on active touch. Psychological Review. 1962;69:477-491.

20. Lederman SJ, Klatzky RL. Haptic classification of common objects: knowledge-driven
exploration. Cognitive Psychology. 1990;22:421-459.
21. Hernandez-Reif M, Field T, Diego M, Largie S. Weight perception by newborns of depressed
versus non-depressed mothers. Infant Behavior and Development. 2001;24:305-316.
22. Molina M, Jouen F. Modulation of the palmar grasp behavior in neonates according to
texture property. Infant Behavior and Development. 1998;21:659-666.
23. Streri A, Lhote M, Dutilleul S. Haptic perception in newborns. Developmental Science.
2000;3:319-327.
24. Lipsitt L, Kaye H. Change in neonatal response to optimizing and non-optimizing sucking
stimulation. Psychonomic Science. 1965;2:221-222.
25. Rochat P. Oral touch in young infants: responses to variations of nipple characteristics in
the first months of life. International Journal of Behavioral Development. 1983;6:123-133.
26. Meltzoff AN, Borton RW. Intermodal matching by human neonates. Nature. 1979;282:
403-404.
27. Pecheux M, Lepecq J, Salzarulo P. Oral activity and exploration in 1-2 month old infants.
British Journal of Developmental Psychology.1988;6:245-256.
28. Gibson EJ, Walker AS. Development of knowledge of visual-tactual affordances of substance.
Child Development. 1984;55:453-460.
29. Kaye KL, Bower TGR. Learning and intermodal transfer of information in newborns.
Psychological Science. 1994;5:286-288.
30. Hernandez-Reif M, Field T, Del Pino N, Diego M. Less exploring by mouth occurs in
newborns of depressed mothers. Infant Mental Health Journal. 2000;21:204-210.
31. Field T. Infants of depressed mothers. Infant Behavior and Development. 1995;18:1-13.
32. Campbell SB, Cohn JF. Prevalence and correlates of postpartum depression in first-time
mothers. Journal of Abnormal Psychology. 1991;100:594-599.
33. Hart S, Jones NA, Field T, Lundy B. One-year-old infants of intrusive and withdrawn
depressed mothers. Child Psychiatry and Human Development. 1999;30:111-120.
34. McCall RB. What process mediates predictions of childhood IQ from infant habituation
and recognition memory? Speculations on the roles of inhibition and rate of information
processing. Intelligence. 1994;18:107-125.
35. Ruff HA, McCarton C, Kurtzberg D, Vaughan HG Jr. Preterm infants’ manipulative
exploration of objects. Child Development. 1984;55:1166-1173.
36. Blass EM, Watt LB. Suckling- and sucrose-induced analgesia in human newborns. Pain.
1999;83:611-623.
37. Radloff L. The CES-D Scale: a self-report depression scale for research in the general
population. Journal of Applied Psychological Measures. 1977;1:385-401.
38. Whiffen VE, Gotlib IH. Infants of postpartum depressed mothers: temperament and
cognitive status. Journal of Abnormal Psychology. 1989;98:274-279.
39. Zuckerman B, Bauchner H, Parker S, Cabral H. Maternal depressive symptoms during
pregnancy, and newborn irritability. Journal of Developmental and Behavioral Pediatrics: JDBP.
1990;11:190-194.
40. Lundy BL, Jones NA, Field T, et al. Prenatal depression effects on neonates. Infant Behavior
and Development. 1999;22:119-129.
41. Carver C, White T. Behavioral inhibition, behavioral activation and affective responses
to impending reward and punishment: the BIS/BAS scales. Journal of Personality and Social
Psychology. 1994;67:319-333.
42. Gray JA. The psychophysiological basis of introversion-extraversion: a modification of

Eysenck’s theory. In: Nebylitsyn VD, Gray JA, eds. Biological Bases of Individual Behavior.
San Diego, Calif: Academic Press; 1972:182-205.
43. Gray J. A critique of Eysenck’s theory of personality. In: Eysenck HJ, ed. A Model for
Personality. Berlin, Germany: Springer-Verlag; 1981:246-276.
44. Field T, Diego MA, Dieter J, et al. Depressed withdrawn and intrusive mothers’ effects
on their fetuses and neonates. Infant Behavior and Development. 2001;24:27-39.
45. Klatzky RL, Lederman SJ, Metzger VA. Identifying objects by touch: an “expert system.”
Perception & Psychophysics. 1985;37:299-302.
46. Lederman SJ, Klatzky RL. Hand movements: a window into haptic object recognition.
Cognitive Psychology. 1987;19:342-368.
47. Fagard J, Jacquet A. Onset of bimanual coordination and symmetry versus asymmetry
of movement. Infant Behavior and Development. 1989;11:229-235.
48. Ramsey D, Weber S. Infants’ hand preference in a task involving complementary roles
for the two hands. Child Development. 1986;57:300-307.
49. Hernandez-Reif M, Field T, Diego M, Largie S. Haptic habituation to temperature is slower

in newborns of depressed mothers. Infancy. 2003;4:47-63.
50. McCall RB, Carriger MS. A meta-analysis of infant habituation and recognition memory
performance as predictors of later IQ. Child Development. 1993; 64:57-79.
51. Colombo J, Frick JE, Gorman SA. Sensitization during visual habituation sequences:
procedural effects and individual differences. Journal of Experimental Child Psychology.
1997;67:223-235.
52. Maikranz JM, Colombo J, Richman WA, Frick JE. Autonomic correlates of individual
differences in sensitization and look duration during infancy. Infant Behavior and
Development. 2000;23:137-151.
53. Fitzgerald M, Gibson S. The postnatal physiological and neurochemical development

of peripheral sensory C fibers. Neuroscience. 1984;13:933-944.
54. Andrews K, Fitzgerald M. The cutaneous withdrawal reflex in human neonates: sensitization,
receptive field, and the effects of contralateral stimulation. Pain. 1994;56:95-101.
55. Klatzky R, Lederman S. Spatial and nonspatial avenues to object recognition by the human
haptic system. In: Eilan N, McCarthy R, Brewer B, eds. Spatial Representation. Cambridge,
Massachusetts: Blackwell Publishers; 1993:191-205.
CHAPTER 3:
M AT E R N A L
TOUCH EFFECTS ON
I N F A N T B E H AV I O R
Hiselgis Perez, PhD

Jacob L. Gewirtz, PhD
Department of Psychology
Florida International University
Ab s t ra c t
Although the positive effects of massage therapy on infant behavior and growth
have been well documented, few studies have evaluated the reinforcing effects of
different types of tactile stimulation on infant behavior or the distinctions between
maternally delivered and infant-originated tactile stimulation. The study detailed
in this chapter assessed 3 types of tactile stimulation delivered at 2 levels of tactile
pressure. Results of this study revealed that the type of touch preferred most by
infants was “intense stroking.” Conversely, the type of touch preferred least by
infants was “intense poking.” The methods used in this study provide a reliable
means of quantifying tactile pressure.
Introd uc t i o n
Touch is an important element in the interactions between infants and their

caregivers. The amount of maternal touch that occurs during brief interaction
periods has been estimated to range from 33% to 61% of infant-caregiver
interactions.1,2 Additionally, the frequency of touch appears to vary with the child’s
developmental level: Day-care teachers have been noted to touch infants more than
toddlers or preschoolers.3
TOUCH EFFECTS ON INFANTS
When the behavioral states of infants have been compared across several conditions:
alone, mother’s face only, mother’s voice only, mother’s touch only and a
combination of mother’s voice, touch and face,2 infants exhibit the lowest frequency
of fussing and crying during the “touch-only” condition. Furthermore, using touch
during the maternal “still-face” procedure has been shown to reduce the frequency
of fussing, crying and grimacing among 3-month-old to 5-month-old infants, as
well as to maintain more smiling and gazing.1 These effects have also been noted
in 3-month-old infants of depressed mothers.4
MASSAGE THERAPY EFFECTS ON NEONATES AND

YOUNG INFANTS
The positive effects of massage therapy (ie, rhythmic stroking) on infant behaviors
and growth rates have been documented in many studies.3-12 Massage has enhanced
growth rates among preterm, cocaine-exposed, HIV-exposed and full-term
neonates.5,6 Massage therapy has also reduced the signs of stress in normal,
preterm, cocaine-exposed and HIV-exposed neonates.13 In addition,
massaged preterm and cocaine-exposed neonates have received higher motor
scores on the Brazelton Neonatal Assessment Scale when compared with control
neonates who did not receive massage.14,15 Massage has also been associated with
improvements on “…emotionality, sociability, and soothability temperament
dimensions…”7 and, most especially, on sleep organization in full-term infants.8
Furthermore, 3-month-old infants who were massaged by their fathers greeted
them with more eye contact, smiling, vocalizing, reaching and orienting responses,
and showed fewer avoidance behaviors, than did those infants who were not
massaged by their fathers.9,16
Chapter 3: Maternal Touch Effects on Infant Behavior 41
Lastly, massage has been shown to affect infant learning: Infants who received
massage prior to an audiovisual habituation task performed better than those
infants who received either play or no stimulation before the audiovisual
habituation task.10
TOUCH AS A REINFORCER FOR INFANT

BEHAVIORS
The role of touch as a reinforcer for various infant behaviors has also been
explored.11,12,17 One of the methods used to assess reinforcers is to follow the
behavior by an event and measure whether the behavior increases. If the behavior
increases from the event that followed, the event is termed a “positive reinforcer.”
Contingent tactile stimulation—alone or along with visual and/or auditory
stimulation—has been found to increase infant behaviors that produce it.11,12
In fact, contingent touching of the infant’s face, abdomen and/or limbs has been
found to increase vocalizations, smiles, eye contact and approach behaviors in
infants 1.5 months of age to 9 months of age.11,12,17,18
U N R E S O LV E D I S S U E S I N T H E L I T E R AT U R E
REGARDING INFANT TACTILE STIMULATION
Although the literature on tactile stimulation has yielded valuable information

about the effects of tactile stimuli for infant behavior and learning, unresolved
issues remain. Firstly, only one study explored the differential reinforcing effect of
different types of tactile stimuli.12 Without measuring tactile pressure, these authors
compared the effects of tickling and poking, versus stroking, that occurred following
the eye-contact responses of 2- to 4.5-month-old infants. The investigators found
that continuously contingent rhythmic stroking was a more-effective reinforcer for
continuous infant eye contact than was continuously contingent arrhythmic poking
plus tickling. However, there was no separation of the poking-plus-tickling stimulus:
Thus, the reinforcing effects of the distinct stimulus components could not be
assessed. Combining tickling with poking may mask the individual effects of each
component: For example, if tickling was a highly effective reinforcer and poking
was neither reinforcing nor aversive, combining them into a single contingent
treatment might, at best, yield only a moderate reinforcing effect. Such moderate
effects may not be representative of the true effects of either of the 2 component
stimuli because different types of touch may have different effects.
Maternal stroking has been noted to occur very often during mother-infant
interactions.19 Maternal stroking seems to occur most frequently, followed by
maternal tickling, while maternal poking is rare. Some have suggested that each
of these forms of touch conveys a different message.2 The more negative types of
touch also occur more often during disturbed interaction: For example, compared
with nondepressed mothers, depressed mothers show more frequent jabbing and
poking of their infants.20 In turn, such jabbing and poking seem to be associated
with negative affect and gaze aversion exhibited by infants.21
Other relevant stimulus qualities, such as duration, pattern and pressure of touch
need further investigation. Specifying the pressure of tactile stimuli is important
because it may help determine the quality and/or quantity of behavioral effects.
Without accurate measures of the attributes of tactile stimuli, such as pressure,
it may prove difficult to predict the effects of these stimuli on infant behaviors.
Specifying and quantifying the qualities of the touch stimulation would advance
the sciences of touch measurably.
Exploring the Ef f e c t s o f D i f f e re nt Typ e s and

Inten s i t i e s o f To u c h : Cu r re nt St u d y
OVERVIEW
This study analyzed the effects of 3 different types of tactile stimulation at 2 levels
of tactile pressure. Stimuli were considered reinforcers if the infant’s target behavior,
leg kicking, increased. Secondly, the relative efficacies of different tactile stimuli as
positive reinforcers were evaluated.
SUMMARY OF METHODOLOGY
The experiment was conducted using 6 different orders of the stimulation on an

A-B-A-C-A-D–phase design. During the first A phase, a baseline rate for the leg
kicks per 30 seconds was assessed by instructing the mothers to behave as they
would normally around the child; that is, by ignoring the leg kicks. Each of the B,
C and D phases consisted of 2 or 3 alternating treatments at one of the 2 intensity
levels: for instance, mild stroking versus mild poking versus mild tickling; intense
stroking versus intense poking versus intense tickling. During the learning phases,
every leg kick produced the touch stimulus. The second and third A phases were
reversals of the phase that had preceded it immediately. During the reversal phases,
nontarget behaviors received contingent stimulation. For example, nonkicking
behaviors were followed by tactile stimulation. The frequency of stimulation was
equated with that of the previous treatment phase by ensuring that the mothers
provided the earlier contingent stimulus for their child’s nontarget (ie, nonkicking)
responses at least as often as they did for the earlier target (ie, kicking) response in
the corresponding time period of the previous treatment phase: For example, if a
mother delivered the touch contingent on the target behavior 20 times during the
first 30 seconds of the treatment phase, she was required to deliver the same touch
noncontingently at least 20 times during the first 30 seconds of the reversal phase.
The sample split equally along gender lines and included 10 Hispanic and
2 African American infants who were 2 months of age to 5 months of age (mean,
3.6 months). The experiment was conducted in the infants’ homes. Each child sat
on their infant seat 3 feet from a video camera, which recorded his or her behavior
as well as the touch stimulation. Each mother sat next to her child (Figure 1) so that
she could touch her infant according to the instructions given. The experimenter,
who was located on the other side of the room, scored the child’s behavior as shown
on the computer-monitor screen. The leg-kick measurement device consisted of an
infrared beam that was projected 5 inches above the child’s calves. When the child
kicked and either or both legs crossed the beam, a leg kick was registered. The
Figure 1. Representation of
positioning of mother and child
during experiment examining
maternal touch intensity
and infant response 21
microphone of the meter that measured sound level was positioned directly on the
infant’s skin. The mother was instructed to administer the touch within a 2-inch
radius of this very-sensitive microphone that detected the loudness (in decibels
[dBs]) of the sounds that were made as the mother’s skin and the infant’s skin
rubbed against each other during contact. More-intense touching produced higher
dB readings than did less-intense touching on the sound-level meter. The mother
could monitor the intensity of her touch via an analog display on the sound-level
meter. She was instructed to keep her touch intensity between –10 and –5 dB
for the mild-touch condition and between +5 and +10 dB for the intense-touch
condition.
The target behavior was an infant leg kick, defined as the leg going from a
bent-at-the-knee position to a straight position, or the reverse (straight to bent).
The leg-kick response was chosen because it occurs reliably in 2-month-old to
5-month-old infants. Occurrences of the leg-kick response, and of the touch
response and amplitude, were represented automatically by a highly visible blip
on a computer monitor.
Stroking, tickling and poking were produced by the infants’ leg kicks. Thus,
initially, the touch was used as a reinforcer for the infant response and directly
reflected infant preferences for the different types and intensities of touch. The
duration of every touch, 3 seconds, was measured by a timer operating within
the data-acquisition program. Stroking was defined as soft, rhythmic, continuous
touching of the infants’ limbs or abdomens using one hand. Tickling consisted of
arrhythmic touching of the infants’ limbs or abdomens using the fingertips of one
hand. Poking was comprised of a one-finger continuous touch of the infants’ limbs
or abdomens. Additionally, there were 2 levels of pressure for the stroking, tickling
and poking treatments.
SUMMARY OF FINDINGS
With the exceptions of mild poking and intense poking, which appeared only
occasionally, the other forms of touch proved reinforcing. The most-preferred type
of touch was intense stroking, while the least-preferred type of touch was intense
poking.
Results of this study indicated that there were numerous differences among the
treatments. Firstly, intense stroking proved to be a more-effective reinforcer than
did intense poking, mild poking, mild stroking and mild tickling. Additionally,
intense tickling proved more effective than intense poking and mild poking. Lastly,
mild stroking proved more effective than intense poking and mild poking, while
mild tickling proved more effective than intense poking and mild poking. These
results are depicted in Figure 2.
Figure 2. Interactions between type of touch and touch intensity in experiment

examining maternal touch intensity and infant response 21
Conclusions
Results of the current experiment help expand our understanding of the effects
of maternal touch on infant behavior. Firstly, this study quantified the pressure
of maternal tactile stimulation. Finding a reliable way to quantify the pressure of
touch represents a milestone in the field of touch research, as the intensity of touch
can now be monitored with accuracy. Secondly, results of this study yielded
valuable information about the effects of different types of touch. While
tickling/tapping, stroking and occasionally poking can all be reinforcing to the
infant, intense stroking was more effective than tickling/tapping or poking as
a means of increasing the rate of infant leg kicks. Higher pressure stroking and
tickling/tapping appeared to be associated with more frequent kicking, while
higher pressure poking seemed to be associated with less frequent kicking. This
information can help parents understand the types and pressures of touch that
may be used most effectively during their interactions with their infants.
Re f e re n c e s
1. Stack DM, Muir DW. Adult tactile stimulation during face-to-face interactions modulates
five-month-olds’ affect and attention. Child Development. 1992;63:1509-1525.
2. Tronick EZ. Touch in mother-infant interaction. In: Field TM, ed. Touch in Early
Development. Mahwah, NJ: Lawrence Erlbaum Associates; 1995:53-65.
3. Cigales M, Field T, Hossain Z, Peláez-Nogueras M, Gewirtz JL. Touch among children

at nursery school. Early Child Development and Care. 1996;126:101-110.
4. Peláez-Nogueras M, Field T, Hossain Z, Pickens J. Depressed mothers’ touching increases
infants’ positive affect and attention in still-face interactions. Child Development.
1996;67:1780-1792.
5. Field TM. Infant massage therapy. In: Field TM, ed. Touch in Early Development. Mahwah,
NJ: Lawrence Erlbaum Associates; 1995:105-114.
6. Field TM, Schanberg SM, Scafidi F, et al. Tactile/kinesthetic stimulation effects on preterm
neonates. Pediatrics. 1986;77:654-658.
7. Field T. Massage therapy effects. American Psychologist. 1998;53:1270-1281.
8. Field T, Grizzle N, Scafidi F, et al. Massage therapy for infants of depressed mothers. Infant
Behavior and Development. 1996;19:107-112.
9. Cullen C, Field T, Escalona A, Hartshorn K. Father-infant interactions are enhanced by
massage therapy. Early Child Development and Care. 2000;164:41-47.
10. Cigales M, Field T, Lundy B, Cuadra A, Hart S. Massage enhances recovery from habituation
in normal infants. Infant Behavior & Development. 1997;20:29-34.
11. Peláez-Nogueras M, Gewirtz JL, Field T, et al. Infants’ preference for touch stimulation in
face-to-face interactions. Journal of Applied Developmental Psychology. 1996;17:199-213.
12. Peláez-Nogueras M, Field T, Gewirtz JL, et al. The effects of systematic stroking versus
tickling and poking on infant behavior. Journal of Applied Developmental Psychology.
1997;18:169-178.
13. Wheeden A, Scafidi FA, Field T, Ironson G, Valdeon C, Bandstra E. Massage effects on
cocaine-exposed preterm neonates. Journal of Developmental and Behavioral Pediatrics: JDBP.
1993;14:318-322.
14. Scafidi FA, Field TM, Schanberg SM, et al. Effects of tactile/kinesthetic stimulation on
the clinical course and sleep/wake behavior of preterm neonates. Infant Behavior and
Development. 1986;9:91-105.
15. Scafidi FA, Field T, Schanberg SM. Factors that predict which preterm infants benefit
most from massage therapy. Journal of Developmental and Behavioral Pediatrics: JDBP.
1993;14:176-180.
16. Scholz K, Samuels CA. Neonatal bathing and massage intervention with fathers: the
behavioural effects 12 weeks after birth of the first baby. The Sunraysia Australia Intervention
Project. International Journal of Behavioral Development. 1992;15:67-81.
17. Lum Lock KL. Infants’ Approach and Avoidance to Strangers Influenced by Maternal
Contingencies [master’s thesis]. Miami, Fla: Florida International University; 1997.
18. Weisberg P. Social and nonsocial conditioning of infant vocalizations. Child Development.
1963;34:377-388.
19. Eckerman CO, Rheingold HL. Infants’ exploratory responses to toys and people.
Developmental Psychology. 1974;10:255-259.
20. Field T. Early interactions between infants and their postpartum depressed mothers. Infant
Behavior and Development. 1984;7:517-522.
21. Cohn JF, Tronick E. Specificity of infants’ response to mothers’ affective behavior. Journal
of the American Academy of Child and Adolescent Psychiatry. 1989;28:242-248.
CHAPTER 4:
TOUCHING DURING
MOTHER-INFANT
INTERACTIONS
Dale M. Stack, PhD

Department of Psychology and Centre for Research in Human Development
Concordia University and The Montreal Children’s Hospital
Ab s t ra c t
The primary objective of this chapter is to provide an overview of the research on

touch as it occurs in social contexts (ie, during interactions) between adults and
infants. In particular, this chapter focuses on face-to-face and still-face interactions,
and stresses the touching of infants by their respective mothers. Studies are
summarized that reflect on the contributions of parental touch and adult touch to
early interactions and the developing relationship with the infant, as well as how
touch is used in different ways depending on the interactive context or modalities
of communication available to parents. These reviews focus on the first 3 months
to 6 months of the healthy, typically developing infant’s life. Special emphasis is
given to this period because it is well-researched, and dyadic communications
and face-to-face interactions are prominent at this time. Another objective of this
chapter is to communicate the importance of parental touch in early social and
emotional development, specifically with regard to its roles in communication
and development. Further objectives are to highlight the advances that have been
made in this area of investigation, noting some of the new and emerging directions
being taken as a result of these investigations and to delineate some of their
important contributions.
Parental touching
and holding
are important in
early infant
development.
Photo used by permission of SL Labs.
The Impo r t a n ce o f To u c h
Long-held evidence shows the importance of tactile stimulation for the normal
development of nonhuman species. Examples include the survival functions of
maternal washing of the young1 and the specific beneficial effects of handling,
licking and grooming on survival,2,3 growth, development and resistance to disease,4-7
as well as increased exploratory behavior.8 The classic body of work done by
Harlow on rhesus monkeys substantiated the importance of tactile stimulation by
demonstrating that the development of social attachment was more dependent on
physical contact than it was on reducing the feeding drive.9 Additional data from
primate and rodent models have implicated physical contact and touch (tactile
stimulation) as significant concomitants of the infant’s ability to regulate its own
responses to stress.10-12 More complete coverage of the animal research can be found
in a chapter by this author13 and in the texts written by Montagu14 and Field.15
While the importance of touch for the normal development of nonhuman species
is not in question, the importance of touch to human infants has also been
recognized. Classic observations regarding maternal deprivation of human infants
and the lack of tactile stimulation by human adults has underscored the value of
tactile contact including reports of infants who were deprived of mothering for
lengthy periods of time as well as institutionalized infants who were given only
Chapter 4: Touching During Mother-Infant Interactions 51
essential care with no extra attention from the staff.16,17 This work, at least in part,
led to more systematic work on maternal attachment, bonding and deprivation.
Unfortunately, the literature on maternal deprivation is replete with methodological
problems, permitting only cautious interpretations. Recent research, however,
suggests that extreme forms of limited touch affect children’s growth and
development.18,19
Despite advances and the acknowledgment of the importance of tactile stimulation,

the specific contribution of tactile stimulation to early development (ie, during
the first year of life) remains relatively undefined. Evidence, however, suggests
that touch regulates physiological and behavioral reactions during early infant
development.14,20 That is, touch can aid in controlling the infant’s state of arousal
(behavioral state; ie, maintaining alertness, reducing drowsiness, and more). Touch
is also an effective stimulus for soothing human neonates.21,22 For example, tactile
contact and vestibular-proprioceptive stimulation are often paired to soothe the
baby (eg, Byrnes and Horowitz, 1981; Korner and Thoman, 1972 22).
The effects of touch/contact have also been seen in studies where touch has been
used to induce or change behavioral state. In many studies, touch is used between
intervals of a study to maintain alert state in babies, to calm them or as an
attention-getting stimulus.23,24 In the neonate, touch has been shown to reduce
stimulation, thereby acting like a control system to maintain state.25 According to
Brazelton, touch can also stimulate the infant, thereby illustrating how touch can
both instigate and maintain communication.25 Nonverbal maternal behaviors also
provide a means of modulating the overall level of stimulation to which the infant
is exposed, potentially facilitating regulation of its own state and level of arousal.26
Based on reviews of the literature, more emphasis has been placed on the physical
benefits of touch to high-risk infants and the development of the perception of
touch, while less attention has been devoted to touch as it relates to infants’ social
and emotional development.13 Moreover, there has been a general lack of research
on the tactile modality versus auditory and visual modalities in perceptual research
as well as facial and vocal channels in social-emotional research.
Touch, Emo t i o n a nd Co m m u ni c a t ion
If touch is so fundamental, its role(s) in social-emotional development should not

be overlooked, and its ability to communicate deserves attention.13 Emotions can
be communicated through touch: That is, touch conveys meaning (eg, soothe, calm,
active, happy). Although there are very few systematic studies that support such
claims, touch appears to convey emotion very directly, and the type of touch
conveys particular emotions. Emotions (ie, feeling states or messages) that might be
communicated through touch include love and caring, sympathy, empathy, anger
and a sense of security. According to Montagu, for example, through feeling we refer
frequently to emotional states, such as happiness, joy, sadness, melancholy and
depression, and by the term we often imply a reference to touching.14 According to
Virel (an anthropologist and neurologist), the skin may be seen as a mirror of the
organism’s functioning: “Its color, texture, moistness, dryness and every one of its
other aspects, reflect our own state of being, psychological as well as physical. We
blanch with fear and turn red with embarrassment. Our skin tingles with excitement
and feels numb with shock; it is a mirror of our passions and emotions.”27
Tronick has explored ideas (Figure 1) along the same lines, arguing that “…certain
forms of touch, such as gentle holding, might convey the message, you are safe,
whereas other forms of touch, such as poking or jabbing, may convey the message,
you are physically threatened.”28 Touching seems to communicate; to bring out
meaning: Montagu said, “Although touch is not itself an emotion, its sensory
elements induce those neural, glandular, muscular, and mental changes which in
combination we call an emotion.”14
Figure 1. Different types of touch convey different meanings.
Gentle holding You are safe
Poking or jabbing You are in danger
Touch Communication
In a study designed to examine the effects of touch on infants’ emotions,

Hertenstein and Campos had mothers stimulate their 12-month-old infants tactilely
in specific ways while objects were presented to them.29 The study conditions
included the following: a group of mothers who tensed their fingers around their
infants’ abdomens and inhaled (designed to elicit negative emotion); a group
of mothers who relaxed their grips around their infants’ abdomens and exhaled
(designed to elicit positive emotion); and a control condition, where there was no
tactile stimulation. Infants in the former condition touched the objects presented
to them less frequently and exhibited more negative emotional displays. Contrary
to the expectations of Hertenstein and Campos, there was no effect on positive
emotion in the condition where the grip was relaxed.
Touch is also involved in mother-infant attachment and the affectional systems.30-32

The amount and quality of physical contact (ie, touching, proximity) between the
mother and her infant are essential to their relationship.14,33 Contact behaviors are
integral features of emotional communication between mothers and infants, and
higher levels of touch are related to secure positive attachment.30,31 Key components
of current theories of attachment include physical proximity and proximity-seeking
measures. In her examination of the relationship between kissing, hugging, patting
and attachment behavior in infancy, Landau argued that there is an intricate
relationship between infant affectionate behavior and attachment.32 More studies
are needed, however, to relate specific types of touch to attachment constructs, such
as proximity.
Maternal sensitivity, long considered a key contributor to synchronous and

mutually reciprocal interactions, has recently been related to optimal patterns of
attachment.34 The key characteristics of maternal sensitivity and responsiveness
often involve physical closeness and physical touching behaviors. In addition, the
frequency and duration of touch are also considered by some researchers to be an
index of mothers’ attachment.35-37 One of the routes used to examine the more
direct role(s) of touch in communication and social-emotional development is to
study the interactions between mothers and their infants systematically and to
examine touching during these interactions specifically.
Moth e r -In f a n t To u c h i ng D u r i ng S ocial Inte ractions:

Gen e ra l Fi n d i ngs
Examples abound of how touch is related to tender, loving care—from the

beginnings of life at the breast; to being held and cuddled, rocked to sleep, stroked
and swayed to reduce distress; to being tickled and played with, inciting giggling
and laughter; and to being hugged for affection and comfort. Even during the
newborn period, the importance of physical contact between parents and their
neonates has been demonstrated. For example, research studies have examined
specific patterns of touching when first handling newborns,38-43 the shorter- and
longer-term effects of early and extramaternal contacts,39,40,44 types of early maternal
tactile contact45,46 and the effects of infant massage.47-49 On the bases of these
studies and others, it seems clear that touching as a sensory system and as a means
of communication is important to that first relationship—between parent and
infant—as well as individually important to the infant.
A typical
face-to-face
interaction
between a
mother and
her infant.
Photo by author.
As infants develop over the first few months of their lives, social interactions
become increasingly more frequent and represent a means of communicating and
learning. Face-to-face interactions are a frequent way that mothers and infants
interact. Commonly, researchers have focused on the mothers’ faces and voices as
means of communication during these interactions.52,61 However, while mothers’
facial and vocal expressions are important, they are only 2 of the many behaviors
that are used to express and communicate. Researchers have only recently begun
to explore the role of touch in early interactions.57,65-67,78 In the past, the focus was
on the more distal gaze and affect behaviors—to the relative exclusion of touch
and gesture.13 Yet, touch is employed by mothers quite commonly, along with
their vocal and visual expressions, during face-to-face interactions and during play:
For example, during face-to-face interactions, the infant and adult (primarily the
mother) are seated at eye-level to each other during a series of brief interaction
periods. The caregivers interact spontaneously, using their facial, vocal and tactile
expressions, while the infants respond to, and even initiate, interactions. Face-to-
face interactions have been one of the primary means used to study infants’ social
communications,50 emotional expressions and responses to stressful episodes,51
and the development of social expectations.52 However, typically, researchers have
analyzed maternal and infant facial and vocal behavior50,61 but not touch, although
incidental reports reveal that maternal touch occurs during 33% to 61% of brief
interaction periods.50,53,54 Other measures, such as posture,55 manual hand actions56
and gesture,57 have also been documented. Contextual features, such as location
during play, position, inclination of position and proximity of contact, are other
examples of important factors influencing infants during face-to-face play.58,59
Mo t he r -In f a nt To u c h i ng D u r i ng S ocial Inte ractions:

St i ll-f a c e Stu d i e s
The still-face (SF) procedure,60 a modification of the face-to-face procedure, has

been a valuable tool when examining the role of touching (Figure 2). The mother-
infant interaction is divided into 3 brief periods of 90 seconds to 120 seconds each.
In period 1, mothers interact normally, using facial expression (face), voice and
Figure 2. Still-face (SF) procedure.
NORMAL STILL-FACE NORMAL

Maternal face, Neutral face, Maternal face,
voice and touch no voice, no touch voice and touch
Decreased gazing and
smiling at mothers
touch (“normal”). In period 2, mothers assume a “neutral,” nonresponsive SF

and provide neither vocal nor tactile stimulation. While in period 3, they resume
“normal” interaction. During the SF, compared with the “normal” periods, infants
typically decrease gazing and smiling at mothers,61-63 increase neutral to negative
affect and increase vocalizing.64,65
Gusella, Muir and Tronick compared the responses during SF periods between
groups of infants wherein some received maternal touch in the preceding “normal”
period, while others received only maternal face and voice.61 These investigators
found that 3-month-old infants smiled and gazed less at their mothers during the
SF period than during the “normal” periods (compared with a “no-change” control
group who received 3 “normal” periods), but that their behavior was significantly
different from the control group only when maternal touch was permitted during
the “normal” period preceding the SF. That is, these 3-month-old infants exhibited
the SF effect only when maternal touching was part of the prior “normal” periods
and their attention declined over time without the tactile stimulation. This suggests,
according to Gusella, Muir and Tronick,61 that maternal touch during the “normal”
periods facilitated the maintenance of attention in these very young infants.
Indirect evidence that supports the importance of touch in early interactions is

derived from a study undertaken by Field et al.51 These investigators compared the
responses of 4-month-old infants to an SF episode versus those to a separation
sequence. Results showed that these infants found the SF sequence more stressful:
For example, these infants demonstrated more motor activity, gaze aversion and
distress brow and crying, and less smiling. More maternal tactile-kinesthetic
behavior was shown following the SF period: That is, maternal proximal and
comforting behaviors were potentiated. These findings support the view that
infants are sensitive to maternal cues and suggest both the soothing role and the
communicative nature of touching.
Studies examining infants’ responsiveness to their mothers’ touch and their

sensitivity to touch when other forms of stimulation are absent provide important
insight into why mothers use touch during the first 6 months of life. My colleagues
and I have conducted several studies that have helped elucidate these important
issues. By comparing a standard SF with one where mothers could touch during
the SF period (Figure 3), Muir and I showed that by adding touch, infants did not
become distressed, they showed increased smiling and they maintained the high
levels of gaze that are typical in “normal” interactions.65 This new role for touch in
moderating the SF effect has been replicated several times.33,66,67 Moreover, Muir
and I demonstrated that it was the tactile stimulation, not the visual stimulation,
from the adults’ hands that was responsible for the effects.33
Figure 3. Modified still-face (SF)-with-touch procedure.65
NORMAL STILL-FACE NORMAL

+ TOUCH
Maternal face, Maternal face,
voice and touch Neutral face, no voice, voice and touch
but with touch
Increased gazing and
smiling at mothers
Beyond the general issues of infants’ responsiveness and sensitivity, Muir and I went
on to address 2 additional questions: Are infants sensitive to subtle manipulations
of maternal touch? And, are mothers able to use touch to obtain specific responses
from their infants? Infants’ sensitivity to changes in their mothers’ touch has been
demonstrated through the SF procedure by providing a group of mothers with
differing verbal instructions.33,65,67 This is what might be referred to as “changes
in context”: For example, it has been shown that mothers can use touch to elicit
specific behaviors from their infants—such as maximizing their infants’ smiling67
and shifting their infants’ attention to their mothers’ hands.33 The resulting changes
in infants’ behaviors, as a function of the changes in the mothers’ verbal instructions,
implies that there were changes in maternal touch.
As an illustration, in a study by LePage and my lab, experimental infants who

received only touch from their still-faced mothers smiled as much as did control
infants who received “normal” periods of face-to-face interaction during the
period when mothers were asked to obtain the most smiling from their infants.67
Moreover, mothers who were asked to touch their infants on only one area of their
bodies tended to use one type of touch most frequently, and there was a significant
association between the area of the body touched and the type of touch the
mothers used. Lastly, using sequential analyses to examine how touch contributes
to the reciprocal and contingent aspects of communicative exchange, LePage and
I were able to show that cyclical patterns of mother-and-infant behaviors were
demonstrated “around” maternal touching.67 Before their mothers touched them,
infants were likely to be gazing at either their mothers’ hands or their mothers’
faces. After their mothers touched them, however, the infants were more likely to
gaze at their mothers’ hands. Smiling was frequent and was always accompanied by
gazing at mothers’ faces or hands, rather than occurring simply in isolation. These
findings suggest that infants may have been directed to the active component of the
hands and that they were enjoying the tactile stimulation that they were receiving,
which was reflected in their smiling both before and after being touched.
Using an SF-with-touch procedure and making comparisons to a no-change control

group, Arnold and I examined whether touch and gesture alone are used to obtain
specific infant responses.57 Following a period of “normal” interaction, the mothers
in the experimental group received the following additional instructions: try to get
your infants to imitate you; engage your infants in a playful interaction; use touch
only to attract and maintain your infant’s attention on your face with as much eye-
to-eye contact as possible. We found that the infants were sensitive to changes in
maternal touch and hand gestures, and that, when instructed, the mothers appeared
successful in eliciting specific behaviors from their infants using only nonverbal
channels of communication: For example, maternal touch and hand gestures
attracted the infants’ attention to their mothers’ faces even when the mothers’ faces
were still and expressionless during the period when the mothers were instructed to
attract their infants’ attention to their faces. During the period when the mothers
were instructed to engage their infants in a playful interaction, the infants in the
experimental group smiled more frequently in comparison with the infants in the
control group. Moreover, within the experimental group and across the periods, the
infants also smiled more during this playful interaction period compared with the
period when the mothers were asked to get their babies to imitate them.
It is important to note that for the previous 2 studies mentioned,57,67 the changes
in maternal touch were inferred on the bases of changes in the infants’ behaviors.
The actual patterns and types of touching were not measured directly. Therefore,
Arnold 73 and I13,72 have now examined some of the specific changes in touching
using the Caregiver-Infant Touch Scale.113,68 Some of these findings are described in
a following section that deals with patterns of touching during social interactions.
In a different but original direction, LePage used a modification of the SF-with-

touch procedure to investigate infants’ abilities to perceive a tactile contingency (or
the lack of a contingency) during social interactions.69,70 Infants in the “contingent”
condition were reinforced with standardized tactile stimulation (SF interaction,
with touch as the reinforcer) for gazing at the experimenter’s neutral face. Infants
in the “noncontingent” condition received the same tactile stimulation as their
matched counterparts regardless of their behaviors. All infants in the contingent
condition learned the contingency: For example, their level of gazing at the
experimenter’s neutral face was higher, and gazing away was lower, relative to
infants in the noncontingent condition. LePage demonstrated that 4- and 7-
month-old infants could perceive and learn a contingent relationship presented
through the tactile modality during social adult-infant interactions, underscoring
the idea that infants of this age are both sensitive to, and reinforced by, touch.69

Inf lue n c e s of Mo d a l i t y
While facial, vocal and tactile components are used frequently during social
interactions, less is known about how they are used in combination, how they are
used to achieve goals, and, if they do convey messages, how this is accomplished.
Moreover, examination is warranted regarding how touching is integrated with the
other communication channels that are available to parents.13,71,72 Although it is
important to understand each component’s discrete and independent roles, the
context when much of early development occurs is social, and information is
typically specified multimodally for the developing infant: That is, touch is often
combined with other modalities of interaction. Consequently, how the modalities
are used in combination becomes an important issue: That is, do combinations
of modalities communicate more clearly? Are messages embedded in touch? After
examining infants’ sensitivity to subtle changes in maternal tactile behavior we
investigated the influence of “modality.” We studied this by comparing mother-
infant interactions in touch-only periods (ie, SF periods where touching only
was permitted) to conditions where mothers could use all modalities including
touch.57,71-73 We used changes in infants’ gaze and affect to imply responsiveness
to changes in their mothers’ behavior.
Our general research objectives were threefold.71 If infants are sensitive to maternal
touching, we needed to answer the following questions:
• How do infants’ reactions differ when all modalities are used?
• Is touch used differently during unimodal versus during

multimodal interactions?
• Can the same behavioral goals be achieved?
Using the same interaction periods as used by Arnold and my lab,57 results
indicated that when mothers were restricted to using only touch, they employed
more touch to compensate for the absence of other modalities of interaction.
Contrastingly, when all modes of interaction were available, mothers employed
varied communication strategies. Importantly, while these infants smiled and gazed
at their mothers’ faces more often during interactions involving all modalities,
mothers were nonetheless successful in directing infant attention and eliciting
smiling using only touch and gesture.
Using a different set of instructions, Arnold compared both experimental and

control groups in touch-only circumstances versus conditions where mothers were
permitted to use all of the modalities of communication available to them.73
Arnold discovered that while there was more touching done by mothers in the
“tactile” group, mothers in the “all-modes” group nevertheless employed touch
frequently (greater than 68% of the time). Mothers in the tactile group gestured
more frequently during the period when they were instructed to get their infants
excited and happy. Notably, the absence of group differences in gesture during
the period when mothers were instructed to attract their infants’ attention to their
faces suggests that, despite availability of face and voice, mothers in the all-modes
group employed gestures as an attention-getting strategy. Infants in the all-modes
group also gazed more at their mothers’ faces during all periods, except during the
“normal” period. There were no differences in gazing away: That is, infants in the
tactile conditions did not visually disengage more frequently.
Taken together with the studies that examined infants’ sensitivity to changes in
maternal touch, these 2 studies,71,73 examining the influence of modality, illustrate
that dramatic changes in infants’ responses to subtle perturbations can occur within
a restricted communicative context, underscoring the importance of nonverbal
behaviors (ie, touch and gestures). Maternal behavior, therefore, changes as a

function of instructional context. Moreover, 5-month-old infants are sensitive
to subtle changes in maternal touch and hand gestures.
Table 1: Brief summary of some of Stack and colleagues’ past findings I

1. Mothers frequently touch their infants during face-to-face interactions
2. Positive affective responses to their mothers during face-to-face, touch-only interactions
3. Touch moderates still-face effect: Maintains and directs infant attention (gaze)
and elicits positive affect (smiling), influences negative affect
4. Across age
5. Across experimenter and mother

6. Touch (not visual) stimulation of hands that is responsible for the effects
7. Active (not static) touch that is responsible for the effects
8. Infants are responsive to touch and are sensitive to subtle changes in maternal
tactile behavior
9. Modality: In general, infants smile and gaze at their mothers’ faces during interactions
involving all modalities, mothers are nonetheless successful in directing infant
gaze / attention and eliciting smiling using only touch and gesture
10. Maternal behavior changes as a function of instructional context
These findings underscore the value of studying touch and how it can illuminate our
understanding of communication

Be yo n d t he St i l l - Fa c e
Several studies have addressed touch—directly or indirectly—through procedures

that either do not use the SF or that attempt to bring together multiple behaviors
in some unified way: For example, in one of the few attempts to separate the social
components of maternal behavior systematically, Roedell and Slaby explored the
preferences of 24-week-old infants for 3 adults who interacted in different ways.74
One adult (distal) smiled, talked, sang and made facial expressions. Another adult
(proximal) carried, rocked, bounced, patted and stroked the infant, but remained
silent with a neutral face. A third adult (neutral) was silent, unresponsive and made
no eye contact. Over a 3-week period, infants increased their time spent near the
distal adult, while the infants made no changes in their relationship to the proximal
and neutral adults. Moreover, infants chose to look more at the distal adult than at
the proximal adult. Unfortunately, in this trial there were no measures of affect, the
adult was not permitted to maintain eye contact in the proximal condition and the
infant did not need to establish eye contact with the proximal adult in order to
receive stimulation. This lack of “natural social interaction” may have contributed
to the poor responses elicited by the proximal adult, and may have driven the
infants to look away.65
During a free-play interaction with toys, Leiba examined interactions longitudinally

between mothers and their 5-month-old and 12-month-old infants.75,76 One
component of the coding scheme employed by Leiba was displays of passive
affection (ie, touching and maintaining close physical proximity) and active
affection (ie, hugs, kisses, rubs). She found that while mothers engaged in longer
bouts of active affection with their children at 5 months of age, compared with
their children at 12 months of age, the frequency of both passive and active
affectionate displays increased at 12 months of age (from that at 5 months of age).
No age differences were found in the amount of time spent in close physical
proximity, implying that close physical contact was important regardless of age.
A typical layout for a floor-play

session with an arrangement
of standardized toys.
Photo by author.
Rhythmic touch is also preferred over nonrhythmic touch in dyads,77 and infants
are more responsive when touch is added to face and voice.78 To reinforce the
position that touch can elicit specific responses, Wolff studied the development
of smiling and found that, between the fourth and sixth weeks of life, “patty-cake”
becomes an efficient stimulus for smiling.79 To ensure that it was the
proprioceptive-tactile stimuli rather than extraneous stimulation that elicited the
smile, Wolff played the game in such a way that the infant could not see or hear
the elicitor during the test.79 The smiles evoked in this manner were described as
broad, the intensity of smiling was high and it was difficult to habituate the
response with repeated stimulation. Patty-cake, of course, is not the only game that
elicits such positive responses from infants: There are other parent-infant games
that also involve much touching and physical contact—lap games, tickle games,
“I’m-gonna-get-you” games, finger-walking games and even bouncing games and
“horsey” games.
As yet another example, the contention made previously by Gusella, Muir

and Tronick that touch may have served to maintain attention during an SF
procedure61 is consistent with observations made by Roggman and Woodson.80
They compared a pair of 3-minute play sessions during which mothers refrained
from touching their 3-month-old to 4-month-old infants during one session but
not during the other. These investigators reported that maternal touch facilitated
attention during face-to-face interactions. However, it is important to note that
in the studies by Roggman and Woodson80 and by Gusella, Muir and Tronick,61
touch was confounded by concurrent stimulation from both visual and auditory
modalities.
In a study that examined distinct types of infant smiles and their relationship to
the social context in which they typically occur, Dickson, Walker and Fogel81 coded
for basic, play and Duchenne smiles during parent-infant interactions. These
investigators found that it was the physical play that included tactile stimulation
that elicited the most “play smiles” (45% of the time), and that these play smiles
occurred less often during object play, vocal play and book reading.
Lastly, in their examination of infant-affective and behavioral states across a series

of conditions—“normal play,” “baby can only see mother,” “baby can only hear
mother,” “baby can only feel mother’s touch” and “baby is alone”—Tronick found
that the lowest levels of infant fussing and crying were displayed in the “touch-
only” condition.28 Interestingly, Tronick also showed a low level of scanning, high
levels of object attention and less smiling by the infants than during the “normal”
and “face-only” conditions. Consistent with the findings reported by my colleagues
and my lab,33,57,65,67 touching had calming effects on the infants, as reflected in
their decreased fretting, and seemed to permit an openness to the stimulation, as
reflected in the high levels of attention and continued smiling. On the bases of
these findings and others, Tronick suggested that touch is a component of the
mutual regulatory process of the caregiver-infant dyad, and he contends that touch
may serve a regulatory function.28,82
The importance of multiple measures and the examination of patterns of responses

from mothers and infants have been emphasized in recent research. Previously,
emphasis was placed on facial expressions, with the result that attention to touch,
gestures, postural changes, vocalizations and the relations between measures has
been inadequate.82 According to functionalist theories of emotion, it is insufficient
to rely solely on facial expressions as indications of affect and arousal. Emotional
signals are context-specific and are shaped by the immediate goals of organism-
environment relationships.83,84 The implications of these findings are threefold:
• must include multiple measures
• must focus on the importance of context
• must include approaches that address discrete measures, relations

between measures and the interaction itself (eg, patterns, sequencing)
According to Toda and Fogel,“emotional” responses in young infants cannot be

judged entirely from the status of the face: Rather, these responses must involve
the whole body and the patterns of temporally organized actions in context.55,56
Weinberg and Tronick provide some support for these contentions.85 In their
study, which examined mothers and their 6-month-old infants, these investigators
examined multiple modalities, including the infants’ gazes, vocalizations, gestures,
facial expressions and self-regulatory and withdrawal behaviors. They found
evidence of behavioral clusters or “affective configurations” that they argued
conveyed information about the infants’ states and intentions. Increased emphasis
on coding and integrating touch and gesture are warranted to better understand
their roles in communication.

Pa t t e r n s o f To u c h i ng
Although the importance of the studies discussed previously is not in question,

these past findings are limited in at least 2 ways. Firstly, the functions and
adaptability of touch have largely been inferred based on evidence taken from
infant responses to their caregivers, rather than from direct measures of caregiver
touch. Secondly, even in those studies in which touch was assessed directly, the
measures of results have largely been the duration of all touching61 or they have
been the levels of intensity.65
Despite the fact that the overall levels of touching are, in fact, important, they do
not inform us about the qualitative aspects of touching or how particular types
of touch may be used more or less often under specific circumstances. It makes a
difference, for example, whether one strokes, caresses, pats or pokes. Moreover, all
touch may not be used or interpreted similarly: Different types of touch and the
way the touch is applied may have different meanings. The way in which touch is
used, and how one is touched, reflects in some way how touch is communicated or
transmitted. Both the quality and quantity of tactile stimulation, as well as the
parameters, become important.68,86-88
To illustrate these issues, my colleagues and I developed the Caregiver-Infant Touch

Scale (CITS) to measure the different types of touch and the associated quantitative
characteristics (eg, intensity and speed) in social contexts, such as mother-infant
play, and to examine changes across age.68 The CITS is designed to code touch,
second by second, from videotapes of play and interactions. After assessing its
psychometric properties, we applied the CITS in an experimental paradigm (SF-
with-touch procedure) that was known to produce reliable shifts in infants’ social
responses.68 This application was designed to test not only the scale’s sensitivity in
describing differential maternal tactile stimulation, but also to determine whether
mothers used different types of touch as a function of different perturbation
periods. Following a period of natural face-to-face interactions, mothers and their
5-month-old infants participated in 3 SF-with-touch periods. The SF-with-touch
perturbation periods included the following: normal touch, touch to maximize
infant smiling and touch restricted to one area of the body. A baseline comparison
group received 4 periods of natural face-to-face interaction.
Analyses revealed several important findings. Firstly, the natural and all SF
(touch-alone) periods were significantly different, and there were clear differences
among all SF periods. Specific patterns or profiles of touching were shown across
perturbation periods: For example, when asked to maximize infant smiling,
mothers used more-active types of touch (eg, lifting and tickling), more surface
area and greater intensity and speed. During the SF period when mothers were
asked to touch their babies in only one area, there was increased stroking and far
less shaking. Touching was also less intense, and most types of touch were judged
to be executed more slowly during this period. Thus, the more tactilely active
profile was revealed during the period when smiling was maximized. This finding
supports the notion of heightened activity during playful interactions. From these
results it is clear that mothers’ profiles of touching change during brief interactions,
as a function of experimenter instruction, suggesting that what was being
communicated through touch was different.
In her comparison of tactile and all-modes experimental and control groups in the
4 instructional contexts described previously, Arnold also examined these periods
for type of touch.73 Among her findings was that, within the tactile groups,
mothers of 3-month-old infants touched their babies more often than did mothers
of 5-month-old infants. Interestingly, there were also differences in the type of
touch used, as a function of instructional period: For example, relative to the
controls, more stroking was used in the excited and happy periods, and more
tickling and shaking were used in the attention-to-face period. Notable effects were
also found for the area of the body on which the touch occurred. Lastly, effects
were also demonstrated in comparisons across modalities.
Table 2: Brief summary of some of Stack and colleagues’ past findings II

11. Mothers change their touching during different conditions
12. Different types of touch are used in different contexts and for different purposes
13. Different types of touch are used across infants’ age

14. There is a relationship between the type of maternal touch and the area of the
infants’ bodies being touched
These findings underscore the value of studying touch and how it can illuminate our
understanding of communication
Consistent with the theme that mothers use qualitatively different types of touch
during interactions with their babies, Tronick reported that mothers use affectively
positive types of touch with their 6-month-old infants.28 In this study, mothers
used stroking, rhythmic touching and holding with their 6-month-old babies for
the largest proportion of time, followed by tickling and kissing. “Negative” forms
of touch, such as pinching and poking, were rarely observed.
Together, these findings suggest that simple touch “duration” is not a sufficient
index with which to characterize adult behavior. Qualitative and quantitative
variations in touching occur for a variety of reasons and are important to measure
and describe. What is also clear is that mothers use different patterns of touching
for different functions.68
Mo t he r -In f a nt To u c h i ng i n At - Risk Pop ulations
There is abundant evidence regarding the effects of touch in preterm infants:

This is addressed in another chapter in this book. Evidence from those few studies
that have examined touch in early development directly supports the claim that
touch is important to the quality of the parent-child relationship and to the overall
development of the infant. This conclusion is also revealed in studies that involve
at-risk infants: For example, Polan and Ward found support that physical affection
and physical interaction were reduced in mothers of failure-to-thrive infants.89
Similarly, positive touch stimulation has enhanced positive affect and attention
in infants of depressed mothers.66 Paradoxically, in their study of parenting
stress, depression and anxiety and its relationship to behaviors exhibited during
interactions between mothers and their infants, Fergus, Schmidt and Pickens
found that mothers who reported more symptoms of depression touched their
infants more frequently relative to nondepressed mothers.90 However, they also
found that the pattern of interaction shown by these depressed mothers was more
intrusive and overstimulatory in nature. Moreover, these investigators reported that
more poking and tickling were used, and these symptomatic mothers, who were
diagnosed with mild-to-moderate levels of depressive symptoms, attempted more
attention-getting strategies, such as finger snapping. Similarly, Cohn and Tronick
described depressed mothers as using more poking and jabbing with their infants,
touching behaviors that were associated with negative affect and gaze aversion on
the part of the infants.91 Combined, these studies point to the importance of the
types of touch that are used by mothers with their infants, and subsequently
converge to suggest that touch might be an important parenting measure.
Using a sample of young mothers during a 3-minute interaction with their

infants, Malphurs and colleagues demonstrated that mothers who had depressive
symptoms were more likely to use negative types of touch, such as rough tickling,
poking, tugging and pulling, and were more likely to be classified as “intrusive” as
compared with “withdrawn” or “good,” their other 2 possible classifications.92 In a
related study, Lundy and colleagues reported no differences in the types of maternal
touching between depressed and nondepressed mothers and their newborns during
a postfeeding period within the second 12 hours following delivery.93
Sensory-impaired infants reveal yet another way with which touch can be used as
an important communicative channel. In these cases, whether the infant is visually
or hearing impaired, the tacto-gestural modality might be seen as assuming some
of the roles that vision or audition might have otherwise subsumed. Of course,
both the visual and tactile channels are of significance to the deaf infant.94 Yet,
according to Meadow-Orlans and Steinberg, when compared with mothers of
18-month-old hearing infants, mothers of same-age deaf infants used less-frequent
and positive touch, and they were less sensitive and more intrusive.95 Nonetheless,
as Koester, Papouek and Brooks discovered, when deaf mothers use these
more-intrusive touch behaviors with their infants a style of communication is
formed to which infants respond.98 MacTurk and coinvestigators also found that
maternal visual-tactile responsiveness was lower in mothers of deaf infants at
9 months: However, this level of visual-tactile responsiveness contributed to
positive interactions at 18 months.97
As part of the Concordia Longitudinal Risk Project,98,99 Bentley100 examined the

relationship between maternal risk status and mother-child touch behavior in an
intergenerational sample of mothers who had a history of childhood aggression
and social withdrawal. Although the mothers’ childhood levels of aggression and/or
social withdrawal did not predict total positive touch behavior significantly, and
little negative touch was observed, the findings revealed that the age of the child
was associated with the frequency that the mothers touched: That is, mothers of
younger children engaged in more touching behaviors than did mothers of older
children, underscoring the important role of touch in early development.
In an empirical study published in 2000, Weiss and colleagues examined maternal

touch in a sample of low-birth-weight infants during a feeding situation at
3 months of age and its subsequent relationship to attachment status at 12 months
of age.101 While no relationship was found between the frequency and overall
duration of touch used at 3 months with infants’ attachment classifications at
1 year, the investigators did demonstrate that healthy 3-month-old infants who
received more nurturing touch were more likely to be classified as “secure” at
1 year: Heavier weight and minimal perinatal risk contributed to the effects. In
another study by Weiss and colleagues, the investigators attempted to establish links
between the early tactile experiences of low-birth-weight children and subsequent
later mental health and social adaptation, highlighting the contribution of touch
to psychosocial development.102
Fa t he r -In f a nt To u c h i ng D u r i ng Inte ractions
Although the majority of “touch” studies involving infants conducted to date have
involved mothers or female adults, there is an accumulating literature involving
fathers. It is known that fathers engage in more vigorous, physically stimulating
play with their infants,103-106 and it is believed by some that their style of play
serves to create a critical means for the development of attachment.107 According
to some research, fathers’ actual touching of their infants is rare immediately after
birth, while physical closeness and gazing at the infant are more typical.108 This
observation of less-frequent touching by fathers, as compared with mothers, has
also been noted in the neonatal intensive care unit.45
There have been only a scant few studies of fathers’ systematic touching of their
infants in different cultures: As an example, Hewlett observed the Aka pygmies
of the tropical forest region of the southern portion of the Central African
Republic.104 While Aka fathers held their infants substantially less often than did
their mothers, several “positive” patterns did emerge. The fathers who held their
infants often did so in a context-specific situation (eg, during leisure time).
Interestingly, however, it was the Aka fathers who were more likely to engage in
minor physical play, such as tickling and bouncing, with their infants. Aka fathers
did not engage in the vigorous types of play characteristic of American fathers.
In their observations of Italian fathers, New and Benigni described fathers’

interactions with their infants as more distal, rather than proximal, and involving
more looking and talking.109 The touches of Italian fathers were described as
“awkward” and “brief,” while their holding was often limited to times when the
mothers were preparing the feedings. Their physical contacts typically included
tickling and poking.
The characteristics of holding and play in different cultures can also be a revealing
means of examining some aspects of physical touch and affection during parent-
infant interchanges. In a study describing Indian (New Delhi) mothers’ and fathers’
holding patterns, Roopnarine and colleagues found that mothers held their babies
more than did their fathers, and they were more likely to pick them up, feed them
and comfort them while holding and display affection while holding.110 However,
the overall duration of holding was reported as less than what is typically reported
for the North American family. This could be explained by the fact that, on a daily
basis, many other family members and friends hold the typical Indian baby. When
fathers were holding their infants, affection was commonly displayed. Tickling
and lap bouncing were found to be rare occurrences between Indian parents and
their infants. However, playing “peekaboo” was seen more commonly among the
mothers and their infants. Lastly, these infants themselves were more likely to
vocalize to, smile at and follow their mothers, compared with their fathers, but
there were no differences in the amounts of touch or approach behaviors between
the mothers and the fathers.
Taken together, findings from this brief overview of some of the father-infant
interaction studies where touch is considered have shown that touch is important—
and even intrinsic—to fathers and to other cultures. Touch can also be used
differently in some situations or be more frequent. These results converge to
suggest that touching is used to bring people together, for closeness and intimacy
and for proximity and play. More research with fathers is warranted, particularly
pertaining to fathers’ use and styles of touching during interactions.
Su mm a r y a n d Co nc l u s i o ns
Taken together, interaction studies have provided important insights and have
advanced our knowledge of the young infant’s sensitivity to manipulations in
facial, vocal and now tactile expressions. These findings emphasize the complexity
and sophistication of mother-infant dyadic interactions and the importance of
including measures of touch. Moreover, through these studies, an abundance of
new findings has been revealed: For example, mothers frequently use touch during
normal face-to-face interactions with their infants (65% of the time).54,65 For brief
periods of time, touch alone can maintain infants’ attention and elicit positive
affect at least as well as can vocal and facial expressions.111 In addition, infants are
sensitive to subtle changes in maternal touch,56,57 and therefore prefer stimulus
compounds that include touch.78 Results from these studies have also enlightened
us about infants’ sensitivity to maternal behavior; in particular, their sensitivity
to their mothers’ touch. Moreover, the influences of modality have been
demonstrated,57,71-73 and different patterns of touching are shown under different
instructional conditions.68,73 Most importantly, perhaps, these studies have provided
evidence for a functional context for touch that is not limited to the regulation
of distress. Rather, these findings indicate a role for touch in social-emotional
development, and they imply that touch serves an important communicative
function. However, the processes and mechanisms through which touch
communicates require additional research attention.
Touch is emerging as a diverse and adaptable modality that, although often used
alone, accompanies other modalities and channels of communication. Touch
is used frequently during the first year of life, serving a multitude of purposes:
• maintaining the infant’s state
• providing comfort and warmth
• serving as a means of social communication
• adjusting posture
• acting as an important means of developing the early

parent-infant relationship
In essence, touch is considered an integral part of the complex communication

system that exists between parents and infants. However, more research is needed
to describe and determine its specific contributions.
Although facial and vocal expressions are important forms of communication, they
are only 2 of the many behaviors that are used to express and communicate. Far
less data exist regarding “tactile expressiveness” and communication through touch.
Field18 reaffirmed the observation that the communicative functions of touch have
been neglected relative to the other senses and to facial and vocal expressiveness—
points also made by Frank,112 myself13 and subsequently by Hertenstein.86
Moreover, in her article in Human Development, Field18 argues the important role
for touch in understanding and improving infants’ well-being, emphasizing that
significant effects on growth, development and emotional well-being are suggested
from numerous studies and many observations of extremes of limited touch in
infants and children. However, little is known about the long-term effects of
parental touch and touch patterns on the subsequent mental health, social and
emotional development and adjustment in healthy infants or more vulnerable
groups of children.
The findings discussed in this chapter emphasize the points that the tactile modality
provides an important means for parents and infants to develop and maintain a
connection with each other, as well as with the environment and to the self. They
also illustrate the flexibility and adaptability of touch, as well as the adaptability of
the communication system, such that both partners can modify their behaviors to
adjust and compensate for the situation: Therefore, both partners are responsive to
each other. It has also been made clear that patterns of touching may be different.
Given that the contexts within which much of early development occurs are social
and multimodal, how touching is integrated with other communication channels
that are available to parents is an important research pursuit in our search for
unraveling the mysteries and challenges of the sense of touch.
Beyond the aforementioned, several additional questions and pivotal issues are
prominent and have emerged from the extant literature.13 Firstly, what is being
communicated through touch? Secondly, assuming that communication is
occurring (based on the evidence to date), and that touch is serving a multitude of
important roles and functions, the next issue involves determining how to measure
this process. Establishing clear roles for, and contributions to, social and emotional
development are important. Thirdly, it is absolutely essential to address the
quantitative and qualitative characteristics (patterns) of touch and their salience
over age, as well as the changes that occur in infants’ and caregivers’ communicative
behaviors (eg, affective, gazing and touching behaviors) over time. Fourthly,
particularly throughout the first few years of life, the infant (and its parents) is
developing, changing and adjusting. It is important for future research to pursue
and be aware of how development itself plays a role in the changes we see related
to touch, physical contact and affection. Fifthly, relationships between patterns
of touching and/or discrete types of touch and emotional/affective displays are
required. Sixthly, the long-term implications for touch and its contributions to
development and adjustment are important endeavors, yet they present difficult
research challenges. The last 2 issues stress the importance of studying individual
differences and examining different trajectories. Lastly, it is essential that the
research be integrated into existing models and theory and that the development of
new, more comprehensive models be added to the mix as the research develops. The
value of such pursuits must not be overemphasized. Hertenstein has taken a positive
step in this direction by describing a general mode for tactile communication.86
The future of touch is promising. While examining touching during mother-infant

interactions is only one means through which to study touch systematically, it is
an enlightening avenue that is invaluable to understanding social-emotional
development in a comprehensive way and to developing the communication system
between parents and their infants. This direction also offers important possibilities
for cross-fertilization and convergence. Indeed, touch offers us a rich world to
discover—one that has implications for a variety of fields of inquiry.
Acknowledgments
This chapter was written with the support of the Social Sciences and Humanities
Research Council of Canada (SSHRC) and Fonds pour la Formation de
Chercheurs et l’Aide à la Recherche (FCAR). The author expresses her gratitude
to Yves Beaulieu for his detailed comments on an earlier version of this chapter.
Thanks are also extended to Julie Pepin and Laurie Gelfand for their help with
literature searches, library work and preparation of the final version of this chapter.
Some of the ideas for, and the content of, this chapter originated from the author’s
2001 studies on touch and physical contact during infancy.13 The publisher of this
handbook is also gratefully acknowledged.
Ref ere n c e s
1. Montagu A. Touching: The Human Significance of the Skin. New York, NY: Columbia
University Press; 1971.
2. Hamnet FS. Studies in the thyroid apparatus. American Journal of Physiology. 1921;56:
196-204.
3. Hamnett FS. Studies of the thyroid apparatus. Endocrinology. 1922;6:221-229.

4. Denenberg VH. A consideration of the usefulness of the critical period hypothesis as applied
to the stimulation of rodents in infancy. In: Newton G, Levin S, eds. Early Experience and
Behavior. Springfield, Ill: Charles C Thomas; 1968:142-167.
5. Hofer M. Developmental roles of timing in the mother-infant interaction. In: Turkewitz G,

Devenny DA, eds. Developmental Time and Timing. Hillsdale, NJ: Lawrence Erlbaum
Associates; 1993:211-231.
6. Kuhn CM, Schanberg SM. Responses to maternal separation: mechanisms and mediators.
International Journal of Developmental Neuroscience: The Official Journal of the International
Society for Developmental Neuroscience. 1998;16:261-270.
7. Schanberg SM, Field TM. Sensory deprivation stress and supplemental stimulation in the rat
pup and preterm human neonate. Child Development. 1987;58:1431-1447.
8. Denenberg VH. The effects of early experience. In: Hafez ESE, ed. The Behavior of Domestic
Animals. 2nd ed. London, England: Baillure, Tindall and Cox; 1969:95-130.
9. Harlow HF. Love in infant monkeys. Scientific American. 1959;200:68.
10. Levine S. A further study of infantile handling and adult avoidance learning. Journal of
Personality. 1956;25:70-80.
11. Levine S. Stimulation in infancy. Scientific American. 1960;202:80.

12. Levine S, Stanton ME. The hormonal consequences of mother-infant contact. In: Barnard
KE, Brazelton TB, eds. Touch: The Foundation of Experience. Vol 4. Madison, Wis:
International Universities Press; 1990:165-194.
13. Stack DM. The salience of touch and physical contact during infancy: unraveling some of the
mysteries of the somesthetic sense. In: Bremner G, Fogel A, eds. Blackwell Handbook of Infant
Development. Malden, Mass: Blackwell Publishers; 2001:351-378.
14. Montagu A. Touching: The Human Significance of the Skin. 3rd ed. New York, NY: Harper
and Row Publishers; 1986.
15. Field T. Touch. Cambridge, Mass: MIT Press; 2001.

16. Provence S, Lipton RC. Infants in Institutions: A Comparison of Their Development With
Family-Reared Infants During the First Year of Life. New York, NY: International Universities
Press; 1962.
17. Spitz RA, Wolf KM. Anaclitic depression. An inquiry into the genesis of psychiatric
conditions in early childhood II. The Psychoanalytic Study of the Child. 1946;2:313-342.
18. Field T. Infants’ need for touch. Human Development. 2002;45:100-103.
19. Rutter M. Developmental catch-up, and deficit, following adoption after severe global early
privation. English and Romanian Adoptees (ERA) Study Team. Journal of Child Psychology
and Psychiatry, and Allied Disciplines. 1998;39:465-476.
20. Brazelton TB. Touch as a touchstone: summary of the round table. In: Barnard KE, Brazelton
TB, eds. Touch: The Foundation of Experience. Vol 4. Madison, Wis: International Universities
Press; 1990:561-566.
21. Birns B, Blank M, Bridger WH. The effectiveness of various soothing techniques on human
neonates. Psychosomatic Medicine. 1966;28:316-322.
22. Korner AF, Thoman EB. The relative efficacy of contact and vestibular-proprioceptive
stimulation in soothing neonates. Child Development. 1972;43:443-453.
23. Barrera ME, Maurer D. The perception of facial expressions by the three-month-old. Child
Development. 1981;52:203-206.
24. Muir D, Field J. Newborn infants orient to sounds. Child Development. 1979;50:431-436.
25. Brazelton TB. Introduction. In: Brown CC, ed. The Many Facets of Touch. Skillman, NJ:
Johnson & Johnson Baby Products Co; 1984:xv-xviii. Pediatric Round Table Series, No. 10.
26. Koester LS, Papouek H, Papouek M. Patterns of rhythmic stimulation by mothers with
three-month-olds: a cross-modal comparison. International Journal of Behavioral Development.
1989;12:143-154.
27. Montagu A. Touching: The Human Significance of the Skin. 3rd ed. New York, NY: Harper
and Row Publishers; 1986.
28. Tronick EZ. Touch in mother–infant interaction. In: Field TM, ed. Touch in Early
29. Hertenstein MJ, Campos JJ. Emotion regulation via maternal touch. Infancy. 2001;2:
549-566.
30. Ainsworth MD, Blehar MC, Waters E, Wall S. Patterns of Attachment: A Psychological Study
of the Strange Situation. Hillsdale, NJ: Lawrence Erlbaum Associates; 1978.
31. Bowlby J. Attachment & Loss. Vol 1. New York, NY: Basic Books; 1969.
32. Landau R. Affect and attachment: kissing, hugging, and patting as attachment behaviors.
Infant Mental Health Journal. 1989;10:52-69.
33. Stack DM, Muir DW. Adult tactile stimulation during face-to-face interactions modulates
five-month-olds’ affect and attention. Child Development. 1992;63:1509-1525.
34. Isabella RA, Belsky J, von Eye A. Origins of infant-mother attachment: an examination
of international synchrony during the infant’s first year. Developmental Psychology.
1989;25:12-21.
35. Anisfeld E, Lipper E. Early contact, social support, and mother-infant bonding. Pediatrics.
1983;72:79-83.
36. Grossmann K, Thane K, Grossman KE. Maternal tactual contact of the newborn after
various postpartum conditions of mother-infant contact. Developmental Psychology.
1981;17:158-169.
37. Schaller J, Carlsson SG, Larsson K. Effects of extended post-partum mother-child contact
on the mother’s behavior during nursing. Infant Behavior & Development. 1979;2:319-324.
38. Cannon RB. The development of maternal touch during early mother-infant interaction.
Journal of Obstetric, Gynecologic, and Neonatal Nursing: JOGNN/NAACOG. 1977;6:28-33.
39. Carlsson SG, Fagerberg H, Horneman G, et al. Effects of amount of contact between mother
and child on the mother’s nursing behavior. Developmental Psychobiology. 1978;11:143-150.
40. de Chateau P. The influence of early contact on maternal and infant behaviour on
primiparae. Birth and the Family Journal. 1976-1997;3:149-155.
41. Klaus MH, Kennell JH, Plumb N, Zuehlke S. Human maternal behavior at the first contact
with her young. Pediatrics. 1970;46:187-192.
42. Rubin E. Maternal touch. Nursing Outlook. 1963;11:328-331.

43. Trevathan WR. Maternal touch at 1st contact with the newborn infant. Developmental
Psychobiology. 1981;14:549-558.
44. Klaus MH, Jerauld R, Kreger NC, McAlpine W, Steffa M, Kennel JH. Maternal attachment.
Importance of the first post-partum days. The New England Journal of Medicine.
1972;286:460-463.
45. Harrison LL, Woods S. Early parental touch and preterm infants. Journal of Obstetric,
Gynecologic, and Neonatal Nursing: JOGNN/NAACOG. 1991;20:299-306.
46. Robin M. Neonate-mother interaction: tactile contacts in the days following birth. Early
Child Development and Care. 1982;9:221-236.
47. Field TM. Massage therapy effects. The American Psychologist. 1998;53:1270-1281.
48. Field T. Touch therapy effects on development. International Journal of Behavioral
Development. 1998;22:779-797.
49. Field T. Touch Therapy. New York, NY: Churchill Livingstone; 2000.
50. Kaye K, Fogel A. The temporal structure of face-to-face communication between mothers
and infants. Developmental Psychology. 1980;16:454-464.
51. Field TM, Vega-Lahr N, Scafidi F, Goldstein S. Effects of maternal unavailability on
mother-infant interactions. Infant Behavior & Development. 1986;9:473-478.
52. Cohn JF, Tronick EZ. Three-month-old infants’ reaction to simulated maternal depression.
53. Field TM. Early interactions between infants and their postpartum depressed mothers.
Infant Behavior & Development. 1984;7:517-522.
54. Symons DK, Moran G. The behavioral dynamics of mutual responsiveness in early
face-to-face mother-infant interactions. Child Development. 1987;58:1488-1495.
55. Fogel A, Dedo JY, McEwen I. Effect of postural position and reaching on gaze during
mother-infant face-to-face interaction. Infant Behavior & Development. 1992;15:231-244.
56. Toda S, Fogel A. Infant response to the still-face situation at 3 and 6 months. Developmental
Psychology. 1993;29:532-538.
57. Stack DM, Arnold SL. Changes in mothers’ touch and hand gestures influence infant
behavior during face-to-face interchanges. Infant Behavior & Development. 1998;21:451-468.
58. Lavelli M, Fogel A. Developmental changes in early mother-infant face-to-face
communication (abstract). In: Program and abstracts of the Biennial Meeting of the
International Conference on Infant Studies; April 2-5, 1998; Atlanta, Ga.
59. Stack DM, Arnold SL, Girouard N, Welburn B. Infants’ reactions to maternal unavailability
in very low birth weight preterm and full-term infants. Paper presented at: the Biennial
Meeting of the Society for Research in Child Development; April, 1999; Albuquerque, NM.
60. Tronick EZ, Als H, Adamson L, Wise S, Brazelton TB. The infant’s response to entrapment
between contradictory messages in face-to-face interactions. Journal of the American Academy
of Child Psychiatry. 1978;17:1-13.
61. Gusella JL, Muir DW, Tronick EZ. The effect of manipulating maternal behavior during
an interaction on three- and six-month-olds’ affect and attention. Child Development.
1988;59:1111-1124.
62. Lamb ME, Morrison DC, Malkin CM. The development of infant social expectations in
face-to-face interaction: a longitudinal study. Merrill-Palmer Quarterly. 1987;33:241-254.
63. Mayes LC, Carter AS. Emerging social regulatory capacities as seen in the still-face situation.
64. Ellsworth CP, Muir DW, Hains SM. Social competence and person-object differentiation:
an analysis of the still-face effect. Developmental Psychology. 1993;29:63-73.
65. Stack DM, Muir DW. Tactile stimulation as a component of social interchange: new
interpretations for the still-face effect. British Journal of Developmental Psychology.
1990;8:131-145.
66. Peláez-Nogueras M, Field TM, Hossain Z, Pickens J. Depressed mothers’ touching increases
infants’ positive affect and attention in still-face interactions. Child Development.
1996;67:1780-1792.
67. Stack DM, LePage DE. Infants’ sensitivity to manipulations of maternal touch during
face-to-face interactions. Social Development. 1996;5:41-55.
68. Stack DM, LePage DE, Hains SM, Muir DW. Qualitative changes in maternal touch as a
function of instructional condition during face-to-face social interactions. Poster presented at
the 10th Biennial International Conference on Infant Studies; April 18-21, 1996; Providence,
RI.
69. LePage DE. Four- and 7-Month-Old Infants’ Sensitivities to Contingency During Face-to-Face
Social Interactions [dissertation]. Montréal, Québec, Canada: Concordia University; 1998.
70. LePage DE, Stack DM. Four- and 7-month-old infants’ abilities to detect tactile
contingencies in a face-to-face context (abstract). In: Program and abstracts of the Biennial
Meeting of the Society for Research in Child Development; April 3-6,1997; Washington,
DC.
71. Arnold SL, Brouillette J, Stack DM. Changes in maternal and infant behavior as a function
of instructional manipulations during unimodal and multimodal interactions (abstract).
In: Program and abstracts of the Biennial Meeting of the International Society for the Study
of Behavioral Development; August 12-16, 1996; Quebec City, Quebec, Canada.
72. Stack DM. Touching during mother-infant face-to-face interactions: influences of context
and modality. Paper presented at: The Touch Research Institutes Symposium Preconference,
XIIIth Biennial International Conference on Infant Studies; April 18-21, 2002; Toronto,
Ontario, Canada.
73. Arnold SL. Maternal Tactile-Gestural Simulation and Infants Nonverbal Behaviors During Early
Mother-Infant Face-to-Face Interactions: Contextual, Age, and Birth Status Effects [dissertation].
Montréal, Québec, Canada: Concordia University; 2002.
74. Roedell WC, Slaby RG. The role of distal and proximal interaction in infant social preference
formation. Developmental Psychology. 1977;13:266-273.
75. Leiba E. Maternal Nonverbal Behaviors and Infant Gaze During Triadic Play With Toys at
5 and 12 Months [master’s thesis]. Montréal, Québec, Canada: Concordia University; 2000.
76. Leiba E, Stack DM. Maternal nonverbal behaviors and affectionate displays during play at
5 and 12 months (abstract). In: Program and abstracts of the Canadian Psychological
Association Annual Convention; June 21-23, 2001; Sainte-Foy, Quebec, Canada.
77. Peláez-Nogueras M. Rhythmic and nonrhythmic touch during mother infant interactions.
(abstract). In: Program and abstracts of the Society for Research in Child Development;
March 30-April 2, 1995; Indianapolis, Ind.
78. Peláez-Nogueras M, Gewirtz JL, Field T, et al. Infants’ preference for touch stimulation
in face-to-face interactions. Journal of Applied Developmental Psychology. 1996;17:199-213.
79. Wolff PH. Observations on the early development of smiling. In: Foss BM, ed. Determinants
of Infant Behavior II. London, England: Methuen & Co Ltd; 1963:113-138.
80. Roggman LA, Woodson R. Touch and gaze in parent-infant play (abstract). In: Program and
abstracts of the Society for Research in Child Development Conference; April 27-30, 1989;
Kansas City, Kan.
81. Dickson KL, Walker H, Fogel A. The relationship between smile type and play type during
parent-infant play. Developmental Psychology. 1997;33:925-933.
82. Tronick EZ. Emotions and emotional communication in infants. The American Psychologist.
1989;44:112-119.
83. Campos JJ, Mumme DL, Kermoian R, Campos RG. A functionalist perspective on the
nature of emotion. Monographs of the Society for Research in Child Development. 1994;59:
284-303.
84. Thompson RA. Socioemotional development: enduring issues and new challenges.
Developmental Review. 1993;13:372-402.
85. Weinberg MK, Tronick EZ. Beyond the face: an empirical study of infant affective
configurations of facial, vocal, gestural, and regulatory behaviors. Child Development.
1994;65:1503-1515.
86. Hertenstein MJ. Touch: its communicative functions in infancy. Human Development.
2002;45:70-94.
87. Weiss SJ. Measurement of the sensory qualities in tactile interaction. Nursing Research.
1992;41:82-86.
88. Weiss SJ, Campos R. Touch. In: Lindeman CA, McAthie M, eds. Fundamentals of
Contemporary Nursing Practice. Philadelphia, Pa: WB Saunders Company; 1999:941-967.
89. Polan HJ, Ward MJ. Role of the mother’s touch in failure to thrive: a preliminary
investigation. Journal of the American Academy of Child and Adolescent Psychiatry.
1994;33:1098-1105.
90. Fergus EL, Schmidt J, Pickens J. Touch during mother-infant interactions: the effects of
parenting stress, depression and anxiety (abstract). In: Program and abstracts of the Biennial
Meeting of the International Society of Infant Studies; April, 1998; Atlanta, Ga.
91. Cohn JF, Tronick E. Specificity of infants’ response to mothers’ affective behavior. Journal
of the American Academy of Child and Adolescent Psychiatry. 1989;28:242-248.
92. Malphurs JE, Raag T, Field T, Pickens J, Peláez-Nogueras M. Touch by intrusive and
withdrawn mothers with depressive symptoms. Early Development & Parenting. 1996;5:
111-115.
93. Lundy BL, Field T, Cuadra A, Nearing G, Cigales M, Hashimoto M. Mothers with
depressive symptoms touching newborns. Early Development & Parenting. 1996;5:129-134.
94. Koester LS. Effects of maternal or infant deafness on early interaction patterns (abstract).
In: Program and abstracts of the Biennial Meeting of the International Conference on Infant
Studies; May 6-10, 1992; Miami, Fla.
95. Meadow-Orlans KP, Steinberg AG. Effects of infant hearing loss and maternal support
on mother-infant interactions at 18 months. Journal of Applied Developmental Psychology.
1993;14:407-426.
96. Koester LS, Papouek H, Brooks L. The role of tactile contact in deaf and hearing
mother-infant dyads (abstract). In: Program and abstracts of the Biennial Meeting of the
Society for Research in Child Development; March 2-April 2, 1995; Indianapolis, Ind.
97. MacTurk RH, Meadow-Orlans KP, Koester LS, Spencer PE. Social support, motivation,
language, and interaction. A longitudinal study of mothers and deaf infants. American Annals
of the Deaf. 1993;138:19-25.
98. Schwartzman AE, Ledingham JE, Serbin LA. Identification of children at risk for adult
schizophrenia: a longitudinal study. International Review of Applied Psychology. 1985;34:
363-380.
99. Serbin LA, Cooperman JM, Peters PL, Lehoux PM, Stack DM, Schwartzman AE.
Intergenerational transfer of psychosocial risk in women with childhood histories of
aggression, withdrawal, or aggression and withdrawal. Developmental Psychology.
1998;34:1246-1262.
100. Bentley VM. Maternal Childhood Risk as a Predictor of Emotional Availability and Physical
Contact in Mother-Child Interactions: an Intergenerational Study [master’s thesis]. Montréal,
Québec, Canada: Concordia University; 1997.
101. Weiss SJ, Wilson P, Hertenstein MJ, Campos R. The tactile context of a mother’s caregiving:
implications for attachment of low birth weight infants. Infant Behavior & Development.
2000;23:91-111.
102. Weiss SJ, Wilson P, St Jonn-Seed M, Paul SM. Early tactile experience of low birth weight
children: links to later mental health and social adaptation. Infant and Child Development.
2001;10:93-115.
103. Arco CM. Infant reactions to natural and manipulated temporal patterns of paternal
communication. Infant Behavior & Development. 1983;6:391-399.
104. Hewlett BS. Intimate fathers: patterns of paternal holding among Aka pygmies. In: Lamb
ME, ed. The Father’s Role: Cross-cultural Perspectives. Hillsdale, NJ: Lawrence Erlbaum
Associates; 1987:295-330.
105. Parke RD, O’Leary S. Family interaction in the newborn period: some findings, some
observations and some unresolved issues. In: Riegel KF, Meacham J, eds. The Developing
Individual in a Changing World. Vol 1. The Hague, The Netherlands: Mouton; 1976:
653-663.
106. Yogman MW. Development of the father-infant relationship. In: Fitzgerald HE, Lester BM,
Yogman MW, eds. Theory and Research in Behavioral Pediatrics. Vol 1. New York, NY:
Plenum Press; 1982:221-279.
107. Lamb ME. The Role of the Father in Child Development. New York, NY: John Wiley; 1981.
108. Tomlinson PS, Rothenberg MA, Carver LD. Behavioral interaction of fathers with infants
and mothers in the immediate postpartum period. Journal of Nurse-Midwifery. 1991;36:
232-239.
109. New RS, Benigni L. Italian fathers and infants: cultural constraints on paternal behavior.
In: Lamb ME, ed. The Father’s Role: Cross-cultural Perspectives. Hillsdale, NJ: Lawrence
Erlbaum Associates; 1987:139-167.
110. Roopnarine JL, Talukder E, Jain D, Joshi P, Srivastav P. Characteristics of holding, patterns
of play, and social behaviors between parents and infants in New Delhi, India.
111. LePage DE, Stack DM. Do manipulations of maternal face, voice and touch differentially
effect infant responses? (abstract). In: Program and abstracts of the Canadian Psychological
Association; June 11-13, 1992; Québec City, Canada.
112. Frank LK. Tactile communication. Genetic Psychology Monographs. 1957;56:209-255.
113. Stack DM. Qualitative components of maternal touch during mother-infant interactions:
application of the Caregiver-Infant Touch Scale. Invited speaker at the Touch Research
Institutes Symposium Preconference at the Biennial Meeting of the Society for Research
in Human Development: April 24-27, 2003; Tampa, Fla.
114. Casler CR. Maternal deprivation: a critical review of the literature. Monographs of the Society
for Research in Child Development. 1961;26:2.
115. Casler CR. Perceptual deprivation in institutional settings. In: Newton G, Levine S, eds.
Early Experience and Behavior. New York, NY: Springer; 1968.
116. Ainsworth MDS. The effects of maternal deprivation: a review of findings and controversy
in the context of research strategy. Deprivation of maternal care: a reassessment of its effects.
Geneva, Switzerland: World Health Organization; 1962.
117. Byrnes JM, Horowitz FD. Rocking as a soothing intervention: the influence of direction
and type of movement. Infant Behavior and Development. 1981;4:207-218.
CHAPTER 5:
DIFFERENCES ACROSS
C U LT U R A L G R O U P S
IN MOTHERS’ NONVERBAL
TEACHING METHODS
Mary Martini, PhD

Department of Family and Consumer Sciences
University of Hawaii at Manoa
Ab s t ra c t
The primary purpose of the study described in this chapter was to determine how
mothers used touch to “teach” their infants; that is, to help direct their attention
and shape their actions. The study also examined the changes in mothers’ preferred
teaching methods from infancy through toddlerhood and described the cultural
differences in nonverbal teaching strategies. Mothers in 4 American cultural
groups—Caucasian, Japanese, Filipino and Hawaiian—participated in the
study, which assessed how the individuals in each of these groups touched, held,
repositioned and physically interacted with their infants and toddlers (3 months
to 36 months of age) during videotaped mealtimes and how they organized the
mealtime setting. The teaching methods used by the different groups and the
results of this study are explored in detail in this chapter.
Introd uc t i o n
Parents teach children in many ways.1 They instruct children directly, they explain
to them how things work and they show them how to solve problems. Parents
naturally encourage their children to see and respond to reality in the same ways
that they do and to adopt the same “goals for action” that they pursue. Parents also
model actions that they want their children to copy. In addition, they structure
their children’s day-to-day lives so that they will engage in activities that the parents
feel are important and so that their children will come into contact with good
teachers and role models.
Bruner described carefully how parents teach their infants nonverbally.2 He noted
that they hold and move their babies’ limbs in ways that are valued culturally, and
that they direct their infants to attend only to certain objects and perform only
certain “goal-directed actions.” For example, mothers teach their babies that “bells
are for ringing” by wrapping the baby’s fingers around the stem of the bell, shaking
the bell and showing excitement when the bell rings. The baby then takes on this
goal for action and stops “mouthing” bells in order to ring them. Similarly, babies
learn to bang with mallets, to stir with spoons, to match shapes to slots and to say
“hi” and “bye-bye” into telephones.
Harkness and Super pointed out that parents teach their children by structuring
their environments.1 Different sets of parents in this study set up mealtimes quite
differently. Mealtimes differed in one or more of the following ways: whether
the family sat together at a table; whether adults and children ate together; how
many adults were present; how many other children were present; whether
infants/toddlers were constrained to high chairs, held on laps or allowed to roam;
whether mothers attended exclusively to infants; and whether the television was
turned on. Parents effectively convey the messages of their cultural and social groups
and their personal beliefs and habits by carefully (albeit, often subconsciously)
structuring their children’s environments and activity settings and by guiding their
children’s actions.
Chapter 5: Differences Across Cultural Groups in Mothers’ Nonverbal Teaching Methods 85
Cur re n t St u d y 3
METHOD
DATA COLLECTION
Families were recruited from 2 preschools and one elementary school in Honolulu,
Hawaii, for a study of dual-earner families. Parents set up a camera provided
to them, pointed it at their typical mealtime area and turned it on for at least
20 minutes during mealtimes. Videotapes from 80 families who had at least one
young child (ages 3 months to 36 months) were viewed and coded for this study.
PARTICIPANTS
Twenty families were studied from each of the following 4 American cultural
groups: Caucasian American, Japanese American, Filipino American and Hawaiian
American. The mothers in these families had stated on a prestudy questionnaire
that they identified most closely with that cultural group.
Within each cultural group in this sample, 10 families had an infant less than 23
months of age, while the other 10 families had an older toddler, from 24 months
to 36 months of age. Equal numbers of male and female infants/toddlers were
studied within each cultural group. The groups did not differ significantly in terms
of average age of the infants, gender of the infants, number of people present at
the meal, number of parents present at the meal and number of siblings present
at the meal.
CODING
After repeated viewings of a set of these videotapes, a coding system was developed
to define observed forms of touch, seating patterns for the infants and other setting
features. Two observers coded the mothers’ actions during 10 minutes of each
videotaped mealtime. They coded the 10 minutes that followed the first 3 minutes
of each videotape, thereby omitting coding on that part of the tapes during which
families typically adjusted to the camera while setting up their meals.
ANALYSIS
A one-way analysis of variance was used to compare age groups to each other,
cultural groups to each other and males to females. Data were analyzed to compare
the effects of mealtime setting features, the amounts of touch and the forms
of touch. Comparisons involved the extent to which mothers touched infants
in particular ways in these groups.
R E S U LT S
HOW DO MOTHERS SHAPE (“TEACH”) INFANT ATTENTION AND ACTIONS

THROUGH HOLDING AND TOUCH?
During meals, the mothers used the following 4 major forms of touch, which
reflect 4 forms of teaching:
• Instrumental touch—consisted of actions performed upon the baby for

purposes of feeding, cleaning, comforting or caring for the infant: For
example, a mother adjusted her infant’s bottle, spoon-fed her, wiped her
face and repositioned her to feed her more easily. With instrumental
touch, parents showed their infants the routines of mealtime: The spoon
was put into the bowl, picking up food. Then, the food-laden spoon was
put into the infant’s mouth. The mouth then got wiped with the spoon
or washed with the cloth. Lastly, at the end of the meal, the infant’s
hands got wiped. The babies developed expectancies about these actions
and signaled to their mothers to do them by performing their steps in
these sequences (eg, holding up their “dirty” hands for wiping).
• Directive touch—consisted of the mother physically shaping what she

wanted the child to do, or physically blocking the child from performing
unwanted actions. Infants participated more actively during cases of
directive touch. For example, a mother shaped her baby’s hand around
the handle of his feeding spoon and then moved his hand toward his
mouth. Another mother pushed her baby’s hands away when he reached
for his oatmeal. In this way, children took on their parents’ “goals for
actions.”
• Responsive touch—consisted of actions by a mother to help her child

complete actions the child initiated: For example, a mother gave her
baby a toy the infant had reached for and helped the baby place a puzzle
piece she had struggled to place. Mothers often converted responsive
touch into more directed structuring by subtly shaping their infants’
self-initiated actions to bring them more in line with what the
parents wanted: For example, when a baby spooned oatmeal onto his
high-chair tray, his mother moved a bowl underneath and subsequently
complimented him for “putting” the oatmeal in the bowl.
• Supportive touch—consisted of comforting and affectionate, playful

touching, such as kissing, hugging, patting the infant on the head,
nuzzling, rocking, tickling, jiggling and actions involved in playing
games. These actions maintained (supported) the mother-child
relationship and kept the baby focused on the mother.
WHAT KINDS OF NONVERBAL TEACHING DID MOTHERS USE MOST FREQUENTLY?
It was determined that mothers touched their infants an average of 2.11 times per
minute. As Figure 1 shows, most instances of touch consisted of “instrumental
touching” (at an average rate of 1.33 instances/minute), during which the babies
experienced passively how feeding or cleaning should proceed. The second most
frequently used forms of touch were “supportive” (affectionate) and “responsive”
touch combined (at an average rate of 0.55 instances/minute). Touch was used least
frequently to “direct” or “restrict” infants (at an average of 0.24 instances/minute).
DID FORMS OF TEACHING CHANGE AS I NFANTS GOT OLDER?
As indicated in Figure 2, mothers used significantly more nonverbal teaching

(“touch”) with infants than with toddlers and more verbal instruction with
toddlers than with infants. Mothers touched infants statistically significantly
more than they touched toddlers (3.4 touches/minute versus 0.8 touches/minute,
respectively; P<.0001). They used “affectionate” touching statistically significantly
more with infants than with toddlers (1.10 touches/minute versus 0.12
touches/minute, respectively; P<.04). They also touched infants for “instrumental”
purposes (to feed and care for them) statistically significantly more than they
touched toddlers (P<.0001) for these purposes.
Figure 1. Observed forms of touch during mealtimes among mothers and their infants
in different cultural groups.3
Figure 2. The types of touch and the frequency with which mothers from different
cultural groups touched their infants versus their toddlers during mealtimes.3
Naturally, from early infancy to late toddlerhood, mothers’ teaching methods

changed from mainly performing actions and routines on the passive child—
demonstrating the structure of the valued routine—to shaping the infant and
toddler into participating actively in this shared routine. This then progressed to
instructing the child verbally to perform the steps of the routine without using
much touch or positioning in these instructions. Instruction becomes removed
increasingly from the child’s body.
DID PARENTS FROM DIFFERENT CULTURAL GROUPS STRUCTURE MEALTIMES

DIFFERENTLY?
The manners in which mothers from the various cultural groups involved in this
study structured their mealtime settings affected how they interacted with their
infants. The structuring of the activity setting and guiding children through the
activities are ways of conveying cultural preferences and messages.
Caucasian American mothers and Japanese American mothers tended to place their
infants and toddlers in high chairs in order to feed them (Figure 3). In addition,
Japanese American parents tended to provide their babies with play toys during
feeding. Caucasian American parents tended not to provide toys, explaining that
they thought the toys would distract their babies from eating.
Throughout mealtimes, Japanese American mothers tended to sit directly in front

of their children’s high chairs and pay undivided attention to their babies. In most
of these families, the baby was fed before the rest of the family sat down to eat, so
that the mother’s attention could be directed completely to the baby (Figure 4).
In the remaining Japanese American families, the father interacted with the older
children while the mother fed the baby.
Caucasian American parents tended to place the high chair at the periphery of
the larger family meal gathering. Cacausian American infants were commonly
given finger foods and were attended to sporadically. These parents attended most
consistently to their children who were verbal, as meals were seen as times for
exchanging information.
Figure 3. Cultural-group differences regarding mothers placing their infants or

toddlers in high chairs during mealtimes.
FA = Filipino American; JA = Japanese American; CA = Caucasian American; HA = Hawaiian American
Figure 4. Cultural-group differences among mothers devoting undivided attention to

their infants during mealtimes.

Filipino American mothers tended to hold their infants and toddlers on their laps,
facing outward. From this position—being fed from behind—Filipino American
babies were touched much more frequently than were infants from the other
cultures studied. Since she could not see her child’s facial cues, the Filipino
American mother fed her child on her own schedule and not in relation to
readiness cues from her child.
Hawaiian American parents held their babies on their laps to feed them. They also
“passed” their infants to other adults and older children sitting around the meal
table. They allowed their toddlers to “roam” during meals (Figure 5), and they
were permitted to come and go to a central adult holding the toddler’s food bowl.
In this way, the child experienced maximum autonomy in deciding what and when
to eat. Roaming occurred more frequently in the Hawaiian American group than
in any of the other groups studied.
DID MOTHERS FROM DIFFERENT CULTURAL GROUPS USE DIFFERENT NONVERBAL

TEACHING METHODS?
Filipino American: As seen in Figure 6, Filipino American mothers touched their

infants the most frequently (average, 3.17 touches/minute). This outcome was
largely because they held their infants on their laps more frequently, and physically
manipulated their infants into performing desired actions. Filipino American
mothers used the “directive” form of teaching/touching more than did mothers
in the other groups: For example, these mothers kept feeding their babies even
after the infants signaled that they did not want any more food. They also stopped
their babies from touching objects or food, spoon-fed their older toddlers and
discouraged self-feeding.
Japanese American: Japanese American mothers touched their infants the next
most frequently (mean, 2.57 touches/minute). During mealtimes, they placed their
infants in high chairs more frequently than did Filipino American or Hawaiian
American parents, and then attended to them carefully. Japanese American mothers
used the “responsive” form of teaching/touching more often than did mothers
in the other groups. They facilitated exploratory play and helped their children
complete self-initiated goals. And, although they did not block their children’s
unwanted actions, they reshaped them subtly until they were more acceptable.
Figure 5. Cultural-group differences among mothers who allowed their infants to

roam or to sit in an open chair during mealtimes.
Figure 6. Cultural-group differences in rates of touching among mothers and their

infants during mealtimes.3

Caucasian American: Caucasian American mothers touched their infants third

most frequently (mean, 1.51 touches/minute) during this study. They attended
more carefully to their husbands and verbal children, and therefore less carefully to
their infants and toddlers. Caucasian American mothers encouraged self-reliance by
providing finger food, and they allowed their infants to explore and play with their
food. They encouraged their toddlers, in particular, to develop “goals for action”
that were the opposite of expected actions by laughing at these actions.
Hawaiian American: Of the 4 cultural groups studied, Hawaiian American

mothers touched their children least during mealtimes (average, 1.21
touches/minute). They allowed the most free-ranging wandering and exploration,
and their children were often “out of touch.” They also attended more to other
adults who were present during mealtimes. At times, the Hawaiian American
mothers encouraged “rascal”-type violations of social expectations. At other times,
they scolded their toddlers for disrupting the group.
WERE PARENTS’ PREFERRED TEACHING METHODS SUPPORTED BY HOW THEY

STRUCTURED MEALTIMES?
Trial-and-error, exploratory learning—during which children were encouraged to

learn by doing—was facilitated by the following conditions: children being free to
move around; parents providing toys, utensils and finger foods to infants; parents
allowing, and even praising, their child’s explorations; and parents not monitoring
their children closely, particularly as parents engaged in other conversations. Hawaiian
American parents allowed these conditions to occur the most frequently, while
Filipino American parents allowed these conditions to occur the least frequently.
Shaped, exploratory learning—during which infants were allowed to initiate their

own “projects,” but were then structured into turning these into conventional
actions—was facilitated by the following conditions: infants being “restrained”
in high chairs where they could be monitored; mothers attending to their infants
carefully; mothers helping their infants reach their signaled goals; and parents
providing toys, utensils and food to restrained children, and letting them play
with these items. Japanese American parents allowed these conditions to occur
the most frequently.
Adult-directed learning—during which parents introduced and shaped the skills

to be learned by their infants—was facilitated by the following conditions: infants
and toddlers being “restrained” to a parent’s lap; mothers monitoring their infants
and toddlers carefully; parents providing their children with few objects to play
with; mothers intervening quickly to stop their children’s unwanted actions; and
mothers shaping their infants and toddlers physically to perform the actions that
the parents want. Filipino American mothers allowed these conditions to occur
the most frequently.
Cu ltura l Me ss a g e s Ab o u t To u c h
Parents in the various cultural groups studied in this study conveyed different
messages about togetherness and independence during mealtimes. Filipino
American mothers and Japanese American mothers were highly affectionate and
highly controlling toward their infants during mealtimes. They, in fact, taught their
infants that being a member of a tight-knit family group involves both the warmth
of belonging and the obligations of conforming to group ways.
Although Caucasian American mothers were more distant relationally, they allowed
their infants a greater sense of autonomy and choice. This autonomy involved
reduced closeness but increased individual self-expression through talk. Whereas,
even though Hawaiian American mothers enabled autonomy, they also allowed
their infants to seek more contact when their babies wanted such contact.
These groups of American parents also sent messages to their infants about the
appropriate ways to make contact with others. Caucasian American parents, in
particular, emphasized making contact through talking during mealtimes. They
effectively “separated” mealtime members into formal, boundaried spaces at the
dinner table, and then tried to close the distance with conversation. Therefore,
it is not surprising that Caucasian American children learn to talk early, loudly
and often.
Conclusions
In this study,3 the mean rate at which American mothers touched their infants and
toddlers during mealtimes was very low. Mealtimes may be settings during which
parents encourage independence and discourage touch. The low rate of touching,
the decrease in touch as children got older and, in particular, the decrease in
affectionate touch may reflect current American patterns of increasing physical
distance among people as they mature.
Touching their infants was most frequent among the most traditional mothers;
that is, those who held their infants on their laps. However, these were also the
most “directive” mothers. Touching was least frequent in mothers who rely on
baby furniture, such as high chairs and booster chairs: High chairs emphasize the
“separateness” of baby and mother, while they also constrain the infant. While they
allow the baby to sit and eat by himself/herself, freeing up the mother, they also
keep the infant from seeking more contact when he/she wants it.
Mothers who sat on the floor or on low chairs while their infants or toddlers
roamed touched their children more, in large part because the infants approached
the adults for more touch. Babies who roamed and “grazed” experienced autonomy
in choosing when to come for more food. In addition, such infants were free to
approach adults to request touch, which could not be done by the infants who
were constrained in high chairs.
Ackn owle d g m e nt s
This paper was presented at the Touch Research Symposium in Minneapolis,

Minnesota on April 19, 2001. This research was supported by funds provided by
the Johnson & Johnson Pediatric Institute, L.L.C. to the Touch Research Institutes
and to the author.
Ref ere n c e s
1. Harkness, Super, 1989. Personal communication.
2. Bruner JS (1973). Organization of early skilled action. Child Development, 44, 1-11.
3. Martini M. Differences across cultural groups in mothers’ nonverbal teaching methods.
Paper presented at: Touch Research Symposium; April 19, 2001; Minneapolis, Minn.
SECTION II.
T H E R A P E U T I C A P P L I C AT I O N S
O F T O U C H I N P R E G N A N C Y,
L ABOR AND POSTBIRTH

CHAPTER 6:
P R E G N A N C Y, L A B O R
AND INFANT MASSAGE
Tiffany Field, PhD

Miguel Diego, MA
Abstract
Massage therapy during pregnancy benefits the mother-to-be and the fetus.
Massage during labor also benefits the expectant mother, reducing her labor
pain and shortening her labor. Elderly volunteers who massaged infants
lowered their own stress hormone levels, as well as improved the health of
the newborns. In addition, depressed mothers who massaged their infants
experienced decreased depression. In this chapter studies are reviewed on
massage therapy during pregnancy and labor, as well as infant massage for
full-term and preterm infants, and the benefits experienced by both those
who give and those who receive the massages.
100 Section II. Therapeutic Applications of Touch in Pregnancy, Labor and Postbirth
Massa g e Dur i ng Pre g na nc y
In our study on pregnancy massage, the massage therapy was expected to decrease
stress hormones and anxiety, leg pain and back pain. Improvements in mood and
sleep were also expected. In this study, 26 pregnant women were assigned to a
massage therapy or a relaxation therapy group for 5 weeks.1 The therapies consisted
of 20-minute sessions twice weekly. Both groups reported feeling less anxious after
the first session and less leg pain after the first and last sessions. Only the massage
therapy group, however, reported reduced anxiety, improved mood, better sleep
and less back pain by the last day of the study. In addition, urinary stress hormones
(ie, norepinephrine) decreased for the massage therapy group, these women
had fewer obstetric complications and their infants experienced fewer postnatal
complications, most especially a lower incidence of prematurity.
Prenatal depression has been noted to affect the developing fetus. In one of our
studies, depressed pregnant women who had elevated cortisol and norepinephrine
(stress hormones) gave birth to newborns who had depression-like symptoms2:
These newborns also had higher levels of cortisol and norepinephrine,3 as well
as greater relative right frontal electroencephalogram (EEG) activation, thereby
mimicking their depressed mothers’ stress hormone profile and EEG patterns.4
In an attempt to reduce maternal depression and its negative effects on the fetus
and neonate, we conducted a study on massage therapy with depressed pregnant
women (Field et al, 2003, unpublished data). These depressed women were
randomly assigned to a massage therapy, relaxation therapy or a standard
treatment control group and were compared with each other and with a group of
nondepressed women at the end of pregnancy. The massage and relaxation groups
received two 20-minute therapy sessions each week, given by their significant
others, during the last trimester of their pregnancies. Data analyses revealed the
following (Figure 1):
• Immediately after both types of therapy on the first and last days,
the women reported lower anxiety and reduced leg and back pain.
• Both therapy groups had more optimal postnatal complications scores

than did the depressed control group, although only the massage therapy
group had more optimal obstetric complications scores including a
lower incidence of prematurity.
Chapter 6: Pregnancy, Labor and Infant Massage 101
• The massage therapy group had lower depression scores.
• Fetal movement decreased for all 3 groups.
• The massage therapy group had better Brazelton Neonatal Behavior

Assessment Scale (BNBAS) scores, including those representing
habituation, motor ability, range of state, withdrawal and depression
scores.
Figure 1. In depressed pregnant women, massage therapy during the last trimester
was associated with lower depression scores and decreased fetal movement.T.F., M.D., M.H-R.,
S. Schanberg, C. Kuhn, unpublished data, 2003
(CES-D = Center for Epidemiological Studies of Depression scale)
These data suggest that depressed pregnant women and their offspring can benefit
from these alternative therapies, particularly massage therapy.
Ma s sa g e Du r i ng L a b o r
Touch and massage have been used effectively during labor in nearly every culture
for hundreds of years.5 In contrast, physical support during delivery has been
available only recently to Western women.6 In the past, the massage and support
provided during labor were used to improve or correct the position of the fetus, to
stimulate uterine contractions, to “…prevent the fetus from rising back up in the
abdomen…” and to exert mechanical pressure to aid in the expulsion of the child.7
However, today the focus tends to center more on relaxation to reduce anxiety and
alleviate pain.5
This study compared significant others massaging their pregnant partners during
labor versus partners simply being present and doing what came naturally during
labor (typically, coaching the breathing exercises they had learned in prenatal
classes).8 Massage coupled with breathing exercises, versus breathing exercises alone,
was expected to reduce anxiety and pain, as well as the length of labor. Twenty-eight
women were recruited from prenatal classes and assigned randomly to receive
massage and coaching in breathing from their partners during labor, or to receive
coaching in breathing alone. The mothers who received massage reported decreases
in depressed mood, anxiety and pain, showed less agitated activity and anxiety,
and exhibited more positive affect following the first massage. In addition, the
mothers who were massaged had significantly shorter labors (Figure 2), less labor
medication, shorter hospital stays and less postpartum depression.
Figure 2. Mothers who received massage therapy from their partners during labor,
in addition to coached breathing exercises, had significantly shorter labors than those
who received coached breathing exercises alone.8
Inf a n t Ma ss a ge
Infant massage is practiced in most countries of the world, especially in Africa and
Asia.9,10 In many countries, including Nigeria, Uganda, India, China, Bali, Fiji,
New Guinea, New Zealand (the Maiori), and the Soviet Union, infants are given
a massage with oil after the daily bath and before sleep time for the first several
months of their lives.
Infant massage has been discovered and researched only recently in the Western
world. Suddenly, in the United States, there are massage therapy schools in almost
every city teaching parents how to massage their infants. The techniques they are
learning and using are based on the teachings of 2 massage therapists who trained in
India.9,10 Indian infant massage involves a daily routine that begins during the first
days of life. The infant is laid on his or her stomach on the mother’s outstretched
legs, and each body part is stretched individually. Warm water and soap are applied
to the legs, arms, back, abdomen, neck and face. The massager looks like she is
scrubbing clothes on an old washboard and the process seems extremely rigorous.
After they are massaged and swaddled, the infants then sleep for prolonged periods.
Although data have not been collected on infant massage as it is practiced in India,
some infant massage therapists have attributed the precocious motor development
of these infants to their daily massages. Infant massage therapists have made
several claims based on anecdotal data, including that the massage provides
both stimulation and relaxation that helps respiration, circulation, digestion and
elimination.11 They have claimed that infants who are massaged sleep more soundly,
that the massage relieves gas and colic, and that it helps the healing process during
illness by easing congestion and pain.12 Furthermore, they assert that infant massage
helps enhance parent-infant bonding and warm, positive relationships, reduces
distress in the infant following painful procedures (such as inoculations), reduces
pain from teething and constipation, reduces sleep problems and makes parents
“feel good” while they are massaging their infants. Infant massage therapy groups
have also reported that infants who have special needs—such as those who are
blind and deaf or those who are paralyzed, have cerebral palsy or are premature—
seem to become more aware of their bodies, among receiving other benefits.
Figure 3. Premature babies do not like light touch, probably because it feels like
tickling. Babies who gained weight in the studies of infant massage were those who
received deeper-pressure massage that stimulated both tactile and pressure receptors.13
Figure 4. Massaged infants gained 47% more weight per day than control infants.14
Pre t e r m In f a nt Ma s s a ge
Most data available on the effects of infant massage come from studies of preterm/
premature infants. During the last 2 decades, several studies were conducted on
what we called “tactile/kinesthetic stimulation” (due to the negative connotations
attached to the word “massage”). A meta-analysis of data from 19 of these studies
revealed that 72% of the massaged infants were affected positively.13 Most of
them experienced greater weight gain and better performance on developmental
assessments. In those studies that did not report significant weight gain, investigators
had used a light-stroking procedure. Babies do not like light touch (Figure 3),
probably because it feels like tickling. The babies who gained weight had been given
deeper-pressure massage, thereby stimulating both tactile and pressure receptors.
One of the studies used in this global analysis was conducted in our laboratory
at the Touch Research Institute in Miami.14 In that study, massage therapy was
given to preterm newborns 3 times per day, for 15 minutes per session, for 10 days.
As shown in Figure 4, the massaged infants averaged 47% greater weight gain
than infants who were not massaged, even though the groups consumed equal
amounts of formula. The massaged infants were awake and active more of the time,
even though we expected they would sleep more. They were also more alert and
responsive to the examiner’s face and voice, and they showed more organized limb
movements on the BNBAS. Finally, they were discharged from the hospital 6 days
sooner, saving approximately $3000 per infant in hospital costs. The comparable
cost savings today in the US would be $10,000 per infant. If every one of the
470,000 premature infants born each year in the US was massaged, the hospital
cost savings would approximate $4.7 billion per year. That dollar figure could
double based on more recent data, suggesting that the same weight gain can be
achieved in 5 days, versus 10 days, of massage therapy.15
Replication studies have been conducted in Israel16 and the Philippines.17 In the
study performed in the Philippines,17 which was done with an exact replication
of the Field et al methodology,18 preterm infants who were massaged gained 45%
more than infants who were not massaged. In the study conducted in Israel,16 a
31% greater weight gain and more organized sleep were reported for the massaged,
versus the control, preterm infants. In addition, the mothers who provided the
massage experienced less postpartum depression. These studies approximated
the weight gain data (47% and 31%, respectively) published by Field et al14 and
Scafidi et al.19 A recent study by Dieter et al15 suggests that a 46% greater weight
gain can be achieved in preterm infants following only 5 days of massage.
At approximately the same time that the previously mentioned premature infant
studies were being done,14 colleagues at Duke University Medical School were
conducting similar studies on rat pups.20 They separated rat pups from their
mothers to explore touch deprivation. The researchers stroked the deprived rat pups
with a paintbrush—much like the mother rat would tongue-lick them—so that the
pups would grow normally. In several studies, the team at Duke noted that growth
hormone decreased when the pups were separated from their mothers. This
decrease was noted in all body organs, including the heart, liver and brain. These
values, however, returned to normal once the pups were stroked with the paintbrush.
The more recent discovery of a growth gene that responds to touch suggests a
strong genetic influence on the relationship between touching and growth,
although the underlying mechanism is unclear at this time.21
This observation, plus the results of a study done in Sweden,22 led myself and
my colleagues to some theories about mechanisms that might explain the
touch/weight-gain relationship. The investigators in Sweden reported that
stimulating the mouths of the newborn, as well as the breasts of the breastfeeding
mothers, led to an increase in food absorption hormones, such as gastrin and
insulin.22 Schanberg and I argued that massage therapy delivered to several parts
of the body would lead to an even greater increase in food absorption hormones,
which could itself explain the weight gain.23 Assays on the insulin levels before,
during and after massage therapy suggested that insulin levels were elevated in
those preterm infants who received massage therapy versus those who did not
(a 61% increase versus a 4% decrease, respectively, in insulin levels).23 Of course,
future studies could assess the relative changes caused by several food absorption
hormones and other vagal stimulated changes, such as an increase in gastric
motility.
Mass a g i n g In f a nt s Wi t h Oi l
To determine whether massaging infants with oil enhanced the effects of massage, a
study was conducted in which 60, 1-month-old, full-term infants were randomized
to one group that was massaged with oil, while the other group was massaged
without oil.24 Results of this trial showed that the infants massaged with oil
experienced greater soothing and calming effects. The infants who received massage
with oil were less active, showed fewer stress behaviors, demonstrated less head
averting and had greater declines in saliva cortisol levels. In addition, vagal activity
increased more following massage with oil versus massage without oil.
Ma s sa g i n g Inf a nt s W h o Ha ve Colic and Sle e p

Pro b le ms
“Colic” and sleep problems are the complaints presented most frequently to
pediatricians by parents of infants. In a study on these problems, we taught parents,
whose infants attended a university nursery school, to massage their 3-month-old
to 6-month-old infants for 15-minute periods prior to bedtime.18 The massaged
infants, versus a group of control infants, became less irritable, fell asleep faster,
experienced fewer night awakenings and spent more time in quiet, alert states
during the daytime.
Parents who participated in this study, and who years later encountered the
investigators on campus, suggested that their marriages had been “saved” by
having received help to calm their infants and get them to sleep. Others expressed
chagrin that their children, now as old as 7 years, still needed a massage in order
to go to sleep.
De pre s se d Mo t h e r s Ma s s a gi ng T h e ir Infants
In our studies, we routinely teach parents to massage their infants so they can be
massaged on a daily basis at no cost. In addition, these parents are told that they,
too, might benefit from giving the massages. A study in which depressed mothers
were taught to massage their infants resulted in decreased maternal depression and
less distress behavior and disturbed sleep patterns in the infants.18 For this study,
the infants’ mothers were asked to perform 15-minute massages once daily for
2 weeks. At the end of the study period, the massaged infants fell asleep faster, slept
longer and were less fussy. The mothers enjoyed better face-to-face interactions
with their infants. During these interactions,

Figure 5. the infants were positioned in an infant
Mothers seat on a table facing their mothers. Video
massaging cameras, which were partially hidden, filmed
the mothers’ faces and torsos and the infants’
babies.
faces and bodies. The videotapes were
subsequently coded for the mothers’ and
infants’ eye contact, facial expressions
and vocalizations. Based on their filmed
behaviors, improved mother-infant
interactions were documented.
Fath e r s a s Ma s s a g e T h e ra p i s t s
A recent Australian television show featured American fathers giving massages to

their infants. Since this film was broadcast, a study was conducted on Australian
fathers massaging their infants.25 In this study, fathers who had first-born babies
were given a 1-month training program in a bathing-massage technique. During
home observations after 3 months, the fathers showed greater involvement with
their infants. At this time, the massaged infants also greeted their fathers with more
eye contact, smiling, vocalizing and reaching responses and less avoidance behavior.
In a study we conducted, infants were given massages by their fathers for 15 minutes
prior to their daily bedtimes over a 1-month period.26 By the end of the study,
the fathers who massaged their infants were more expressive and showed more
enjoyment and warmth during floor-play interactions with their infants than did
a control group.
El derly Vo lun te e r s a s Ma s s a ge T h e rap ists
Teaching infant massage to elderly volunteers is a no-cost way to deliver massage

therapy. In a study we conducted, elderly volunteers massaged infants and toddlers
in a nursery school environment.27 The study was designed to measure the effects
of massage therapy on the elderly volunteers giving the massage. Surveys indicate
that failure-to-thrive and depression are fairly common among the elderly, with a
frequency of 5% to 26%.28 Symptoms of depression in the elderly are similar to
those found in younger persons.29 In addition, the elderly can experience frequent
nighttime awakenings, increased levels of stress hormones and immune system
problems.30 Pet therapy (ie, having and holding pets) has been effective with
the elderly,31 and in our study massage therapy was also effective. In this study the
elderly volunteers were randomly assigned to either give infants massages or to
receive massage therapy themselves (Figure 6).27 At the end of the first month,
the volunteers then received the opposite treatment. Each of these sessions was
Figure 6. In the
grandparent-aged
volunteer study,
participants gave
massages to infants
and received massage
therapy themselves.27
Figure 7. Levels of stress hormones decreased in volunteer grandparents after giving

infant massages.27
15 minutes in length for the infant massages and 30 minutes for the grandparent-
volunteer massages. The latter sessions were longer simply because a full-body
table massage for an adult requires more time. Both the infant and the elderly
volunteer massages occurred twice per week for 4 weeks. Following the baseline
and end-of-study sessions, the grandparent volunteers reported lower anxiety levels,
fewer symptoms of depression and improved mood after both giving and receiving
massages. Their stress hormones also decreased after giving massages (Figure 7).
After only 1 month of giving or receiving massages, their scores on a lifestlye
questionnaire improved. They had more social contacts, made fewer trips to their
doctors’ offices and consumed less coffee. These changes probably helped improve
their sleep and their self-esteem. A somewhat surprising finding was that these
improvements were greater after 1 month of giving the infant massages than after
1 month of receiving massages. Thus, massage therapy was effective not only for
the infants receiving them but also for the adults who were giving these massages.
Su mm a r y
In summary, massage therapy appears to facilitate growth and development from

as early as the fetal stage. Massage therapy given during pregnancy contributes to
lower prematurity rates, probably via the mechanism of reduced cortisol levels (a
stress hormone) in the mother, leading to lower cortisol levels in the fetus. Massage
therapy given during labor lessens labor pain, shortens labor and thereby reduces
the stress on the mother and fetus. During the neonatal stage, infants benefit from
reduced irritability and more organized sleep, if they are born at term, or enhanced
growth, if they are born prematurely and/or experience exposure to cocaine or the
human immunodeficiency virus (HIV). Using elderly volunteers, mothers or fathers
to deliver the massage is not only cost-effective, but it also benefits the person
giving the massage by lowering stress hormones and likely improving health, given
that cortisol kills immune cells. These data highlight the importance of adding
massage therapy training to birthing classes and in newborn nurseries and neonatal
intensive care units (NICUs). According to a recent survey, 38% of NICUs across
the US offer massage therapy in their programs (Field et al 2003, unpublished
data). Like most other programs, however, only consumer demand will make
massage therapy a standard.
The authors wish to thank the mothers, fathers, elderly volunteers and infants who
participated in these studies, as well as the research associates who assisted in the
data collecting and coding. This research was supported by a National Institute of
Mental Health (NIMH) Research Scientist Award (#MH00331) and an NIMH
Research Grant (#MH46586) given to Tiffany Field, as well as by funding
provided by the Johnson & Johnson Pediatric Institute, L.L.C. to the Touch
Research Institutes.
Ref ere n c e s
1. Field T, Hernandez-Reif M, Hart S, Theakston H, Schanberg S, Kuhn C. Pregnant women
benefit from massage therapy. Journal of Psychosomatic Obstetrics and Gynaecology. 1999;20:
31-38.

3. Lundy BL, Jones NA, Field T, et al. Prenatal depression effects on neonates. Infant Behavior
and Development. 1999;22:119-129.
4. Jones NA, Field T, Fox NA, Lundy B, Davalos M. EEG activation in 1-month-old infants
of depressed mothers. Development and Psychopathology. 1997;9:491-505.
5. Hedstrom LW, Newton N. Touch in labor: a comparison of cultures and eras. Birth.
1986;13:181-186.
6. Kennell J, Klaus M, McGrath S, Robertson S, Hinkley C. Continuous emotional support

during labor in a US hospital. A randomized controlled trial. JAMA: the Journal of the
American Medical Association. 1991;265:2197-2201.
7. Engelman G. Labour Among Primitive Peoples. St Louis, Mo: JH Chambers; 1982.
8. Field T, Hernandez-Reif M, Taylor S, Quintino O, Burman I. Labor pain is reduced by

massage therapy. Journal of Psychosomatic Obstetrics and Gynaecology. 1997;18:286-291.
9. Auckett AD. Baby Massage: Parent-Child Bonding Through Touching. New York, NY:
Newmarket Press; 1982.
10. McClure VS. Infant Massage: A Handbook for Loving Parents. New York, NY: Bantam Books;
1989.
11. Grossman RL. Other Medicines. Garden City, NJ: Doubleday; 1985.
12. Eisenberg D, Wright TL. Encounters With Qi: Exploring Chinese Medicine. New York, NY:
WW Norton & Company; 1995.
13. Ottenbacher KJ, Muller L, Brandt D, Heintzelman A, Hojem P, Sharpe P. The effectiveness
of tactile stimulation as a form of early intervention: a quantitative evaluation. Journal of
Developmental and Behavioral Pediatrics: JDBP. 1987;8:68-76.
15. Dieter J, Field T, Hernandez-Reif M, Emory E, Redzepi M. Stable preterm infants gain more
weight and sleep less after 5 days of massage therapy. Journal of Pediatric Psychology.
2002;28:403-411.
16. Goldstein-Ferber S. Massage in premature infants. Paper presented at: Child Development
Conference; 1998; Bar-Elon, Israel.
17. Jinon S. The effect of infant massage on growth of the preterm infant. In: Yarbes-Almirante
C, De Luma M, eds. Increasing Safe and Successful Pregnancy. The Netherlands: Elsevier
Science, BZ; 1996:265-269.
18. Field T, Grizzle N, Scafidi F, Abrams S, Richardson S. Massage therapy for infants of
depressed mothers. Infant Behavior and Development. 1996;19:107-112.
19. Scafidi F, Field T, Schanberg S, et al. Massage stimulates growth in preterm infants: a
replication. Infant Behavior and Development. 1990;13:167-188.
20. Schanberg S, Field T. Maternal deprivation and supplemental stimulation. In: Field T,
McCabe PM, Schneiderman N, eds. Stress and Coping: Across Development. Hillsdale, NJ:
Lawrence Erlbaum Associates; 1988:3-25.
21. Schanberg S. The genetic basis for touch effects. In: Field TM, ed. Touch in Early
22. Uvnäs-Moberg K, Widstrom AM, Marchini G, Winberg J. Release of GI hormones in

mother and infant by sensory stimulation. Acta Paediatrica Scandinavica. 1987;76:851-860.
23. Field T, Schanberg S. Massage alters growth and catecholamine production in preterm
newborns. In: Field T, Brazelton TB, eds. Advances in Touch. Skillman, NJ: Johnson &
Johnson; 1990:96-104.
24. Field T, Schanberg S, Davalos M, Malphurs J. Massage with oil has more positive effects
on normal infants. Journal of Prenatal and Perinatal Psychology and Health. 1996;11:75-80.
25. Scholz K, Samuels CA. Neonatal bathing and massage intervention with fathers, behavioral
effects 12 weeks after birth of the first baby: the Sunraysia Australia Intervention Project.
International Journal of Behavioral Development. 1992;15:67-81.
26. Cullen C, Field T, Escalona A, Hartshorn K. Father-infant interactions are enhanced by

massage therapy. Early Child Development and Care. 2000;164:41-47.
27. Field T, Hernandez-Reif M, Quintino O, Schanberg S, Kuhn C. Elder retired volunteers
benefit from giving massage therapy to infants. Journal of Applied Gerontology. 1998;17:
229-239.
28. Copeland JRM, Dewey ME, Wood N, Searle R, Davidson IA, McWilliam C. Range of
mental illness among the elderly in the community. Prevalence in Liverpool using the
GMS-AGECAT package. The British Journal of Psychiatry: the Journal of Mental Science.
1987;150:815-823.
29. Gaylord SA, Zung WWK. Affective disorders among the aging. In: Carstensen LL, Edelstein
BA, eds. Handbook of Clinical Gerontology. New York, NY: Allyn & Bacon; 1987:139-151.
30. Post F. Functional disorder II. Treatment and its relationship to causation. In: Levy R, ed.
Psychiatry of Late Life. London, England: Blackwell Scientific; 1982:213-229.
31. Grossberg JM, Alf EF Jr. Interaction with pet dogs: effects on human cardiovascular response.
The Journal of the Delta Society. 1985;32:518-524.
CHAPTER 7:
SKIN-TO-SKIN CARE FOR
B R E A S T F E E D I N G D I F F I C U LT I E S
POSTBIRTH
Gene Cranston Anderson, PhD, FAAN1

Sheau-Huey Chiu, PhD, RN, PNP2
Barbara Morrison, PhD, FNP, CNM1
Maria Burkhammer, RN, CDONA1
Susan Ludington-Hoe, CNM, PhD, FAAN1
1 Frances Payne Bolton School of Nursing
Case Western Reserve University
2 College of Nursing
University of Akron
Abstract
This chapter reports the results of a study evaluating the status of breastfeeding (BF)
for mother-infant dyads having BF difficulties postbirth and given skin-to-skin care
(SSC). The hypothesis was that SSC would facilitate successful BF. The focus was
on exclusive BF (only human milk) as well as on duration, because both measures
correlate positively with health benefits. The study involved 50 healthy, fullterm,
mother-newborn dyads experiencing BF difficulties between 11 hours and 24 hours
postbirth. Participants experienced “SSC with BF” (SB) for 3 consecutive BFs
(SB1–SB3) and one BF (SB4) 24 hours after SB1. Two dyads withdrew during the
hospital phase. At SB4, of the 48 remaining dyads, 39 (81.3%) were BF exclusively,
while 9 (18.7%) were BF partially. At 1-week follow-up, 35 dyads (72.9%) were BF
exclusively, 5 (10.4%) were BF partially, 6 (12.5%) were not BF and 2 (4.2%) were
lost to follow-up. At 1-month follow-up, 25 dyads (52.1%) were BF exclusively,
9 (18.8%) were BF partially, 13 (27.1%) were not BF and 1 (2.1%) was lost to
follow-up. These mother-infant dyads were a select group because they were already
experiencing BF difficulties. Thus, the fact that 81.3% of these dyads breastfed
exclusively during the early postpartum period is impressive, because this compares
favorably with 71.9% based on data from the Ross Mothers’ Survey,1 and the 75%
designated as Objective 16–19 for Healthy People 2010.2 Important insights were
also gained about ways to strengthen the intervention, including allowing infants
to have uninterrupted sleep and awaken spontaneously, placing infants in SSC at
first oral cues and protecting the BF process from interruptions.
Introd uc t i o n
In the United States (US), women have become increasingly motivated to breastfeed
their newborn infants. This motivation emanates from increasing evidence-based
information about the benefits of breastfeeding (BF) for their infants and even
themselves,3-7 as well as the recommendation that BF be exclusive (using human
milk only) for the first 6 months of the infant’s life, with the subsequent
introduction of complementary foods and continued BF thereafter through at
least the first year.8,9 Although the incidence of BF has increased gradually since
1984, exclusive BF, which yields the most benefits, decreased from 87.9% in 1971
to 71.9% in 1998.1
In addition to the health benefits of BF, sources of motivation for health-care

providers and hospital systems to encourage BF include:
• the seminal workshop on BF and human lactation sponsored by former

US Surgeon General C. Everett Koop, MD10
• the Innocenti Declaration11
• the inclusion of objectives for BF in Healthy People 200012 and

Healthy People 20102
• the Baby-Friendly Hospital Initiative13-15
• the Blueprint for Action on Breastfeeding (2000) set forth by former

US Surgeon General David Satcher7
• the strategic plan for implementing the Blueprint16

Chapter 7: Skin-to-Skin Care for Breastfeeding Difficulties Postbirth 117
However, the intention and motivation to breastfeed do not, of themselves,

guarantee success. In fact, mothers frequently fail to successfully initiate and
continue BF, often through no fault of their own. Some healthy infants, also
through no fault of their own, do not attach to the breast successfully during the
first day—a problem often leading to BF failures. Other infants attach but in a
less-than-optimal fashion: They attach to the nipple, but not to the areola (the
area beyond the nipple). Attaching this way, such infants cannot compress the
milk-containing sinuses in the breast with their gums, and thus they receive very
little milk. Instead, their gums compress the mother’s nipple, causing redness,
blisters, fissures, pain and, of course, maternal concern and stress.17,18 Maternal
stress affects the synthesis, ejection and release of milk negatively.19 Infants naturally
become hungry, fussy and disorganized, and they, too, seem stressed. Between
periods of sleep from exhaustion, they cry intermittently, which is unhealthy for
newborn infants.20,21
Stress occurring during the early postpartum period may present serious consequences
for newborn infants. In a randomized, controlled trial of 84 fullterm infants, salivary
cortisol reached inappropriate levels at 6 hours postbirth in the control infants who
were separated from their mothers and received standard procedural care in the
hospital nursery for hours 1 through 5.22 In contrast, the infants who remained
with their mothers, as compared with the control infants, exhibited significantly
lower levels of cortisol at 6 hours, indicating a healthy recovery from the high levels
of cortisol present at birth. Conceivably the inappropriate responses exhibited by
the control infants might result in persistent similar responses to stress across the
life span. This has been documented in animal models, wherein newborn rat pups
that had the least maternal contact (licking, grooming, and arched-back nursing
posture) suffered inappropriate stress responsivity. These inappropriate responses
to stress can lead to cognitive dysfunction,25,27 stress-related diseases23,24,26,27 and
susceptibility to addiction.28
Skin-to-skin
For dyads who have difficulties with BF, contact may
skin-to-skin contact (SSC)—otherwise be a useful,
known as “kangaroo care”—may be a useful, stress-reducing
stress-reducing intervention. With SSC, the intervention
mother holds her diaper-clad infant skin to for dyads
skin and underneath her clothing, using a who have
breastfeeding
blanket draped across her infant’s back for
difficulties.
warmth, if needed. SSC is recognized as an
important kind of touch by leaders in

With skin-to-skin
this area of scientific investigation.29-32
contact, the
mother holds her
Studies of SSC also support its safety
diaper-clad infant and beneficial effects in the care of
underneath her preterm infants and their mothers:
clothing, with Examples include exclusive BF and a loss
a blanket across of only 6.3% of birth weight,33 increased
the infant’s back,
milk volume,34 and, at follow-up, a
if needed, for
longer BF duration.33,35 SSC is now used
warmth.
in many neonatal intensive care units
(NICUs), and it has been used as an
intervention to promote BF. Only 2 small anecdotal studies,37,38 however, have
36
been reported in which SSC was used as an intervention for mothers and their
fullterm infants who were identified as having difficulties with BF.
Harris reported that mothers and their fullterm infants who were having difficulties
with BF during the first few days postbirth often had a history of interruption of
peaceful mother-baby interactions during the immediate perinatal period (the first
2 hours postbirth).37 Harris then described how to treat these difficulties that the
mother-infant dyad were having with BF: She reported that if the mother and baby
are placed together, skin to skin, in a deep bathtub filled with water no warmer
than 38ºC (100.4ºF), even as late as 1 week postbirth, almost all infants will
move gradually to the breast, self-attach correctly with minimal assistance within
30 minutes and breastfeed successfully thereafter.
In the second study, a small pilot trial, 5 mothers and their fullterm infants who had
difficulties with BF were given SSC for 30 to 60 minutes.38 As a result, each infant
in 4 of these dyads not only located their mother’s breast during SSC and attached
independently, but they also attached correctly. The fifth dyad was not successful,
perhaps because at least 10 extended-family members were chatting loudly in the
room during this time. The first 3 dyads, who had not yet breastfed successfully by
18, 20 and 40 hours postbirth, were reported by Meyer and Anderson.38 However,
none of the 5 dyads was followed after their initial, successful BF.
Providing extended SSC postbirth in the hospital results in significantly better BF

outcomes for mother and fullterm infant dyads.39 In fact, when healthy fullterm
infants (if unmedicated) were placed on their mothers’ abdomens immediately after
birth and were allowed to remain there (ie, with SSC), they reached the breast, they
attached unaided and they began to nurse successfully within 30 to 60 minutes
postbirth.31,40 Importantly, one third of 70 fullterm infants born at home and given
unrestricted BFs and essentially continuous SSC from birth lost no weight.41 Even
when mothers and preterm infants, 34 to 36 weeks gestation, were separated during
the first 30 to 40 minutes postbirth and were then reunited for almost continuous
SSC thereafter, the infants achieved successful BF during the first few hours with
minimal assistance.33
Despite this compelling evidence, implementing SSC so soon after birth is

difficult, even in certified Baby-Friendly Hospitals because of increased budget
constraints.15,42-44 Unfortunately, when BF difficulties become evident, they are
often refractory to a variety of interventions. For these reasons, we conducted the
exploratory research reported here, using SSC during BFs (SBs) to answer the
following questions:
• In what percentage of these dyads will SBs be successful?
• What percentage of birth weight is lost at hospital discharge?
• What patterns of BF success will occur during the 3 SBs?
• How often will each mother breastfeed between the end of SB3
and the beginning of SB4?
• At study entry, how much BF-related pain will mothers have already
experienced?
• Following study entry, how much BF-related pain will mothers

experience during each SB?
• How exclusive will BF be at 1 week and at 1 month postdischarge?
• What will be the mothers’ evaluations of their BF experiences with

their infants?
• Do changes occur in the levels of maternal salivary cortisol after

experiencing SBs?
Metho d s
SAMPLE
Fifty culturally diverse mothers—in the postpartum unit of a large teaching

hospital in the upper-midwest US—who had chosen to breastfeed, but who were
experiencing BF difficulties, were enrolled following their informed consent. Two
mothers withdrew from the study soon after enrollment: One mother withdrew—
before the first SB—because the father did not want her to participate. He said
“we have had too many interruptions already.” The second mother withdrew—soon
after the first SB—because her pediatrician told her that her infant would never be
able to nurse due to constricted nasal passages. Thus, data from 48 mothers and
their healthy infants are reported here (Table 1). Only 6 of the 13 multiparous
mothers had breastfed a previous child.
INSTRUMENTS
Demographic and obstetrical data were collected from the medical records of the
mothers and their infants regarding the following: prenatal maternal behavior;
previous experience with BF; the birth experience; maternal and infant
demographics and medical condition; and any previous formula feedings and/or
experiences with SSC with the current infant.
Six instruments were used to measure difficulty and success of BF. The Mother-
Baby Assessment (MBA), a 10-point measuring tool of maternal and infant BF
behavior,45 was used to observe the progress of the mother-infant dyads as they
learned to breastfeed. The MBA has five sequential assessment steps: signaling,
positioning, fixing, milk transfer and ending. For this study, “difficulty” was
defined as a score of 7 or less. Three Visual Analogue Scales (VASs) were used in
the hospital for maternal self-reports of nipple pain, breast pain and BF success.
VASs are single-item indicators that ask respondents for their global ratings of
specific concepts based on factors salient to them.46 After the mothers went home
and because we were interviewing them by telephone, the VASs were converted to
Likert scoring systems. These interviews were done at 1 week and at 4 weeks. The
Index of Breastfeeding Scale (IBS) is a measure of duration but, more importantly
exclusivity, and was developed in 1988 by an interagency group of representatives
from organizations such as the World Health Organization (WHO) and the United
Table 1. Maternal and infant demographic characteristics (n=48).
Variable Mean SD(±) Range

Age (year) 29.0 5.9 18–41
Education (year) 14.9 2.1 10–18
Gestation (week) 39.0 0.95 37–41
Birth weight (gm) 3396 455 2490–4415
Discharge weight (gm) 3171 412 2365–4050
Length of stay (hour) 56.1 29.2* 36.5–230
VARIABLE N %†
Maternal race
African American 16 33.3
White 24 50.0
Asian/Pacific Islander 6 12.5
Other 2 4.2
Marital status
Married 32 66.7
Single 16 33.3
Employment status
Full time 29 60.4
Part time 7 14.6
Unemployed 8 16.7
Student 4 8.3
Delivery method
Vaginal 35 72.9
Cesarean section 10 20.8
Vacuum/forceps 3 6.3
Analgesia/anesthesia used
None 4 8.3
Pain medication 2 4.2
Epidural 30 62.5
Spinal 1 2.1
Pain medication and epidural 11 22.9
Smoking‡
Yes 2 4.2
No 46 95.8
Parity
Primipara 35 72.9
Multipara 13 27.1
Infant gender
Male 24 50.0
Female 24 50.0
* SD is large because one infant was admitted to NICU due to hyperbilirubinemia.

† Column percent may not equal 100 due to rounding.
‡ During this pregnancy.
Nations Children’s Fund (UNICEF). These representatives were known for their
expertise in the area of BF measurement. This group referred to the IBS as a
“schema” for definitions of BF.47 We have since added the acronym and now
refer to the schema as “The IBS” in order to facilitate its clinical usefulness.
“Exclusivity” is defined as BF with no complementation and no supplementation.48
Complementation means BF plus formula.48 Supplementation means formula not
replacing a BF.48 The basic IBS is equivalent to the schema and has 6 categories:
exclusive; almost exclusive; high, medium, and low partial; and token.
For purposes of data collection in this study, the IBS was adapted to form 8
categories by subdividing the medium partial category into medium high and
medium low and adding a category called “none.” Therefore, the 8 newly formed
categories were called exclusive, almost exclusive, high partial, medium high partial,
medium low partial, low partial, token and none. BF exclusivity was coded on
a scale of 1 through 8, where “1 = exclusive” and “8 = none.” We also added
percentages to each category. During analyses for this report, these 8 categories
were collapsed into 4 categories: exclusive (including exclusive and almost exclusive)
100%, partial (including high, medium high, medium low, and low) (<100%
to >5%), token (5% to >0%), and none (0%). The almost exclusive category
essentially did not occur in this study.
The fourth instrument was the LATCH, a BF evaluation tool designed for
professional assessment based on either observations or questions to solicit responses
by the mothers.49 The LATCH evaluates the following 5 components of BF: how
well the infant latches correctly onto the breast; amount of audible swallowing
heard at the breast; type of nipple; comfort level of the mother regarding breasts,
nipples and position; and the amount of help required to position the infant.
Each of the 5 components is scored numerically as a 0, 1, or 2; total possible
points = 10. The higher the score, the greater the success and satisfaction with BF,
and the more likely the mother will continue with BF.50
The fifth instrument, the Maternal Breastfeeding Evaluation Scale (MBFES),

is designed to measure both positive and negative aspects of BF that mothers
have identified as important in defining successful BF. This tool consists of 3
subscales: maternal enjoyment/role attainment, infant satisfaction/growth and
lifestyle/maternal body image.51 Total possible score is 150 with higher scores
indicating greater satisfaction with BF. In addition, we recorded notes regarding
potentially confounding events that occurred prior to enrollment (eg, maternal-

infant separation and formula feedings) and over the course of the 2-day protocol
(eg, maternal distress, procedures such as circumcision or use of a pacifier and the
reason for its use).
PROCEDURE
Eligible dyads were enrolled in the study whenever research staffing appeared
adequate to cover all data-collection points. The original age for eligibility was
approximately 12 hours postbirth. However, because many of the enrolled infants
seemed very sleepy at 12 hours, eligibility was expanded after dyad 29 to include
infants who were between 12 and 24 hours, a time during which the effects of
obstetric medications may have diminished. Recruitment was accomplished in the
following manner. The researcher screened dyads for eligibility using the nurse
assignment sheet, on which BF difficulties were noted and in collaboration with the
nurses. Then, after selecting the mother whose infant’s age was closest to the target
postnatal age, the researcher approached the mother to explain the study, request
her participation and obtain informed consent. If this mother did not consent, the
same recruitment procedure was followed for the mother whose infant was next
closest to the target age.
Following enrollment on day 1 and approximately 30 minutes before the next BF

was expected, the first of 3 consecutive SBs (SB1 to SB3) took place. The fourth
SB (SB4) occurred on day 2, approximately 24 hours after SB1 to control for the
circadian rhythm of cortisol. Also at this time, the progress of BFs between SB3
and SB4 was evaluated. SB4 generally occurred at a time approaching hospital
discharge for the mothers who delivered vaginally and had 2-day stays. At each SB,
the diaper-clad infant was placed in SSC on the mother’s chest and covered across
his/her back with the mother’s clothing and a blanket. Quiet, privacy and freedom
from interruption were provided as best possible to facilitate maternal relaxation,
an apparently important factor in achieving the benefits of SSC for infants.52
Maternal saliva was collected first, before and after SB1 and SB4. Data were
collected for the VASs immediately before SB1 and after each SB. The MBA and
the LATCH were completed immediately after each SB. The IBS was used to
measure level of exclusivity before SB1 to learn what had occurred since birth,
immediately after each SB to measure each feeding and prior to SB2, SB3 and SB4
to measure relevant events that occurred during the interfeeding intervals (SB1 to
SB2 and SB2 to SB3) and during the longer period of time between SB3 and SB4.
If successful BF did not begin after 30 minutes of SSC, a nurse researcher provided
assistance. The end of each SB was defined in one of 3 ways: at the conclusion
of a successful BF; at the conclusion of a successful BF after assistance was given;
or at the conclusion of an unsuccessful BF despite assistance. In the latter case,
whenever possible, another SB opportunity was provided within 2 hours of the
end of the previous feeding. To avoid disturbing the infant and causing crying,
pre- and postfeeding weights were not done.
The MBFES was administered after SB4. In addition, the most recently recorded
infant weight was obtained. At 1 week and 1 month postnatal ages, mothers were
contacted by telephone to evaluate the following: BF exclusivity, using the IBS; BF
satisfaction, using the MBFES; and nipple pain, breast pain and BF success, using
the Likert scoring system. Data were analyzed using descriptive statistics. The
research protocol for each dyad lasted approximately 1 month. The data reported
here were collected from February 2002 to December 2002.
Result s
At SB4, 48 dyads remained in the study. Of these 48 dyads, 39 (81%) were BF

exclusively, while 9 (19%) were BF partially. At 1 week, 2 of the 48 dyads were
lost to follow-up, 35 were BF exclusively, 5 were BF partially, and 6 were being
fed formula. At 1 month, we re-established contact with one of the dyads lost at
1 week. Thus, only 1 dyad was lost to follow-up, 25 were BF exclusively, 9 were
BF partially, and 13 were being fed formula (Figure 1).
Using the last weight measurement before discharge, the average percent of infant
weight loss was 6% (SD = ± 3%; range = –2% to 21%). Three of these infants lost
more than 10% of body weight (11%, 14% and 21%). The infant who lost 21%
was feeding poorly because of “tongue thrust” and did not begin regaining weight
until after formula feeding by bottle was introduced. No obvious reasons for the
weight loss were identified in the other 2 infants. Forty-one mothers (85%) reported
a successful BF after SB1, 40 mothers (83%) reported the same after SB2 and 42
Figure 1. Percentages of dyads who were exclusively BF, partially BF, formula fed and
lost to follow-up at discharge/B4 at 1 week and at 1 month follow-up (n = 48).
SB = skin-to-skin care with breastfeeding; F/U = follow-up; BF = breastfeeding
mothers (88%) reported a successful BF after SB3. Between the end of SB3 and
the start of SB4, 27 mothers (56%) breastfed their infants an average of 8.5 times
(SD = ± 4.42; range = 3 to 22).
Based on reports made by the mothers, breast pain was generally low, rose at week
1 and had begun to subside at month 1 (Figure 2). Nipple pain was only slightly
higher than breast pain, and had an essentially parallel curve. Mothers’ perceptions
of BF success increased rapidly with each SB experience and then decreased
slightly at week 1 and at month 1. Satisfaction with BF increased over time for the
28 mothers who completed the MBFES at SB4, week 1 and month 1 (120 ± 9.1,
121 ± 14.8 and 122 ± 14.6, respectively).
Saliva samples were collected before and after SB1 and SB4 from the last 20 dyads
enrolled. However, samples from only 5 mothers have been assayed to date. In
these 5 mothers, mean cortisol decreased from pre-SB to post-SB during each SB1
and each SB4, and was the lowest after SB4. Overall, pre-to-post change scores at
SB4 were greater than at SB1 (Table 2).
Figure 2. Mothers’ reports on visual analogue scales for BF success, nipple pain and
breast pain before SB1, after each SB (1–4) and at 1-week and 1-month follow-up
(n=24). Scores represent complete data sets at all 7 times.
SB = skin-to-skin care with breastfeeding; BF = breastfeeding
Table 2. Maternal salivary cortisol levels(µg/dL) before and after SB1 and SB4 (n=48).
Mothers
1 2 3 4 5 Cortisol
SB H C H C H C H C H C Mean (SD±)
Pre-1 0920 .26 1000 .61 1100 .53 0900 .48 1010 .43 .46(.13)
Post-1 1100 .19 1130 .60 1213 .34 0935 .32 1420 .09 .31(.19)
Pre-4 0920 .20 0955 .56 1015 .82 1120 .75 1000 .44 .55(.25)
Post-4 1000 .15 1245 .26 1050 .63 1330 .22 1130 .19 .29(.19)
SB = skin-to-skin breastfeeding; H = military time; C = cortisol; SD = standard deviation.

Discussion
These mother-infant dyads were a select group who were chosen for study because
they had been identified approximately 12 hours postbirth as being at greater risk
for BF failure than most. Nevertheless, 81% (n = 39) of these dyads breastfed
exclusively during the early postpartum period. This percentage compares favorably
with the 75% for any BF (Objective 16–19 for Healthy People 2010)2 and the
71.9%, also for any BF, for 1998 as calculated by Cadwell,1 using data from the
Ross Mothers’ Survey.53
Encouraging results were also observed on other measures: For example, 94%
(n = 45) of the infants lost ≤10% of their birth weight before hospital discharge.
SSC was successful for at least one of the first 3 SBs for 42 (88%) of the mothers.
On average, mothers breastfed 8.5 times between SB3 and SB4 (a period of
approximately 14 hours). This frequency of BF falls within the range recommended
in the guidelines of the Academy of Breastfeeding Medicine,54,55 which suggest
8 to 12 BFs during the first 24 hours to 48 hours postbirth. In addition, maternal
perception of success with BF doubled from baseline after the first skin-to-skin BF
(SB1) and remained high thereafter. Upon study enrollment, several mothers had
reddened nipples, small blisters and nipple cracks. By SB3, however, their ratings
of nipple pain had decreased, suggesting that the SB feedings were less injurious.
The perception of breast pain was generally rated low throughout the study
protocol. Although saliva from only 5 mothers has been assayed to date, 4 of these
mothers exhibited decreases in salivary cortisol levels from pre-SB to post-SB, with
the lowest levels at SB4, suggesting lower levels of stress related to SB.
Conclusions are tentative because of the small sample size (N = 48), the one-group
design and the brief follow-up period. However, in the process of conducting
this study, we have gained numerous insights. These insights may be helpful in
understanding some of the problems underlying poor BF outcomes in the US, as
well as some of the reasons why many mothers do not choose to initiate BF, do not
continue BF once begun, do not advise others to BF and do not plan to BF their
subsequent children.
Firstly, although our study was conducted in a women’s hospital that has an
excellent reputation for providing outstanding care, the BF attempts made by these
mothers and their infants were interrupted with astonishing frequency—as often as
19 times in 3 hours and 13 times in just 1 hour. Most guidelines recommend 8 to

12 BFs in the first 24 hours to 48 hours.8,13,55 However, it must be quite difficult
to BF this often, given the frequency of interruptions observed in this study. Some
of these interruptions, however—taking the baby to the nursery for the genetic
screening heel stick; auditory screening; the physician’s physical examination; and
the injection of the hepatitis B vaccine—are considered necessary in our present
culture, and therefore are part of standard care. Each of these interruptions is
done at different times and cannot be done until the health care professional is
available. Furthermore, of the 16 male infants who were taken to the nursery for
circumcision, 15 (94%) were given nothing by mouth, except for acetaminophen
(Tylenol™). These infants were then kept in the nursery, in continued separation
from their mothers until the physician was available to perform the procedure. In
one case, 5 hours had elapsed and the circumcision was still not performed. The
decision was finally made to allow this infant to BF.
Other interruptions resulted from choices that were made, sometimes reluctantly,
by the mothers: For example, some mothers allowed a BF to be interrupted so that
their infants could be photographed, or sometimes they did not begin a pending
BF if visitors arrived or were soon expected. They reported that they wanted to be
hospitable, yet they also desired privacy at times and were reluctant to say this to
their visitors. Mothers who have recently given birth have physical needs as well,
such as pain that makes moving about difficult and feelings of fatigue that make
them fearful that they will fall asleep if they continue to hold their infants.56
Although interruptions are sometimes mentioned in a list of problems in the

literature regarding BF,10,37,57 these studies have often been conducted without
quantification or other emphasis. Thus, hospital personnel are probably unaware
of the high frequency with which these interruptions occur. To some degree, this
is because numerous personnel and visitors are responsible for these interruptions.
Thus, no single person may be aware of the frequency with which these
interruptions occur. If noticed at all, such interruptions may be viewed as having
little, if any, harmful impact, rather than having a cumulative effect that might
interfere seriously with the developing didactic process of mutual caregiving and
mutual regulation.58 The irony here is that the quiet times that are virtually free
of interruptions—late evening and during the night—are also the times when the
infants are usually in the nursery, rather than rooming-in with their mothers.
Secondly, infants’ sleep was interrupted, as well, a phenomenon first documented

with preterm infants by Duxbury and colleagues59 and by Korones.60 “It is by
accident, rather than by intent, that infants are left at rest,” said Korones (p. 94).60
Before long, we realized that we, ourselves, were amongst the interrupters in our
own study. We assumed incorrectly that we could predict the appropriate timing
of the next BF. Instead, we learned that we could not predict this very well at all.
The infants were the ones who knew, perhaps because they became fully rested and
developed hunger pangs. Thus, we learned that the next feeding must be allowed to
occur in a self-regulatory way, based on the quintessential indicators—infant cues,
such as awakening, rooting and hand-mouth activity. If the infants were awakened,
undressed and placed in SSC near their mothers’ breasts, they often fell asleep
again, which delayed the feeding and was upsetting to most mothers, or they failed
to attach (“latch on”) to the breast effectively. In contrast, if we waited until the
infants awakened spontaneously from uninterrupted sleep and gave early oral cues,
and if we then undressed them gently, they rarely cried. When we then placed them
in SSC near their mothers’ breasts, they usually sought and found the breasts rather
quickly, attached appropriately with ease, and fed well. Under these circumstances,
this successful and relatively independent effort sometimes occurred during the
first SB.
Thirdly, being gentle, considerate and respectful can make all the difference, not
only for the infants, but also for the mothers. Some mothers seem offended or
embarrassed by the way some health care personnel assist them with BF. This may
be one reason why many mothers who have chosen to breastfeed decide, while still
in the hospital, that “breastfeeding is not for me.” One common technique for
“getting the baby on” is to take hold of the mother’s breast with one hand and
the back of the baby’s head with the other hand, and, as the baby begins to open
his/her mouth, to push the baby’s mouth as far as possible onto the mother’s
breast.61 This technique is employed with the good intention of having the baby
latch onto the areola rather than the nipple. However, both the mother and her
infant become tense during this process, and the latch is usually neither ideal nor
fully effective. In the study by Mozingo and colleagues, mothers reported that this
kind of “insensitive care” by health care personnel was the chief reason they decided
to terminate BF within the first 2 weeks postbirth.62 One mother in the study
reported here referred to this as “aggressive treatment,” and soon thereafter decided
to feed with formula. Other mothers simply looked away and did not participate
actively in trying to learn how to position their infants most effectively.
Lastly, as time passes without successful BF, concern for hypoglycemia in the infant
intensifies, as do well-intended efforts by staff to rush the BF process or to resort to
supplementary or complementary feedings with formula. However, according to
the guidelines set by the Academy of Breastfeeding Medicine,54,55 and those set
forth by Eidelman,63 concern regarding hypoglycemia is unwarranted for healthy,
fullterm, appropriate-for-gestation infants who have no signs of illness. Eidelman
emphasized that his guidelines apply to infants who have been screened carefully
for risk and who are provided with what is generally considered ideal care, including
the initiation of SSC and BF within the first 30 to 60 minutes postbirth,(cf. 64) BF
that begins on cue before any crying, and BF that occurs every 1.5 to 2 hours in
the first 24 to 48 hours. This kind of care maintains normal temperature63 and
stimulates suckling and milk production.38 Similar care is also associated with
reduced breastfeeding jaundice,65 a later cause of mother-infant separation and even
rehospitalization. An important caveat is that this kind of care is difficult to achieve
consistently for all mother-infant dyads in current, typical hospital systems. Thus,
the concern regarding infant hypoglycemia remains a valid concern.
The first day, and especially the first few hours, postbirth is a crucial time for the
mother and her infant. Their future mutual comfort depends upon an investment
of time and effort by the mother during this brief window of developmental
opportunity when she and her infant are making major physiological adaptations—
the infant to extrauterine life and the mother to her postpregnant state. To
facilitate these processes, everything possible must be done to meet their needs in
a timely, self-regulatory, gentle and respectful way, including helping visitors to
understand. In so doing, hospital personnel can provide an additional benefit—
modeling these kinds of humane behaviors for mothers and their families.
Closing Thoughts
In closing, we are concerned—but also encouraged—by what has been reported

in this chapter. BF that is difficult or painful, or both, makes little sense to many
mothers or to most observers. This current study is relevant to racial, ethnic and
socioeconomic disparities because the most vulnerable women are those who are
most likely to falter when difficulties with BF appear. They say, “How can I do this
if it is so hard?” Perhaps we have identified, for the first time, a very practical
approach to BF difficulties that have already begun: a brief intervention, within

easy reach of practicing nurses, that can be used to assist mother-infant dyads who
are in the midst of settings that do not routinely allow self-regulatory BF and
mutual caregiving that begins with birth and continues thereafter.
The authors would like to thank all research staff (Susan Greene, Dianne Rafferty,
and Regina Tuma), hospital postpartum staff and all of the mothers, babies and
families who participated in the current study, as well as Medela, Inc., McHenry, IL
(breast pump accessories, and the loan of an electric Lactina breast pump) and
Exergen, Inc., Watertown, MA (Temporal ScannerTM, i.e., temporal artery
thermometry). This study was funded by the National Institutes of Health, the
National Institute of Nursing Research (#2R01 NR02444) and the General
Clinical Research Center (#M01 RR00080-36).
The Dr. Korones quotation has been used with permission of Ross Products
Division, Abbott Laboratories Inc., Columbus, OH 43215. From Korones SB.
Disturbance and infants’ rest. In: Iatrogenic Problems in Neonatal Intensive Care:
Report of the Sixty-Ninth Ross Conference on Pediatric Research (1976). Ross
Products Division, Abbott Laboratories Inc.60
Ref ere n c e s
1. Cadwell K. An international policy perspective on breastfeeding in the United States. In:
Cadwell K, ed. Reclaiming Breastfeeding for the United States: Protection, Promotion, and
Support. Sudbury, Mass: Jones & Bartlett; 2002:1-10.
2. US Department of Health and Human Services. Healthy People 2010. With Understanding
and Improving Health and Objectives for Improving Health. Vol 2. 2nd ed. Washington, DC:
US Government Printing Office; 2000.
3. Labbok MH. Effects of breastfeeding on the mother. In: Schanler RJ, ed. Breastfeeding 2001,
Part I: The evidence for breastfeeding. Pediatric Clinics of North America. 2001;48:143-158.
4. Lawrence RA, Lawrence RM. Breastfeeding: A Guide for the Medical Profession. 5th ed. New
York, NY: Mosby-Year Book; 1999.
5. Schanler R, ed. Breastfeeding 2001, Part I: The evidence for breastfeeding. The Pediatric
Clinics of North America. 2001;48:1-264.
6. United States Breastfeeding Committee. Benefits of Breastfeeding [issue paper]. Raleigh, NC:
United States Breastfeeding Committee; 2002.
7. US Department of Health and Human Services: Office on Women’s Health. Breastfeeding: A

Message from the Surgeon General. In: HHS blueprint for action on breastfeeding. Available
at: http://www.4woman.gov/Breastfeeding/bluprntbk2.pdf. Accessed September 12, 2003.
8. American Academy of Pediatrics. Work Group on Breastfeeding. Breastfeeding and the use
of human milk. Pediatrics. 1997;100:1035-1039.
9. World Health Organization. The optimal duration of exclusive breastfeeding: results of a
WHO systematic review. Available at: http://www.who.int/inf-pr-2001/en/note2001-07.html.
Accessed September 12, 2003.
10. US Department of Health and Human Services. Report of the Surgeon General’s Workshop on
Breastfeeding and Human Lactation. Rockville, Md: US Dept of Health and Human Services;
1984. USDHHS publication HRS-D-MC-84-2.
11. UNICEF/WHO. The Innocenti Declaration. Meeting on Breastfeeding in the 1990s: A
global initiative hosted by UNICEF/WHO, co-sponsored by the United States Agency for
International Development (AID) and the Swedish International Development Authority
(SIDA), Spedale degli Innocenti; 1990; Florence, Italy.
12. US Department of Health and Human Services. Healthy People 2000: Conference Edition.
Vol 2. Washington, DC: US Dept of Health and Human Services, Public Health and Service,
Office of the Assistant Secretary for Health; 1990. USDHHS publication PHS 91-50213.
13. United Nations Children’s Fund (UNICEF)/World Health Organization (WHO). The
UNICEF/Baby Friendly Hospital Initiative: Ten Steps to Successful Breastfeeding. New York, NY:
United Nations Children’s Fund; 1992.
14. Naylor AJ. Baby-Friendly Hospital Initiative. Protecting, promoting, and supporting
breastfeeding in the twenty-first century. In: Schanler RJ, ed. Breastfeeding 2001, Part II:
The management of breastfeeding. Pediatric Clinics of North America. 2001;48:475-483.
15. Turner-Maffei C. Using the Baby-Friendly Hospital Initiative to drive positive change. In:
Cadwell K, ed. Reclaiming Breastfeeding for the United States: Protection, Promotion, and
Support. Sudbury, Mass: Jones & Bartlett; 2002:23-33.
16. US Department of Health and Human Services. United States Breastfeeding Committee.
Breastfeeding in the United States: A National Agenda. Rockville, Md: US Department of
Health and Human Services, Health Resources and Services Administration, Maternal and
Child Health Bureau; 2001. USDHHS publication. Available at: http://www.usbreastfeed-
ing.org/USBC-Strategic-Plan-2001.pdf. Accessed September 12, 2003.
17. Riordan J, Auerbach KG. Breastfeeding and Human Lactation. 2nd ed. Boston, Mass: Jones
& Bartlett; 1999.
18. Woolridge MW. The ‘anatomy’ of infant sucking. Midwifery. 1986;2:164-171.

19. Lau C. Effects of stress on lactation. In: Schanler RJ, ed. Breastfeeding 2001, Part I: The
evidence for breastfeeding. Pediatric Clinics of North America. 2001;48:221-234.
20. Anderson GC. Risk in mother-infant separation postbirth. Image — the Journal of Nursing
Scholarship. 1989;21:196-199.
21. Anderson GC. Kangaroo care of the premature infant, In: Goldson E, ed. Nurturing the
premature infant: Developmental Interventions in the Neonatal Intensive Care Nursery. New
York, NY: Oxford University Press; 1999:131-160.
22. Anderson GC, Wood CE, Chang H-P. Self-regulatory mothering vs. nursery routine care
postbirth: effect on salivary cortisol and interactions with gender, feeding, and smoking.
Infant Behavior & Development. 1998;21(April):264.
23. Francis D, Diorio J, Liu D, Meaney MJ. Nongenomic transmission across generations
of maternal behavior and stress responses in the rat. Science. 1999;286:1155-1158.
24. Liu D, Caldji C, Sharma S, Plotsky PM, Meaney MJ. Influence of neonatal rearing
conditions on stress-induced adrenocorticotropin responses and norepinepherine release in
the hypothalamic paraventricular nucleus. The Journal of Neuroendocrinology. 2000;12:5-12.
25. Liu D, Diorio J, Day JC, Francis DD, Meaney MJ. Maternal care, hippocampal
synaptogenesis and cognitive development in rats. Nature Neuroscience. 2000;3:799-806.
26. Liu D, Diorio J, Tannenbaum B, et al. Maternal care, hippocampal glucocorticoid receptors,
and hypothalamic-pituitary-adrenal responses to stress. Science. 1997;277:1659-1662.
27. Sapolsky RM. The importance of a well-groomed child. Science. 1997;277:1620-1621.
28. Piazza PV, Maccari S, Deminiere JM, Le Moal M, Mormede P, Simon H. Corticosterone
levels determine individual vulnerability to amphetamine self-administration. Proceedings of
the National Academy of Sciences of the United States of America. 1991;88:2088-2092.
29. Christensson K, Siles C, Moreno L, et al. Temperature, metabolic adaptation and crying in
healthy fullterm newborns cared for skin-to-skin or in a cot. Acta Paediatrica (Oslo, Norway:
1992). 1992;81:488-493.
30. Field T. Touch. Cambridge, Mass: Massachusetts Institute of Technology; 2001.
31. Widstrom AM, Ransjo-Arvidson AB, Christensson K, Matthiesen AS, Winberg J,

Uvnäs-Moberg K. Gastric suction in healthy newborn infants. Effects on circulation and
developing feeding behaviour. Acta Paediatrica Scandinavica. 1987;76:566-572.
32. Kjellmer I, Winberg J. The neurobiology of infant-parent interaction in the newborn:
an introduction. Acta Paediatrica (Oslo, Norway: 1992). Supplement. 1994;397(suppl):1-2.
33. Syfrett EB, Anderson GC. Very early kangaroo care beginning at birth for 34-36 week
infants: effect on outcome. Paper presented at: Meeting on Kangaroo Mother Care at the
Bureau of International Health; October 24-26, 1996; Trieste, Italy.
34. Hurst NM, Valentine CJ, Renfro L, Burns P, Ferlic L. Skin-to-skin holding in the neonatal
intensive care unit influences maternal milk volume. Journal of Perinatology: Official Journal
of the California Perinatal Association. 1997;17:213-217.
35. Charpak N, Ruiz-Peláez JG, Figueroa de C Z, Charpak Y. Kangaroo mother versus

traditional care for newborn infants ≤2000 grams: a randomized, controlled trial. Pediatrics.
1997;100:682-688.
36. Meier PP. Breastfeeding in the special care nursery. Prematures and infants with medical
problems. In: Schanler RJ, ed. Breastfeeding 2001, Part II: The management of breastfeeding.
Pediatric Clinics of North America. 2001;48:425-442.
37. Harris H. Remedial co-bathing for breastfeeding difficulties. Breastfeeding Review.
1994;2:465-468.
38. Meyer K, Anderson GC. Using kangaroo care in a clinical setting with fullterm infants
having breastfeeding difficulties. MCN. The American Journal of Maternal Child Nursing.
1999;24:190-192.
39. Anderson GC, Moore E, Hepworth J, Bergman N. Early skin-to-skin contact for mothers
and their healthy newborn infants (Cochrane Review). In: The Cochrane Library, Issue 2,
2003. Oxford: Update Software.
40. Righard L, Alade MO. Effect of delivery room routines on success of first breastfeed. Lancet.
1990;336:1105-1107.
41. Odent M. Newborn weight loss. Mothering. 1989;13:72-73.
42. Dodgson JE, Allard-Hale CJ, Bramscher A, Brown F, Duckett L. Adherence to the ten steps
of the Baby-Friendly Hospital Initiative in Minnesota hospitals. Birth (Berkeley, Calif ).
1999;26:239-247.
43. Kovach AC. A 5-year follow-up study of hospital breastfeeding policies in the Philadelphia
area: a comparison with the ten steps. Journal of Human Lactation: Official Journal of
International Lactation Consultant Association. 2002;18:144-154.
44. Punthmatharith B. Randomized Controlled Trial of Early Kangaroo (Skin-to-Skin) Care: Effects
on Maternal Feelings, Maternal-Infant Interaction, and Breastfeeding Success in Thailand
[dissertation]. Cleveland, Ohio: Case Western Reserve University; 2000.
45. Mulford C. The Mother-Baby Assessment (MBA): an “Apgar score” for breastfeeding. Journal
of Human Lactation: Official Journal of International Lactation Consultant Association.
1992;8:79-82.
46. Youngblut JM, Casper GR. Single-item indicators in nursing research. Research in Nursing
& Health. 1993;16:459-465.
47. Labbok M, Krasovec K. Toward consistency in breastfeeding definitions. Studies in Family
Planning. 1990;21:226-230.
48. Cadwell CM. Defining breastfeeding in research. In: Cadwell K, ed. Reclaiming Breastfeeding
for the United States: Protection, Promotion, and Support. Sudbury, Mass: Jones & Bartlett;
2002:81-89.
49. Jensen D, Wallace S, Kelsay P. LATCH: a breastfeeding charting system and documentation
tool. Journal of Obstetric, Gynecologic, and Neonatal Nursing: JOGNN/NAACOG. 1994;23:
27-32.
50. Schlomer JA, Kemmerer J, Twiss JJ. Evaluating the association of two breastfeeding
assessment tools with breastfeeding problems and breastfeeding satisfaction. Journal of
Human Lactation: Official Journal of International Lactation Consultant Association.
1999;15:35-39.
51. Leff EW, Jefferis SC, Gagne MP. The development of the Maternal Breastfeeding Evaluation
Scale. Journal of Human Lactation: Official Journal of International Lactation Consultant
Association. 1994;10:105-111.
52. Gray L, Watt L, Blass EM. Skin-to-skin contact is analgesic in healthy newborns. Pediatrics
(Online) [serial online]. 2000;105:1-6. Available at:
http://www.pediatrics.org/cgi/content/full/105/1/e14. Accessed September 12, 2003.
53. Ryan AS. The resurgence of breastfeeding in the United States. Pediatrics. 1997;99:E12.
54. The Academy of Breastfeeding Medicine. ABM Protocols. Clinical protocol number 1 —
guidelines for glucose monitoring and treatment of hypoglycemia in term breastfed neonates.
November, 1999. Available at: http://www.bfmed.org. Accessed September 14, 2003.
55. The Academy of Breastfeeding Medicine. ABM Protocols. ABM clinical protocol number 3
— hospital guidelines for the use of supplementary feedings in the healthy term breastfed
neonate. 2001. Available at: http://www.bfmed.org. Accessed September 14, 2003.
56. Anderson GC, Chiu S-H, Dombrowski MAS, Swinth JT, Albert J, Wada N. Mother-
newborn contact in a randomized trial of kangaroo (skin-to-skin) care. Journal of Obstetric,
57. Dennis CL. Breastfeeding initiation and duration: a 1990-2000 literature review. Journal
of Obstetric, Gynecologic, and Neonatal Nursing: JOGNN/NAACOG. 2002;31:12-32.
58. Anderson GC. The mother and her newborn: mutual caregivers. Journal of Obstetric,
59. Duxbury ML, Henly SJ, Broz LJ, Armstrong GD, Wachdorf CM. Caregiver disruptions and
sleep of high-risk infants. Heart & Lung; the Journal of Critical Care. 1984;13:141-147.
60. Korones SB. Disturbance and infants’ rest. Paper presented in: Iatrogenic Problems in
Neonatal Intensive Care: Report of the Sixty-Ninth Ross [Laboratories] Conference on
Pediatric Research. February, 1976; Columbus, Ohio.
61. Auerbach KG. Evidence-based care and the breastfeeding couple: key concerns. Journal
of Midwifery & Women’s Health. 2000;45:205-211.
62. Mozingo JN, Davis MW, Droppleman PG, Merideth A. “It wasn’t working.” Women’s
experiences with short-term breastfeeding. MCN. The American Journal of Maternal Child
Nursing. 2000;25:120-126.
63. Eidelman AI. Hypoglycemia and the breastfed neonate. In: Schanler J, ed. Breastfeeding
2001, Part II: The management of breastfeeding. Pediatric Clinics of North America.
2001;48:377-387.
64. Durand R, Hodges S, LaRick S. The effect of skin-to-skin breastfeeding in the immediate
recovery period on newborn thermoregulation and blood glucose values. Neonatal Intensive
Care: The Journal of Perinatology-Neonatology. 1997;10:23-27.
65. Gartner LM, Herschel M. Jaundice and breastfeeding. In: Schanler RJ, ed. Breastfeeding
2001, Part II: The management of breastfeeding. Pediatric Clinics of North America.
2001;48:389-399.
SECTION III.
TO U C H I N P R E M AT U R E I N FA N T S
CHAPTER 8:
TA C T I L E S T I M U L AT I O N
O F N E O N ATA L
INTENSIVE CARE UNIT
PRETERM INFANTS
Lynda Law Harrison, RN, PhD, FAAN

University of Alabama School of Nursing
University of Alabama at Birmingham
Ab s t ra c t
This chapter presents a review of studies on tactile stimulation of preterm

infants in neonatal intensive care units (NICUs) and during the early months
following discharge from the NICU. The studies describe the types of touch
received by infants in the NICU as well as the supplemental tactile stimulation
provided in the NICU, including still, gentle touch, studies of stroking/massage
and studies of stroking combined with kinesthetic or vestibular stimulation.
This chapter also summarizes those studies that have described the types of
touch received by preterm infants in the early months following discharge from
the NICU, as well as those trials that have evaluated the effects of home-based,
supplemental, tactile stimulation programs.
140 Section III. Touch in Premature Infants
Introd uc t i o n
Annually, approximately 7.6% of live births in the United States are infants who
are born prematurely and weighing less than 2500 g.1 In the same 12-month period,
1.4% of live births are infants who weigh less than 1500 g, and therefore are
considered very-low-birth-weight (VLBW) infants.1 Goldson reported that 20%
to 60% of LBW infants have some developmental disability, and 10% to 20%
have significant adverse sequelae.2 Recently, there is increasing recognition of
the importance of providing appropriate types and amounts of environmental
stimulation to these fragile infants during the early weeks of life to reduce stress
and prevent complications, such as intraventricular hemorrhage (IVH), which
can increase the risk for subsequent neurological and developmental problems.
Although there has been debate over the years about whether preterm infants are
understimulated or overstimulated in the neonatal intensive care unit (NICU),
there is now a consensus that both problems exist.3
In recent years, there has been increasing recognition of the value of providing
individualized, developmental care (DC) in the NICU as a strategy to reduce
NICU-induced stress and promote optimal health and developmental outcomes in
affected infants.4-9 Although providing appropriate handling is generally considered
an important component of DC, few studies have evaluated specific types of
handling and tactile stimulation provided as a component of DC. Several
researchers have suggested the need to identify the types of touch that might
be provided to fragile preterm infants to afford them comfort and minimize the
stresses associated with the NICU environment.10 Tactile interaction with their
infants is also important for parents of preterm infants, since the desire to maintain
contact between parent and infant is a primary component of positive parent-infant
attachment.11 Parents of preterm infants often have limited interactions with their
babies during the early weeks of life, and thus are at risk for problems in the
development of secure parent-infant attachment relationships.
Most of the touch received by preterm infants who are hospitalized in NICUs
consists of touch associated with medical or nursing procedures. These infants
receive very little comforting touch.12-18 Adverse effects of the touch associated with
medical and nursing procedures include hypoxia, bradycardia, sleep disruptions and
increased intracranial pressure (ICP).14,17,19-24 Repeated episodes of hypoxia and
increased ICP may place preterm infants at increased risk for complications, such
Chapter 8: Tactile Stimulation of Neonatal Intensive Care Unit Preterm Infants 141
as IVH and subsequent neurodevelopmental delays. Because of concerns about the

adverse effects of handling associated with medical and nursing procedures, many
NICUs have instituted minimal handling policies; that is, limiting the amount of
handling provided to preterm infants by both parents and NICU staff members.
Unfortunately, such policies are generally applied to all touch, thereby limiting
potentially beneficial comforting and soothing touch as well as touch associated
with medical and nursing procedures.
Type s o f Ha nd l i ng / Ta c t i l e St i m ulation Re ce ive d

by In f a n t s i n t h e N I CU
Studies that have focused on describing the types of stimulation experienced by

infants in the NICU environment have generally concluded that preterm infants
receive high levels of stimulation in the NICU, but that much of this stimulation
is aversive, not contingent on the infants’ cues and responses, and involves handling
associated with medical and nursing procedures.25-27 Preterm infants may be handled
as much as 3.5 hours per day28 or 5% to 38% of observation times.12 Symon and
Cunningham used time-lapsed video recordings of 12 preterm infants in an NICU
on the first and third days postbirth.28 The number of handling episodes ranged
from 28 to 71, and ventilated infants received more handling than did infants
who were not on ventilators. Contrary to the perceptions of nurses caring for these
infants, there was no correlation between the infants’ gestational ages (GAs) and
their duration of handling.
In contrast to the frequent exposure to procedural touch, preterm infants in

NICUs receive relatively little comforting or soothing touch, such as stroking.
Using time-lapsed video recordings, Blackburn and Barnard12 examined caregiving
provided to 102 preterm infants: They reported that social stroking was the type
of contact observed least frequently. During the 6, 24-hour observation periods,
25 of the infants received no social stroking. Werner and Conway reported similar
findings based on 2, 55-minute observations of 11 preterm infants who were
23.5 weeks to 28.5 weeks GA.18 Only 4.4% of the contacts observed were
described as “comfort measures,” such as soothing touch, while only 3 of the
11 infants received any comforting touch. Such touch was provided by only one
of the nurses observed.
Harrison and Woods examined the types of touch provided by parents and
grandparents during 3 visits to their infants in the NICU and found that the types
and amounts of touch varied considerably.29 The sample included 36 preterm
infants who were between 25 weeks and 33 weeks GA at birth. The mean total
duration of touch during the parental visits was 17.5 minutes, with a standard
deviation of 22.6 minutes. Infants less than 28 weeks GA received less touch than
did infants who were more than 28 weeks GA. Mothers touched more than did
grandmothers, who touched more than fathers. The most frequent types of touch
provided by the parents included holding, stroking, simple contact or rubbing. In
another report from the same study, Harrison, Leeper and Yoon noted that there
were no differences in infants’ mean heart rate (HR) or oxygen saturation (OS)
levels during parent touch, compared with baseline or posttouch periods, but there
was more variability in these measures during the touch periods.30 In addition,
there were significantly more episodes of abnormally low OS levels during periods
of parent touch than during baseline period.30
Miller and Holditch-Davis observed preterm infants from 7:00 pm to 11:00 pm in

the NICU and compared the types of contacts provided to these infants by nurses
and parents.16 Nurses spent more time providing general care, such as feeding,
changing, bathing and taking vital signs, or providing high-level care, such as
respiratory care or needle sticks. Parents provided more contact, such as touching
and holding.
Preter m Infants’ Responses to Handling A ssociated

With Me d i c a l/ Nu r s i ng Pro c e d u re s
Findings from studies that have evaluated preterm infants’ responses to handling
associated with medical and nursing procedures have consistently demonstrated
adverse effects from such procedures, including hypoxia, bradycardia, sleep
disruptions, increased ICP and behavioral agitation.12-24,31 Long, Philip and Lucey
tracked continuous recordings of the heart rates, transcutaneous O2 (tcPO2) levels
and respiratory rates of 30 preterm infants for 20 hours per day during the first
5 days after birth.21 Half the infants were assigned to an experimental group and
half were assigned to a control group. Nurses and caregivers of control-group
infants were not allowed to observe the tcPO2 recordings, but nurses and caregivers
of the experimental-group infants were allowed to observe these measures and, as a

result, modified their care to minimize handling that was associated with decreased
O2 levels. Of the incidents of hypoxia that occurred in the control-group infants,
75% were associated with handling. Peters noted that even procedures such as
auscultating heart rates or applying blood pressure cuffs were associated with
hypoxia, O2 desaturation or increased ICP.23
Several researchers have evaluated the effects of swaddling, tucking or containment

as an intervention to reduce stress associated with aversive procedures in preterm
infants. Corff found that preterm infants who received facilitated tucking during
heel sticks demonstrated lower mean HR, crying time and sleep disruption time
than infants in a control group.32 Fearon, Kisilevsky, Hains, Muir and Tranmer
studied the effects of swaddling in 15 preterm infants following a heel lance.33 The
swaddling resulted in increased OS scores, suggesting that this type of touch may
help ameliorate the adverse effects of handling associated with medical and nursing
procedures. Infants less than 31 weeks postconceptional age (PCA) returned to
presampling facial activity patterns almost immediately after the heel stick, whereas
those older than 31 weeks PCA demonstrated prolonged behavioral disturbance, as
evidenced by facial activity indicating stress, which was reduced significantly by the
swaddling intervention. These findings suggest that very small preterm infants may
not have sufficient energy or neurological development to maintain behavioral
responses to stressors, and, therefore, future research should control for the effects
of gestational age when examining behavioral outcomes.
St ud i e s Eva l u a t i ng Su p p l e m e nt al Tactile
St i mula t i o n Int e r ve nt i o ns i n t he NIC U
To reduce the stressors associated with the NICU environment and excessive
handling associated with medical and nursing procedures, several researchers over
the past 30 years have examined the effects of interventions designed to provide
preterm infants with supplemental comforting touch. These interventions have
included still, gentle touch, stroking/massage or stroking/massage combined with
vestibular stimulation.
GENTLE TOUCH
Because of concerns that in the early weeks of life many preterm infants are too
fragile to tolerate stroking or massage, several researchers have evaluated the effects
of still, gentle human touch (GHT) or touch without stroking or massage. Jay
provided gentle touch to 13 preterm infants ventilated mechanically for 12 minutes,
4 times a day, beginning when the infants were less than 96 hours old.34 Compared
with a matched control group, the experimental infants had higher hematocrits
and required less O2. Tribotti studied the responses of 4 stable preterm infants
(33 weeks to 35 weeks gestation) to gentle touch provided for 15 minutes, 3 times
daily, for 3 days.35 During the first session, infants demonstrated decreased tcPO2,
increased respiratory regularity and a slight decrease in motor activity. By the
third session, they exhibited increases in tcPO2 and respiratory regularity, and a
continued decrease in motor activity.
Findings from 4 additional studies of GHT have been published in 6 separate

reports by Harrison and colleagues and by Modrcin-McCarthy.36-41 Infants enrolled
in these studies have ranged from 25 weeks to 33 weeks gestation, and the touch
interventions have been provided for 5 days to 10 days, beginning 6 days to 9 days
postbirth. The touch consisted of one of the investigators placing a hand on the
infant’s head and the other hand on the infant’s lower back and buttocks for 10
minutes to 20 minutes. In the largest of these studies, the sample consisted of 84
preterm infants between 27 weeks and 33 weeks GA at birth.38,39 In addition to the
usual care given in the NICU, infants in the experimental group received 3 periods
of GHT each day for 10 days, beginning when they were between 6 days to 9 days
of age. Each GHT lasted for up to 10 minutes, but was discontinued if the infant
demonstrated signs of physiological distress (HR <100 BPM or >200 BPM for
12 seconds or more, or arterial OS levels <90% for longer than 30 seconds). Fewer
than 23 out of 1280 GHT sessions (<2%) had to be discontinued early. Infants
whose GHT sessions had to be discontinued early had significantly lower GAs and
birth weights, higher morbidity and neurobiological risk scores and spent more days
on supplemental O2 and more days in the hospital compared with infants who did
not require early discontinuation of any GHT session. Continuous computerized
recordings were made of infant HR, OS and O2 concentration levels beginning
10 minutes before and ending 10 minutes after each touch episode. Infant motor
activity and behavioral state were scored during these same time periods. Infants
in the GHT and control groups were compared on several outcome measures, such
as length of hospital stay, average daily weight gain and morbidity status.
No significant differences emerged in mean HR when the investigators compared

baseline (B), touch (T) and posttouch (PT) phases. Significant decreases occurred
in OS from B to PT and from T to PT, although the differences in the means
during these phases were very small and not clinically significant. Significantly
lower levels of active sleep, motor activity, behavioral distress and modified
behavioral distress occurred during periods of GHT when compared with the
B and PT periods. No differences were noted on outcome variables between the
experimental and control group infants.
Findings from the studies of GHT suggest that this type of touch has no adverse
effects on mean HR or OS levels, but that such touch resulted in increased
respiratory regularity, decreased levels of active sleep, motor activity and behavioral
distress, and increased levels of quiet sleep during periods of gentle touch when
compared with B and PT periods.35-41 However, no differences were noted on
longer-term outcome measures, such as weight gain, morbidity scores or length of
hospital stay, in infants who received the GHT intervention when compared with
infants in the randomly assigned control groups.36,40,41
These findings from the studies of gentle touch suggest that nurses should
encourage parents and other caregivers to use GHT when handling physiologically
fragile preterm infants in the NICU. Gentle touch might be a useful intervention
to soothe infants who are agitated and stressed.
Studies Examining Preter m Infants’ Responses

to St ro ki n g Wi t h o u t Ki ne s t h e t i c Stimulation
Studies examining responses to stroking have focused primarily on both immediate

and longer-term outcomes. The findings from these studies have been inconsistent,
most likely due to differences in the GAs and morbidity levels of the infants
in the studies and differences in the types and amounts of touch provided.
Adamson-Macedo and Attree noted that the interchangeable use in the literature
of terms such as “massage,” “stroking” and “rubbing” has led to confusion and
misinterpretation of findings across studies.42 These authors suggested that the term
“massage” implies kneading, whereas the term “rubbing” implies friction. They
defined stroking as “…to pass the hand (or the fingertips) softly in one direction.”42
IMMEDIATE OUTCOMES OF STROKING

INTERVENTIONS
Kattwinkel, Nearman, Fanaroff, Katona and Klaus provided an intervention that

consisted of rubbing the extremities of 6 preterm infants (26 weeks to 31 weeks
GA at birth and 2 days to 35 days of age at the time of the study) for 5 minutes
of every 15 minutes for 3 hours.43 Infants exhibited decreased apnea during the
stimulation, compared with control periods. In contrast, McGehee and Eckerman
found that preterm infants (27 weeks to 32.5 weeks gestation) exhibited more
gasping, grunting, moving and state transitions during periods when they used
lighter tactile and auditory/tactile stimulation, compared with periods when they
had only visual and/or auditory stimulation.44 Oehler reported that tcPO2 levels of
preterm infants (26 weeks to 30 weeks gestation) tended to decrease during periods
of light stroking or stroking/talking, but that these levels remained the same or
increased during periods of auditory stimulation only.45 Oehler, Eckerman and
Wilson found that light stroking had adverse effects only for the more high-risk
infants, and the authors advocated further research to examine subgroup differences
among high- and low-risk preterm infants’ responses to touch.46 Infants’ responses
to stroking also differed according to their individual behavioral states at the time
of the stroking intervention. Light stroking led to increased activity levels when the
infants were in a quiet state or were visually alert, but there was no change in activity
levels or avoidance cues in response to stroking when they were in an active state.
Adamson-Macedo, de Roiste, Wilson, Hayes, Eaton and de Carvalho evaluated the

immediate responses of 11 ventilated preterm infants to 3 minutes to 4 minutes of
a systematic, light-stroking intervention called Touching and Caressing, Tender in
Care (TAC-TIC) therapy.47 There were no significant decreases in tcPO2 levels
during or after the TAC-TIC sessions. Data were also collected that examined
changes in tcPO2 levels before, during and after comparable periods of maternal
intuitive touch: Consistent with the findings reported by Harrison and Woods
described earlier in this chapter,29 there were significant decreases in O2 levels
during and after maternal touch.47 The researchers suggested that patterned
sequences of stroking movements might be more soothing than random touching
to preterm infants.
Daga, Ahuja and Lunkad studied the effects of maternal stroking of the backs of
7 preterm infants—during alternate gavage feedings—who were less than 32 weeks
gestation and weighed less than 1600 g.48 O2 saturation levels were significantly
higher in infants who received the stroking at 20 minutes and 30 minutes after the
feeding, compared with a control group.
Findings from the studies examining immediate responses to stroking/massage are

mixed, most likely influenced by the GAs and physiological statuses of the infants
who were studied and by the types of stroking that were provided. Beneficial effects
have included reduced apnea and increased O2 saturation levels. Adverse effects
have included increased motor activity and increased levels of behavioral distress
cues (eg, gasping, grunting and state transitions). Further research is needed to
evaluate the immediate effects of massage among infants of varying GAs and to
determine whether these effects are influenced by infant characteristics such as GA,
morbidity status and behavioral state at the time the touch is initiated.
LONGER-TERM OUTCOMES OF MODERATE

STROKING/MASSAGE INTERVENTIONS
Solkoff, Yaffe, Weintraub and Blasé provided stroking to the neck, back and arms
for 5 minutes each hour during the first 10 days of life to 10 preterm infants,
each of whom weighed between 1190 g and 1590 g at birth.49 Compared with
a control group, infants who received the stroking were more active during the
hospitalization period, regained their birth weights faster and had fewer
developmental anomalies at 7 months to 8 months postdischarge. Solkoff and
Matuszak provided 7.5 minutes of stroking during each of 16 hours per day for
10 days to 6 preterm infants.50 The infants, who were an average of 14.2 days of
age when the interventions began, had a mean GA of 31.2 weeks. Infants in the
experimental group demonstrated more rapid habituation to light and sound,
improved body tone, increased alertness, more consolability, more state changes
and more rapid avoidance of noxious stimuli than did control-group infants.
Kramer, Chamorro, Green and Knudtson provided 2 minutes to 3 minutes of

stroking before and after feedings each day for a minimum of 2 weeks—for a total
of 48 minutes of extra touch each day—to 8 preterm infants who had a mean
GA of 33 weeks.51 Infants in the experimental group had increased rates of social
development at 6 weeks and 3 months after transfer to a crib, compared with
control-group infants, although there was no difference in weight gain or plasma
cortisol levels following heel sticks.
Adamson-Macedo provided supplemental stroking using the TAC-TIC therapy

protocol for 10 minutes twice a day during the first week of life to a group of
31 preterm infants who had a mean GA of 32 weeks.52 For 15 infants, the tactile
intervention began within the first 48 hours after birth. For the remaining infants,
the touch began 49 hours to 120 hours postbirth. One week after the intervention,
compared with a control group of 35 newborns, infants in both experimental
groups lost significantly less weight, and infants in the early-stimulation group lost
the least amount of weight. Adamson-Macedo et al reported results of a follow-up
study of 8 preterm infants who had received TAC-TIC therapy in the NICU.53
At age 7 years, these 8 children were compared with 6 matched controls who had
received systematic tactile or auditory stimulation. The children who received the
TAC-TIC therapy scored higher on tests of mental, sequential and simultaneous
processing, and they also scored significantly higher on tests of general intelligence.
de Róiste and Bushnell evaluated the effects of a TAC-TIC intervention on 13

preterm infants who were compared with a matched control group of 13 newborns.54
The TAC-TIC stimulation was provided for 20 minutes each day, beginning on
the second or third day after birth and continuing to the day prior to discharge
(mean, 17 days). Infants in the experimental group progressed to total nipple
feeding and were discharged earlier, and they had significantly higher Bayley mental
development scores at age 15 months.
Harrison conducted a pilot study to evaluate a 2-week GHT plus 2-week massage
intervention on 12 infants who were 27 weeks to 30 weeks GA at birth.55 Infants
were assigned randomly and equally to an intervention group and a control group.
Infants in the control group received standard care in the NICU. In addition to
receiving standard care in the NICU, infants in the intervention group received
10 days of GHT for 10 minutes twice each day, beginning when they were 6 days
to 9 days old. These infants then received 10 days of massage intervention for 10
minutes twice each day, beginning at age 20 days to 23 days. Screens were placed
around the infants’ beds during the interventions so that the research assistant (RA)
who collected observational data on infants’ behaviors and behavioral states would
not know whether the infant was in the touch or control group. Five times during
each 10-minute intervention phase, the nurse placed her hands on the infant’s
bed—whether the infant was in the touch group or control group—and the RA
then looked around the screen to observe and record infant behavioral state,
motor activity and behavioral distress. The GAs and birth weights of infants in the
experimental and control groups were similar, although the mean weight of each
infant in the experimental group was 129 g more than each control-group infant.
Infants in the experimental group exhibited lower total and average daily morbidity
scores than those in the control group (total score, 25.8 versus 59.3, respectively),
fewer days in the hospital (44.0 versus 59.8 days) and on supplemental O2 (13.4
versus 15.0), lower neurobiological risk scores (0.6 versus 1.2), higher average daily
weight gain (18.2 g versus 16.9 g) and more optimal scores on the habituation,
orientation and autonomic stability Brazelton Neonatal Behavioral Assessment
Scale (BNBAS) subscales.55 Infants in the control group demonstrated slightly
higher scores on the motor, state regulation and range-of-state BNBAS subscales.
Experimental-group infants experienced slightly more quiet sleep and slightly
reduced levels of motor activity during the massage interventions, while control-
group infants experienced slightly less quiet sleep during the intervention (for
them, no touch) phase. Few changes were noted in O2 saturation comparing
baseline, touch and posttouch phases for experimental-group infants, but there
were slight increases in mean HR (approximately 6 BPM) during the massage
interventions. Because of the small sample size, it was not possible to assess the
statistical significance of these group differences. However, these findings suggest
that adding the 2-week massage component to the 2-week gentle touch intervention
resulted in positive middle-term outcomes that were not noted in previous studies
that included only a gentle touch intervention.
Findings from studies examining middle- and long-term outcomes of moderate

stroking/massage interventions suggest that such interventions have no adverse
effects. In addition, these interventions offer other benefits, including increased
weight gain, reduced levels of morbidity, reduced lengths of hospital stays,
improved behavioral organization and improved mental development and
increased activity levels.
Studies of Massage Combined With K ine sth e tic

or Ve st i b ula r St i m u l a t i o n
Studies examining the combined effects of massage and kinesthetic (passive

movement or range of motion) or vestibular (rocking) stimulation have generally
included infants who were physiologically stable at the time of the intervention.
Researchers have generally provided the interventions,56,57 although in at least one
study the intervention was provided by parents.58,59
The intervention protocols used have included stroking and massage combined
with passive range of motion of the limbs or rocking.56,58,61-71 Findings from these
studies suggest that providing supplemental tactile and kinesthetic or vestibular
stimulation offers middle- and long-term benefits, including the following:
increased daily weight gain63,64; increased secretion of urinary epinephrine and
norepinephrine62; increased levels of alertness, and more mature orientation,
motor skills and range of state61,63,71; and more mature habituation behaviors.61,73
Field and colleagues have conducted several studies evaluating the effects of a
15-minute massage protocol on stable preterm infants.61-64 This intervention
protocol consisted of 5 minutes of stroking, 5 minutes of kinesthetic stimulation
(passive flexion/extension movements) and then 5 minutes of stroking. The infants
were placed in a prone position during stroking and in a supine position during
kinesthetic stimulation. The initial study included 40 preterm infants, who had a
mean GA of 31 weeks and a mean birth weight of 1280 g and who were equally
assigned randomly to an experimental or control group.61 Experimental-group
infants received the intervention 3 times daily for 10 days, beginning when they
were considered medically stable and admitted to the “grower” nursery. These
infants averaged a 47% greater daily weight gain (mean, 25 g versus 17 g), were
more active and alert and had more mature habituation, orientation, motor and
range of state behavior on the BNBAS. In addition, they were discharged an
average of 6 days earlier than those in the control group, resulting in an average
hospital cost savings of nearly $3000 per infant.
In a second report from the same study, Scafidi, Field, Schanberg et al described
the responses of 14 of the experimental-group infants to periods of stimulation
and to no stimulation and to differences in responses to the tactile and kinesthetic
components of the intervention.63 During periods of stimulation, infants in the
experimental group spent more time in active sleep, less time in drowsy wakefulness
and more time exhibiting active limb movements. The infants were more active and
more likely to be in active sleep during tactile stimulation, compared with the
kinesthetic-stimulation phase of the intervention.
Similar findings were published 4 years later from a replication study using similar
methods, but with a different sample of 40 stable preterm infants.64 In this study,
infants in the experimental group averaged a 21% greater daily weight gain (mean,
34 g versus 28 g), were discharged on average 5 days earlier and had more mature
performance on the habituation cluster of the BNBAS scale. In an associated study,
levels of urinary norepinephrine and epinephrine increased from Day 1 to Day 10
for the experimental group, but not for the control group.62 The authors concluded
that the intervention promoted greater maturity of the sympathetic nervous system,
as evidenced by the increased catecholamine levels for the experimental-group
infants.
White-Traut and colleagues have conducted a series of studies evaluating an

intervention that includes auditory, tactile, visual and vestibular (ATVV)
stimulation.66,67,70,71,73 The intervention included 10 minutes of stroking combined
with infant-directed talk, followed by 5 minutes of rocking and continuous
attempts to initiate and maintain eye contact with the infant. This intervention
has been used with stable preterm infants as well as with infants who had
periventricular leukomalacia (PVL).
Infants have responded to ATVV stimulation, showing slight decreases in body

temperature and slight increases in heart and respiratory rates, “although the
changes in physiologic parameters have been within clinically normal limits.”67
In 2 of the studies by White-Traut and colleagues, that included infants with PVL,
infants who received the ATVV stimulation program in the NICU were discharged
an average of 9 days earlier than were infants in a randomly assigned control
group.70,71 In another of these studies, infants in the experimental group received
the ATVV stimulation from a researcher in the hospital and then at home from
their mothers until 2 months corrected age.73 Compared with infants in a randomly
assigned control group, experimental-group infants exhibited better mental and
motor development scores and a lower incidence of cerebral palsy at 1 year of age.
These differences were not statistically significant, perhaps because the 1-year
follow-up data were available for only 26 of the original 37 infants who were
enrolled in the study.
Morrow et al62 evaluated changes in TcPO2 across 4, 15-minute stroking/

kinesthetic stimulation sessions that occurred on the second and ninth days of
the 10-day massage program developed by Field et al.60 During the first of the
4 sessions, mean TcPO2 levels decreased significantly from baseline to the first
tactile period and from baseline to the kinesthetic period, but increased from the
stimulation periods to the posttouch period. There were no significant effects
caused by the tactile or kinesthetic interventions on TcPO2 during the fourth
session. Most of the instances of decreased TcPO2 levels occurred during the
kinesthetic portion of the intervention, suggesting that movement and pressure
applied to the legs during the range of motion intervention may have resulted in
artifactual decreases in TcPO2.
Rose and colleagues assigned 60 preterm infants (28 weeks to 36 weeks GA)
randomly to either a control or an experimental group.74 Infants in the experimental
group received 3, 20-minute tactile interventions 5 days per week, beginning
within the first 2 weeks after birth and continuing until 1 day to 2 days prior to
hospital discharge (mean, 13 days). The tactile interventions consisted of massaging
the infant while the infant was in prone, supine and sitting positions. When the
infant was stable, one of the 3 sessions consisted of rocking the infant in a rocking
chair. Just prior to hospital discharge, experimental-group infants demonstrated
significantly greater cardiac responses to stimulation with a plastic filament.
The researchers also reported that these infants exhibited improved auditory
responsiveness during the neonatal period and improved visual recognition memory
at 6 months of age.
Several studies have evaluated tactile and kinesthetic/vestibular interventions

provided in the NICU by parents of preterm infants. Whipple assigned 20 parents
of preterm infants randomly to a control or an experimental group.59 Parents of
experimental-group infants received 1 hour of instruction on the appropriate uses
of multimodal stimulation, including massage and music, and ways to identify
and avoid overstimulation of their infants. Parents in the experimental group
subsequently demonstrated more appropriate interactions with their infants and
visited their infants in the NICU more often. Infants in the experimental group
demonstrated fewer stress behaviors, had greater weight gain (7.9 g more per day)
and a shorter length of hospitalization (16.1 days fewer) than did control infants,
although these differences were not significant.
White-Traut and Nelson evaluated the effects of an ATVV intervention

administered by mothers 4 times over the first 3 days postbirth to a group of stable
preterm infants who were 28 weeks to 35 weeks GA.58 Compared with mothers
who were asked simply to talk or sing to their infants at comparable times, and
compared with mothers in a control group, mothers in the ATVV group enjoyed
more positive interactions with their infants, as measured by the Nursing Child
Assessment and Feeding Scale (NCAFS) at the time of discharge from the NICU.
Vickers and colleagues published a review of 14 randomized trials of human tactile

stimulation provided to preterm infants in which at least one outcome was reported
that assessed weight gain, length of hospital stay, behavior or development.10 These
authors reported that most of the subjects in the studies reviewed were 30 weeks
to 33 weeks GA and were medically stable before the start of the intervention.
Tactile interventions were categorized into “massage” and “still, gentle touch.” The
investigators noted that many of the massage interventions also included kinesthetic
stimulation, and that although most interventions were provided for 5 days to
10 days, the study reported by Rice56 included massage provided by mothers for 30
days after hospital discharge. The findings of this review are summarized in Table 1.
Table 1. Findings From Studies on Tactile Stimulation of Preterm Infants56
Massage and kinesthetic stimulation enhanced weight gain by 5.1 g/d.

(This represents a 15% to 20% increase over the 30 g/d typical weight gain for
a growing preterm infant.)
• Hospital stays for infants receiving massage was 6 days less than for infants in control
groups. This finding has significant implications for healthcare costs, given the average
daily charges for an infant in a NICU. Additional research is needed to determine whether
such reduced lengths of stay can be replicated with larger and more-diverse samples.
• Infants receiving massage had more-optimal scores on the BNBAS for habituation,
motor maturity and range of state.
• Evidence is insufficient to indicate the effects of massage on other developmental

outcomes, such as Bayley exam scores.
• Massage reduced postnatal complications, although 95% of the weighting for this
analysis came from a study that had a sample of only 30 infants.
• Gentle still touch had no effect on morbidity or weight gain.
• Massage, kinesthetic stimulation and gentle touch had no adverse effects on the
infants studied.
Vickers and colleagues, however, noted that there were methodological problems
in some of the reviewed studies, including failures to blind observers who assessed
outcomes and to ensure that infants in experimental and control groups were treated
similarly in all aspects other than the experimental intervention.10 Because of these
methodological problems, these investigators concluded that there was insufficient
evidence to determine whether massage should be provided to infants in the NICU.
Furthermore, they advocated additional research to assess the effects of massage
interventions on clinical outcomes, controlling for methodological problems by
concealing treatment allocation until the subject was entered into the trial, ensuring
that infants were treated similarly in all respects other than the experimental
intervention, blinding observers assessing outcomes to the treatment assignment,
assessing whether there are differences in the withdrawals from the study based on
treatment-group assignment and assessing longer-term developmental outcomes.
Stu d i e s o f To u c h i n Pre t e r m Inf a nts A fte r

Di sc ha r g e Fro m N I CU
Studies of touch in preterm infants following discharge from the NICU have
focused on evaluating a massage intervention administered by mothers provided
in the home and examining relationships between characteristics of mothers’ touch
and the subsequent development of the parent-infant attachment relationship and
of the infant who is born prematurely. Rice studied the effects of an intervention
in which mothers were taught to provide their preterm infants a tactile/kinesthetic
intervention at home after discharge from the NICU.56,57 The sample included
30 infants born prematurely who were assigned randomly to an experimental or a
control group. Mothers of infants in the experimental group were asked to provide
their newborns the intervention for 15 minutes, 4 times each day, beginning on
the day when the infants came home from the hospital. Mothers held their infants
during the intervention, which consisted of a sequence of strokes using the
fingertips and palms and designed to provide more touching of the head than
other body parts. Following the stroking, mothers swaddled their infants and
rocked them for 5 minutes. Public health nurses visited the mothers in the
experimental group each day for the 30-day treatment period. Nurses visited
mothers in the control group only once per week, providing them with infant-care
information. Infants who received the massage by their mothers gained more
weight, demonstrated greater neurological maturation and had higher Bayley

Mental Exam scores at 4 months of age.
Weiss and colleagues75 videotaped 131 mothers and their low-birth-weight infants
3 months after discharge from a NICU (as a component of a larger longitudinal
study). The videotapes were coded using the Tactile Interaction Indicator to
describe the types and amounts of touch provided during feedings. Nearly half
(47%) of the mothers’ touch was coded as “nurturing,” but 17% of the mothers
provided no nurturing touch. There was a positive relationship between the use of
nurturing touch and subsequent development of secure attachment (at 12 months
of age) for the less-vulnerable infants. However, more-vulnerable infants, who
received more-nurturing touch, were more likely to develop insecure attachments.
The investigators concluded that the more-vulnerable infants might be more
susceptible to overstimulation.
In a subsequent report from the same study, Weiss and coinvestigators examined
relationships between the qualities of mother touch provided during the feeding
at 3 months after NICU discharge and the child’s development at 2 years of age.76
Infants whose mothers used more-nurturing touch had fewer behavior problems,
such as depression and anxiety, at 2 years of age. The authors concluded that the
types of mother touch provided during the early months may influence subsequent
psychosocial development of low-birth-weight (LBW) infants, and that nurturing
touch may provide a protective effect for the infants by helping them develop self-
regulatory skills. The authors also concluded that there was a need for additional
studies of LBW infants to examine relationships between early mother touch and
subsequent child development, as well as to test the effects of different tactile
stimulation programs on parent-child interaction and on child development.
Su mma r y a nd D i s c u s s i o n
Findings from the studies reviewed suggest that infants in NICUs receive relatively
little nurturing touch.12,18 Most of the touch they receive is related to medical or
nursing procedures, and much of this touch has adverse effects, including hypoxia,
bradycardia, increased ICP and sleep disturbances.21,23,24 Gentle touch is safe for
physiologically fragile infants in the early weeks of life. It also has a soothing effect,
as evidenced by decreased motor activity and lessened behavioral distress.39 Stroking

and massage may have adverse effects on physiologically fragile preterm infants.45,46
However, massage and kinesthetic/vestibular stimulation interventions provided
in the NICU or in the home after discharge from the NICU have many benefits
for stable, “growing,” preterm infants, including enhanced weight gain, improved
performance on the BNBAS exam, a reduced length of hospital stay and improved
developmental outcomes.53,54,58,61,64,66,67 Results from some studies suggest that the
types of touch provided by mothers to their preterm infants may have long-term
effects on the quality of the mother-infant attachment relationship, as well as on
the child’s subsequent behavior and development.75,76 Findings from studies in
which mothers were taught to massage their preterm infants suggest that such
interventions benefit not only the infant, but also have positive effects on the
mother-infant relationship.56-58
Additional research is needed to address the methodological limitations in the

studies of tactile interventions provided in NICUs to control for the limitations
identified in the meta-analysis by Vickers and colleagues.10 Studies are also needed
to examine the relationships between the quality and quantity of touch provided
by caregivers after the infant is discharged from the NICU, the infant’s subsequent
development and the caregiver-infant relationship. Lastly, there is a need for
research that will examine the mechanisms by which early tactile experience
influences infant development.
Re f e re n c e s
1. Martin JA, Hamilton BE, Ventura SJ, Menacker F, Park MM. Births: Final data for 2000.
National Vital Statistics Reports. National Center for Health Statistics. Vol 50. No. 5.
Available at: http://www.cdc.gov/nchs/data/nvsr/nvsr50/nvsr50_05.pdf. Accessed August 27,
2003.
2. Goldson E. The developmental consequences of prematurity. In: Wolraich ML, ed. Disorders
of Development and Learning. 3rd ed. Hamilton, Ontario: BC Decker; 2003:345-360.
3. Goldson E. The environment of the neonatal intensive care unit. In: Goldson E, ed.
Nurturing the Premature Infant: Developmental Interventions in the Neonatal Intensive Care
Nursery. New York, NY: Oxford University Press;1999:3-17.
4. Als H. A synactive model of neonatal behavioral organization: framework for the assessment
and support of the neurobehavioral development of the premature infant and his parents
in the environment of the neonatal intensive care unit. Physical & Occupational Therapy in
Pediatrics. 1986;6:3-55.
5. Als H, Lawhon G, Duffy FH, McAnulty GB, Gibes-Grossman R, Blickman JG.
Individualized developmental care for the very low-birth-weight preterm infant: medical and
neurofunctional effects. Journal of the American Medical Association. 1994;272:853-858.
6. Ashbaugh JB, Leick-Rude MK, Kilbride HW. Developmental care teams in the neonatal
intensive care unit: survey on current status. Journal of Perinatology: Official Journal of the
California Perinatal Association. 1999;19:48-52.
7. Brown LD, Heermann JA. The effect of developmental care on preterm infant outcome.
Applied Nursing Research: ANR. 1997;10:190-197.
8. Buehler DM, Als H, Duffy FH, McAnulty GB, Liederman J. Effectiveness of individualized
developmental care for low-risk preterm infants: behavioral and electrophysiologic evidence.
Pediatrics. 1995;96:923-932.
9. Fleisher BE, VandenBerg K, Constantinou J. Individualized developmental care for

very-low-birth-weight premature infants. Clinical Pediatrics. 1995;34:523-529.
10. Vickers A, Ohlsson AJ, Lacy JB, Horsley A. Massage for promoting growth and development
of preterm and/or low birth weight infants. In: The Cochrane Library. 2003, Oxford:1-24
11. Beal JA. The Brazelton Neonatal Behavioral Assessment Scale: a tool to enhance parental
attachment. Journal of Pediatric Nursing. 1986;1:170-177.
12. Blackburn S, Barnard K. Analysis of care giving events relating to preterm infants in the
special care unit. In: Gottfried AW, Gaiter JL, eds. Infant Stress Under Intensive Care:
Environmental Neonatology. Baltimore, Md: University Park Press; 1985:113-130.
13. Catlett AT, Holditch-Davis D. Environmental stimulation of the acutely ill premature infant:
physiological effects and nursing implications. Neonatal Network: NN. 1990;8:19-26.
14. Gottfried AW, Hodgman JE, Brown KW. How intensive is newborn intensive care? An
environmental analysis. Pediatrics. 1984;74:292-294.
15. Lawson K, Daum C, Turkewitz G. Environmental characteristics of a neonatal intensive-care

unit. Child Development. 1977;48:1633-1639.
16. Miller DB, Holditch-Davis D. Interactions of parents and nurses with high-risk preterm
infants. Research in Nursing & Health. 1992;15:187-197.
17. Murdoch DR, Darlow BA. Handling during neonatal intensive care. Archives of Disease
in Childhood. 1984;59:957-961.
18. Werner NP, Conway AE. Caregiver contacts experienced by premature infants in the neonatal
intensive care unit. Maternal-Child Nursing Journal. 1990;19:21-43.
19. Danford DA, Miske S, Headley J, Nelson RM. Effects of routine care procedures on
transcutaneous oxygen in neonates: a quantitative approach. Archives of Disease in Childhood.
1983;58:20-23.
20. Evans JC. Incidence of hypoxemia associated with caregiving in premature infants. Neonatal
Network: NN. 1991;10:17-24.
21. Long JG, Philip AG, Lucey JF. Excessive handling as a cause of hypoxemia. Pediatrics.
1980;65:203-207.
22. Norris S, Campbell LA, Brenkert S. Nursing procedures and alterations in transcutaneous
oxygen tension in premature infants. Nursing Research. 1982;31:330-336.
23. Peters KL. Does routine nursing care complicate the physiologic status of the premature
neonate with respiratory distress syndrome. The Journal of Perinatal & Neonatal Nursing.
1992;6:67-84.
24. Peters KL. Neonatal stress reactivity and cortisol. The Journal of Perinatal & Neonatal
Nursing. 1998;11:45-49.
25. Harrison L. Research utilization: handling preterm infants in the NICU. Neonatal Network:
NN. 1997;16:65-69.
26. Harrison L, Bodin MB. Effects of tactile stimulation on preterm infants: an integrative
review of the literature with nursing implications. Online Journal of Knowledge Synthesis
in Nursing [serial online]. 1994;1:6. Available at: http://www.stti.iupui.edu/
VirginiaHendersonLibrary/OJKSNMenu.aspx. Accessed August 29, 2003.
27. Linn PL, Horowitz FD, Buddin BJ, Leaker JC, Fox HA. An ecological description of a
neonatal intensive care unit. In: Gottfried AW, Gaiter JL, eds. Infant Stress Under Intensive
Care: Environmental Neonatology. Baltimore, Md: University Park Press; 1985:83-111.
28. Symon A, Cunningham S. Handling premature neonates: a study using time-lapse video.
Nursing Times. 1995;91:35-37.
29. Harrison L, Woods S. Early parental touch and preterm infants. Journal of Obstetric,
30. Harrison L, Leeper JD, Yoon M. Effects of early parent touch on preterm infants’ heart rates
and arterial oxygen saturation levels. Journal of Advanced Nursing. 1990;15:877-885.
31. Speidel BD. Adverse effects of routine procedures on preterm infants. Lancet. 1978;
1:864-866.
32. Corff KE. An effective comfort measure for minor pain and stress in preterm infants:
facilitated tucking. Neonatal Network: NN. 1993;12:74.
33. Fearon I, Kisilevsky BS, Hains SMJ, Muir DW, Tranmer J. Swaddling after heel lance:
age-specific effects on behavioral recovery in preterm infants. Journal of Developmental and
Behavioral Pediatrics: JDBP. 1997;18:222-232.
34. Jay SS. The effects of gentle human touch on mechanically ventilated very-short-gestation
infants. Maternal-Child Nursing Journal. 1982;11:199-259.
35. Tribotti SJ. Effects of gentle touch on the premature infant. In: Gunzenhauser N, ed.
Advances in Touch: New Implications in Human Development. [Pediatric Round Table Series.
Vol 14]. Skillman, NJ: Johnson & Johnson Baby; 1990:80-89.
36. Harrison L, Olivet L, Cunningham K, Bodin MB, Hicks C. Effects of gentle human touch
on preterm infants: pilot study results. Neonatal Network: NN. 1996;15:35-42.
37. Harrison L, Groer M, Modrcin-McCarthy MA, Wilkinson J. Effects of gentle human touch
on preterm infants: results from a pilot study. Infant Behavior and Development. 1992;15:12.
38. Harrison L, Williams AK, Berbaum ML, Stem JT, Leeper J. Effects of developmental,
health status, behavioral, and environmental variables on preterm infants’ responses to a
gentle human touch intervention. International Journal of Prenatal and Perinatal Psychology
and Medicine. 2000;12:109-122.
39. Harrison L, Williams AK, Berbaum ML, Stem JT, Leeper J. Physiologic and behavioral
effects of gentle human touch on preterm infants. Research in Nursing & Health.
2000;23:435-446.
40. Modrcin-McCarthy MA. The Physiological and Behavioral Effects of a Gentle Human Touch
Nursing Intervention on Preterm Infants [dissertation]. Knoxville, Tenn: University of
Tennessee; 1992.
41. Modrcin-Talbott MA, Harrison LL, Groer MW, Younger MS. The behavioral effects of
gentle human touch on preterm infants. Nursing Science Quarterly. 2003;16:60-67.
42. Adamson-Macedo EN, Attree JLA. TAC-TIC therapy: the importance of systematic stroking.
British Journal of Midwifery. 1994;2:264-269.
43. Kattwinkel J, Nearman HS, Fanaroff AA, Katona PG, Klaus MH. Apnea of prematurity.
Comparative therapeutic effects of cutaneous stimulation and nasal continuous positive
airway pressure. The Journal of Pediatrics. 1975;86:588-592.
44. McGehee IJ, Eckerman CO. The preterm infant as social partner: responsive but unreadable.
Infant Behavior and Development. 1983;6:461-470.
45. Oehler JM. Examining the issue of tactile stimulation for preterm infants. Neonatal Network:
NN. 1985;4:25-33.
46. Oehler JM, Eckerman CO, Wilson WH. Social stimulation and the regulation of premature
infants; state prior to term age. Infant Behavior and Development. 1988;11:333-351.
47. Adamson-Macedo EN, de Roiste A, Wilson A, Hayes JA, Eaton B, de Carvalho FA.
Systematic gentle/light stroking and maternal random touching of ventilated preterms: a
preliminary study. International Journal of Prenatal and Perinatal Psychology and Medicine.
1997;9:17-31.
48. Daga SR, Ahuja VK, Lunkad NG. A warm touch improves oxygenation in newborn babies.
Journal of Tropical Pediatrics. 1998;44:170-172.
49. Solkoff N, Yaffe S, Weintraub D, Blasé B. Effects of handling and the subsequent
development of premature infants. Developmental Psychology. 1969;1:765-768.
50. Solkoff N, Matuszak D. Tactile stimulation and behavioral development among
low-birthweight infants. Child Psychiatry and Human Development. 1975;6:33-37.
51. Kramer M, Chamorro I, Green D, Knudtson F. Extra tactile stimulation of the premature
infant. Nursing Research. 1975;24:324-334
52. Adamson-Macedo EN. Effects of tactile stimulation on low and very low birth weight infants
during the first week of life. Current Psychological Research and Reviews. 1985-1986;(winter
issue):305-308.
53. Adamson-Macedo EN, Dattani II, Wilson A, de Carvalho FA. A small sample follow-up
study of children who received tactile stimulation after pre-term birth: intelligence and
achievements. Journal of Reproductive and Infant Psychology. 1993;11:165-168.
54. de Róiste A, Bushnell IWR. Tactile stimulation: short- and long-term benefits for pre-term
infants. British Journal of Developmental Psychology. 1996;14:41-53.
55. Harrison L. Effects of a gentle touch and massage intervention on preterm infants: pilot
study results. Paper presented at: National Conference of the National Association of
Neonatal Nurses; October 3-6, 2001; Salt Lake City, Utah.
56. Rice RD. Neurophysiological development in premature infants following stimulation.

57. Rice RD. The effects of the Rice infant sensorimotor stimulation treatment on the
development of high-risk infants. Birth Defects: Original Article Series. 1979;15:7-26.
58. White-Traut RC, Nelson MN. Maternally administered tactile, auditory, visual, and
vestibular stimulation: relationship to later interactions between mothers and premature
infants. Research in Nursing & Health. 1988;11:31-39.
59. Whipple, J. The effect of parent training in music and multimodal stimulation on
parent-neonate interactions in the neonatal intensive care unit. Journal of Music Therapy.
2000;37:250-268.
61. Kuhn CM, Schanberg SM, Field T. Tactile-kinesthetic stimulation effects on sympathetic
and adrenocortical function in preterm infants. The Journal of Pediatrics. 1991;119:434-440.
62. Morrow CJ, Field TM, Scafidi FA, et al. Differential effects of massage and heel stick
procedures on transcutaneous oxygen tension in preterm neonates. Infant Behavior and
Development. 1991;14:397-414.
63. Scafidi FA, Field TM, Schanberg SM, et al. Effects of tactile/kinesthetic stimulation on the
clinical course and sleep/wake behavior of preterm neonates. Infant Behavior and
Development. 1986;9:91-105.
64. Scafidi FA, Field TM, Schanberg SM, et al. Massage stimulates growth in preterm infants:
a replication. Infant Behavior and Development. 1990;13:167-188.
65. White-Traut RC, Tubeszewski KA. Multimodal stimulation of the premature infant. Journal
of Pediatric Nursing. 1986;1:90-95.
66. White-Traut RC, Goldman MBC. Premature infant massage: is it safe? Pediatric Nursing.
1988;14:285-289.
67. White-Traut RC, Nelson MN, Silvestri JM, Patel MK, Kilgallon D. Patterns of physiologic
and behavioral response of intermediate care preterm infants to intervention. Pediatric
Nursing. 1993;19:625-629.
68. White-Traut RC, Nelson MN, Burns K, Cunningham N. Environmental influences on the
developing premature infant: theoretical issues and applications to practice. Journal of
Obstetric, Gynecologic, and Neonatal Nursing. 1994;23:393-401.
69. White-Traut RC, Nelson MN, Silvestri JM, Cunningham N, Patel M. Responses of preterm
infants to unimodal and multimodal sensory intervention. Pediatric Nursing. 1997;23:
169-175, 193.
70. White-Traut RC, Nelson MN, Silvestri JM, et al. Developmental intervention for preterm
infants diagnosed with periventricular leukomalacia. Research in Nursing & Health.
1999;22:131-143.
71. White-Traut RC, Nelson MN, Silvestri JM, et al. Effect of auditory, tactile, visual, and
vestibular intervention on length of stay, alertness, and feeding progression in preterm infants.
Developmental Medicine and Child Neurology. 2002;44:91-97.
72. Schanberg S, Kuhn CM, Field TM, Bartolome JV. Maternal deprivation and growth
suppression. In: Gunzenhauser N, ed. Advances in Touch: New Implications in Human
Development [Pediatric Round Table Series. Vol 14]. Skillman, NJ: Johnson & Johnson Baby;
1990:3-10.
73. Nelson MN, White-Traut RC, Vasan U, et al. One-year outcome of auditory-tactile-visual-
vestibular intervention in the neonatal intensive care unit: effects of severe prematurity and
central nervous system injury. Journal of Child Neurology. 2001;16:493-498.
74. Rose SA, Schmidt K, Riese ML, Bridger WH. Effects of prematurity and early intervention
on responsivity to tactual stimuli: a comparison of preterm and full-term infants. Child
Development. 1980;51:416-425.
75. Weiss SJ, Wilson P, Hertenstein MJ, Campos R. The tactile context of a mother’s care giving:
implications for attachment of low birth weight infants. Infant Behavior & Development.
2000;23:91-111.
76. Weiss SJ, Wilson P, Seed MS, Paul SM. Early tactile experience of low birth weight children:
links to later mental health and social adaptation. Infant and Child Development. 2001;10:
93-115.
CHAPTER 9:
PRETERM AND
FULL-TERM
INFANT MASSAGE
IN CHINA
Zhu Jianxing, MD, MSc

Xinhua Hospital
Shanghai Second Medical University
Ab s t ra c t
During the current study, 405 healthy and sick term and preterm infants were
provided massage 3 times per day, for 15 minutes each, for 10 days. Three
different types of massage were compared; including the type discussed by Field
and colleagues in their article published in Pediatrics in 1986 that involves
massaging the entire body (overall body massage), a simpler massage covering
the head, abdomen, hands and feet, and a simple massage plus the rubbing of
acupressure points. Although no significant differences were noted in formula
intake, greater increases in daily weight were noted in the group that received
the overall body massage and the group that received massage with acupressure
points. No significant differences were noted in head circumference or body length.
Preterm sick babies also benefited from the overall massage by showing a lesser
decrease in their hemoglobin status than did infants in the other groups.
Introd uc t i o n
A recent meta-analysis performed by Ottenbacher and colleagues documented

positive effects for most infants receiving massage therapy.1 Most of the studies
included in the meta-analysis reported greater daily weight gain and increased
performance on newborn assessments and developmental assessments for the
infants who received massage therapy.2,3
Although practitioners in China developed infant massage many centuries ago,

and American massage was derived from massage therapy developed in China and
India, Chinese infants of this century have not received massage therapy until very
recently. Many hospitals in China—including those in Shanghai, where this study
was conducted—now routinely teach parents infant massage.
The C hi n a St u d y
METHODS AND PROCEDURES
The sample of 405 infants came from 6 neonatal clinics, including a large children’s
hospital and 2 municipal-level and 2 district-level maternal-infant medical care
centers. Three different types of massage were assessed, including the procedure
developed by Field and colleagues,2 currently being used in North America, the
Philippines, Korea and Israel. This involves massaging the infant’s entire body,
using moderate pressure. The second method that was tried is called the domestic
simple-touch massage, in which the head (namely the forehead and the face),
abdomen, hands, ankles and feet were massaged. The third method was the same
domestic simple-touch procedure plus the rubbing of acupressure points. Each
procedure was practiced for 15 minutes, 3 times per day, for 10 days. Of note
was that these massages were performed by specially trained physicians and nurses,
unlike most other massage therapy studies which have used parents or trained
research assistants.
The sample of babies was also different from most other massage therapy studies in
that it included normal term and preterm and sick term and preterm neonates. The
Chapter 9: Preterm and Full-term Infant Massage in China 165
infants labeled “sick” had been hospitalized with any of the following diseases to
a mild degree: pneumonia, apnea, hyperbilirubinemia and scleroderma without
complications. The test neonates and controls were assigned randomly to the
following 6 groups: normal term infants and controls who received the overall
body massage; sick term infants and controls who received the overall body
massage; sick preterm infants and controls who received the overall body massage;
normal term infants and controls who received the simple massage; normal term
infants and controls who received the simple massage plus the rubbing of
acupressure points; and sick preterm infants and controls who received the simple
massage plus the rubbing of acupressure points. Several measures were recorded:
growth items, hemoglobin status, formula intake and performance on a neonatal
behavioral and neurological assessment.
R E S U LT S
The data analyses suggested that the groups did not differ on the consumption
of formula. The groups did differ, however, on a number of growth measures.
The full-term infants who received the overall massage, the sick term infants
who received the overall massage and the normal preterm infants who received
the rubbing of acupressure points showed greater weight gain (49 versus 41 grams
per day for the term infants who received overall massage versus their controls;
35 versus 27 grams per day for the preterm infants who received overall massage
versus their controls; and 27 versus 22 grams per day for the preterm infants who
received the acupressure versus their controls). Body length and head circumference
were positively affected but only for the term group who received the overall body
massage. Similarly, hemoglobin status was positively affected but only for the sick
preterm group who received the overall body massage. This was indicated by a
lesser decrease than usually occurs and that occurred for the control group—1.98
versus 5.20 g/dL per 10 days. The cumulative scores on the neonatal behavioral
and neurological assessment were affected positively for the preterm infants who
received overall body massage (4.16 versus 2.63 for the controls), for the preterm
sick infants who received the overall body massage (2.36 versus 1.78 for the
controls) and for the term infants who received the simple massage plus acupressure
(6.30 versus 5.14 for the controls).
Conclusions
The results of this study, performed on healthy and sick term and preterm infants
in China, suggest that daily weight gain, as well as head circumference and body
length, is affected positively by massage therapy. The results also suggest that even
sick preterm infants can benefit from massage therapy: The infants exhibited a
lesser decrease in their hemoglobin status following the massage. Consistent with
many other studies, the groups who received overall body massage and massage
with acupressure showed significant weight gain. It is also notable that the only
procedures that appeared to benefit the infants were the methods that provide
moderate pressure to the whole body and the method that provided moderate
pressure to acupressure points on several parts of the body. As suggested by Field,4
the provision of moderate pressure, which appears to show superior results to
light pressure,5 may be stimulating an increase in vagal activity, which, in turn,
could be activating food absorption hormones. (Please also see Uvnäs-Moberg in
this volume.)
Ref ere n c e s
of tactile stimulation as a form of early intervention: a quantitative evaluation. J Dev Behav
Pediatr. 1987;8:68-76.
3. Kuhn CM, Schanberg SM, Field T, et al. Tactile-kinesthetic stimulation effects on sympathet-
ic and adrenocortical function in preterm infants. J Pediatr. 1991;119:434-440.
4. Field T. Massage therapy facilitates weight gain in preterm infants. Psychol Sci. 2001;10:51-
54.
5. Field T, Hernandez-Reif M, Diego M, Feijo L, Vera Y, Gil K. Massage therapy by parents
improves early growth and development. IBAD. In press.
CHAPTER 10:
O P T I M I Z I N G G ROW T H
AND BONE MASS IN
P R E M AT U R E I N FA N T S :
ARE DIET AND PHYSICAL
MOVEMENT THE ANSWERS?
Laurie J. Moyer-Mileur, PhD, RD, CCD

Department of Pediatrics, University of Utah
Ab s t ra c t
Premature infants are at increased risk for developing osteoporosis and

subsequent fractures. This chapter describes the results of clinical studies that
assessed growth and bone mass in premature infants subjected to daily physical
movement performed by therapists or mothers, as well as the effects that diet
had on these parameters. Results suggest that daily physical movement of these
infants increases bone growth and development in very-low-birth-weight (VLBW)
infants. Routine care of these infants, which typically involves limitation of
movement, appears to decrease the rate of bone formation in premature,
VLBW infants. In infants who have low calcium intake, physical movement
may decrease bone accretion during periods of rapid bone growth.
Studies performed by my
colleagues and I have
examined the effects
of daily physical
movement (PM)
administered by
occupational therapists
(OT) and infants’
mothers (MT) on
growth and bone mass
in premature infants.
Introd uc t i o n
Premature infants (<37 weeks gestation), especially those who have gestational ages
(GAs) less than 33 weeks and/or birth weights of 1500 g or less, present numerous
physiological and developmental concerns, including those involving physical
growth and the provision of adequate nutrients to sustain growth. The ex-utero
growth rate in such infants often lags behind normal in-utero growth, especially in
very-low-birth-weight (VLBW) premature infants (<1000 g), despite the provision
of energy and nutrients at currently advised levels.1
Premature infants are also at risk of developing osteopenia (low bone mass) and
subsequent fractures due to the following: limited accretion of bone minerals
in-utero; a greater need for bone nutrients than infants who are delivered at
term gestation; and decreased ex-utero calcium retention.1-5 The provision of
bone-related nutrients at advised levels increases bone mineralization; however,
ex-utero mineralization rates do not equal in-utero rates, even with gains in normal
body weight.1,3 In addition, bone mineralization in premature infants does not
approach normal ranges until after the first year of life,6 and it may continue to
be inadequate into childhood, further increasing the child’s risk of fractures.7
Chapter 10: Optimizing Growth and Bone Mass in Premature Infants 169
Several interrelated extrinsic factors have been shown to influence ex-utero growth
and bone mineralization in premature infants, including the following: energy
and nutrient requirements to support the infant’s somatic growth rate; the type
and quality of dietary and nutrient intake; medications that may alter metabolism
or impair renal retention of calcium, such as glucocorticoids, furosemide diuretics
and thyroid-replacement preparations; and the multitude of morbidities commonly
associated with prematurity (Figure 1).8
Fig 1. Extrinsic factors that influence growth and bone mineralization in premature
infants.
Ex-Utero
Preterm Birth Growth & Bone Preterm Birth
Limited In-Utero Mineralization Ex-Utero
Lean & Fat Mass
in Preterm Nutrient Needs
& Mineral Accretion
Infants
Nutrient Intake
Source
Parenteral v. Enteral
Bioavailability
Related Morbidities
Physical Stress,Morbidities
Related Sepsis
Movement Malabsorption
Chronic lung disease
Medications
Diuretics
Medications
Glucocorticoids
Anti-convulsants
Thyroid replacement
Mechanical strain on bones and joints stimulates bone formation and growth.9
Osteoblasts, the cells responsible for bone formation, increase their activity in
response to mechanical strain in vitro.10 Weight-bearing physical activity has been
demonstrated to increase bone mass in children, young adults and older persons.11-13
Conversely, the absence of weight-bearing physical activity, such as that observed
in immobilized adults or astronauts during prolonged spaceflights, increases
bone resorption and renal excretion of calcium and decreases bone mass.14
Hypomineralized bone, such as that reported in newborn term infants who have
neuromuscular diseases, has been attributed to decreased movement in utero.16
Since standard care for hospitalized premature infants includes swaddling or
“nesting” and decreased sensory and physical stimulation,15 these infants receive
limited physical movement, which may increase bone resorption and
demineralization even when adequate dietary bone nutrients are provided.
Desi g n a n d Me t h o d s
My colleagues and I have conducted several clinical trials to evaluate whether a

program of physical movement would promote greater gains in bone mineralization
in premature infants17,18 (also Moyer-Mileur, Ball, Brunstetter & Chan, unpublished
data, 2003). The infants enrolled in these studies were healthy premature infants
recruited from the newborn intensive care unit at University Hospital, Salt Lake
City, Utah. They were eligible under the following criteria: GA of 26 weeks to
32 weeks; appropriate body size; ability to feedings of at least 110 kcal/kg body
weight/day; an absence of medications other than appropriate vitamin supplements;
and informed parental consent. The University of Utah Institutional Review Board
approved these studies.
The infants were matched by birth weight and GA and were fed either mother’s
milk with powdered fortification (Enfamil® Human Milk Fortifier; Mead Johnson®
Nutritionals, Evansville, Ind, USA) or premature infant formula (Enfamil®
Premature LIPIL®; Mead Johnson® Nutritionals) at an equal caloric density of
24 kcal/oz.
In the first 2 studies,17,18 infants were assigned randomly to either physical

movement (PM) or control groups. In the third study (Moyer-Mileur, Ball,
Brunstetter & Chan, unpublished data, 2003), infants were assigned randomly
to PM provided by either an occupational therapist (OT) or the infant’s mother
(MT). The PM consisted of range-of-motion exercises with gentle compression
and extension and flexing of both upper and lower extremities. Each movement
was done 5 times each at the wrist, elbow, shoulder, ankle, knee and hip 5 times
per week for 4 weeks. (Figure 2). Since tactile stimulation may have influenced
growth and development (GD), control subjects were given a daily interactive
period of holding and stroking, but no PM activities. Daily body weights and
lengths and head circumferences, as well as feeding intakes and outputs, were
recorded. Serum and urine samples were collected and analyzed at baseline and
study completion to determine markers of growth and bone turnover. A summary
description of the 3 clinical studies is provided in Table 1.
Figure 2A (left)
demonstrates extension
of infant’s arm by
occupational therapist
(OT) against infant’s
own resistance to the
movement.
Figure 2B (right)
shows extension
and stabilization of
infant’s leg as gentle
decompression is
applied by OT to ankle,
knee and hip joints.
Table 1. Summary of Clinical Trials of Physical Movement in Premature Infants17,18
MOVEMENT PROTOCOL MEASUREMENTS
Study 117 Randomized to physical movement • Daily body weight and nutrient intake
(PM; n=13) or tactile stimulation and output
(n=13) • Weekly body length and occipitofrontal
circumference (OFC)
• Baseline and at 4 weeks: bone mass of
distal one-third radius by single-photon
absorptiometry (SPA), and blood and
urine measures of bone mineral status
Study 218 Randomized to PM (n=16) or • Daily body weight and nutrient intake
tactile stimulation (n=16) and output
• Weekly body length and OFC
• Baseline and at 2-kg body weight: bone
mass of forearm by dual-energy X-ray
absorptiometry (DXA) and blood and
urine measures of bone mineral status
Study 3 Randomized to PM provided by • Daily body weight and nutrient intake

occupational therapist (OT; n=16) and output
or the infant’s mother (MT; n=16) • Weekly body length and OFC
• Baseline and at 2-kg body weight: bone
mass of forearm by DXA and blood and
(LJ Moyer-Mileur, SD Ball, VL Brunstetter, GM Chan,
urine measures for growth factors and
unpublished data, 2003) markers of bone mineral status
Results
There were no differences in mean baseline body weights at study entry or at study
completion among any of the infants in the 3 clinical studies (Table 2)17,18 (also
Moyer-Mileur, Ball, Brunstetter & Chan, unpublished data, 2003). The rates of
change for body length and head circumference were similar and indicated an
overall positive trend for all infants. Despite equal energy and nutrient intakes,
infants who received PM in both Study 117 and Study 218 had greater increases in
average daily weight than did the control infants. In Study 3 (Moyer-Mileur, Ball,
Brunstetter & Chan, unpublished data, 2003), infants who received daily PM by
the OT experienced greater rates of daily weight gain than did the control infants,
Table 2. Subject Characteristics and Nutrient Intake During Clinical Trials

of Physical Movement in Premature Infants17,18 (also LJ Moyer-Mileur, SD Ball,
VL Brunstetter, GM Chan, unpublished data, 2003)
STUDY 1 STUDY 2 STUDY 3
13
16
16
n=
n=
n=
6
13
16
16
1
rol
rol
rol
n=
n=
n=
n=
nt
nt
nt
T
PM
PM
OT
Co
Co
Co
M
Gestation (wk) 28.2 (1.3) 28.9 (1.5) 29.6 (1.6) 29.8 (1.5) 28.4 (1.0) 28.9 (1.5) 29.8 (1.5)
Birth weight (g) 1207 (172) 1240 (182) 1258 (230) 1283 (199) 1188 (276) 1252 (232) 1283 (199)
Gender 7M/6F 7M/6F 7M/8F 10M/6F 10M/6F 6M/10F 10M/6F
Adjusted age 30.0 (1.4) 31.3 (0.9) 31.9 (1.5) 31.9 (1.1) 31.6 (0.8) 32.3 (1.4) 31.9 (1.1)
study: Day 1 (wk)
Days on study 28 28 27 24 29 27 24
Feeding 10HMF/3PTF 9HMF/4PTF 8HMF/7PTF 9HMF/6PTF 6HMF/10PTF 16HMF/0PTF 9HMF/6PTF
Energy (kcal/kg/d) 128 (6) 126 (8) 115 (12) 120 (7) 124 (10) 123 (11) 120 (7)
Protein (g/d) 3.4 (0.2) 3.4 (0.3) 3.3 (0.3) 3.5 (0.3) 3.7 (0.3) 3.7 (0.3) 3.5 (0.3)
Calcium (mg/d) 135 (7) 133 (7) 167 (17) 174 (18) 173 (15) 172 (15) 174 (18)
Phosphorus (mg/d) 68 (2) 67 (2) 84 (7) 88 (7) 88 (7) 87 (7) 88 (7)
Vitamin D (IU/d) 380 (15) 377 (15) 312 (14) 323 (24) 368 (14) 367 (13) 323 (24)
Values presented as the mean (±SD). HMF = Human milk with fortifier. PTF = Premature infant formula fed
at 24 kcal/oz. There are no differences in the infants’ characteristics or nutrient intake within or between the
3 studies. In Study 3, however, all infants randomized to the MT group were fed HMF.
while no significant differences were found between OT and MT infants or

between MT infants and controls (Figure 3).
In addition to greater weight gains found in infants who received PM, significant
differences in forearm bone mass were found between Study group 1 and Study
group 2 (Figures 4, 5 and 6). At baseline, there were no between-group differences
in distal one-third radius or total forearm bone mineral content (BMC; mg), bone
area (BA; cm2) or bone mineral density (BMD; mg/cm2). After approximately 28
days of intervention, however, my colleagues and I observed significant gains over
Figure 3. Despite similar energy and nutrient intakes, rate of growth (g/kg body wt/d)
significantly greater in infants who received daily OT-administered PM vs those
randomized to control group (P≤.01, MANCOVA). No differences in growth rates
between maternal administered (MT) PM and OT or controls. (Moyer-Mileur, Ball,
Brunstetter & Chan, unpublished data, 2003).
Figure 4. Bone mineral content (BMC; g) change (%) in distal one-third radius,
as measured by single-photon absorptiometry (SPA; Study 117), and total forearm,
as measured by dual-energy X-ray absorptiometry (DXA; Studies 218 and 3[LJMM,
unpublished data, 2003]), significantly greater in infants who received daily PM administered
by OT or MT versus control infants (P≤.02, MANCOVA).

Figure 5. Bone area (BA; cm2) change (%) in distal one-third radius, as measured
by SPA; Study 117), and total forearm, as measaured by DXA; Studies 218 and 3.(LJMM,
unpublished data, 2003) Percent changes in BA were significant only for OT and MT infants
versus control infants in Study 3 (P≤.03, MANCOVA).
Figure 6. Bone mineral density (BMD; mg/cm2) change (%) in distal one-third
radius, as measured by SPA; Study 117, and total forearm, as measured by DXA;
Studies 218 and 3(LJMM, unpublished data, 2003) significantly greater in infants who received
daily PM administered by an OT or the MT versus control infants (P≤.03,
MANCOVA).
time for BMC (P<.02; Figure 4) and BMD. The type of feeding source—fortified
human milk or premature infant formula—was not found to be a significant
cofactor for changes in bone mass over time.
In Study 3 (Moyer-Mileur, Ball, Brunstetter & Chan, unpublished data, 2003),

the gains in total forearm BMC, BA and BMD were similar for infants in the OT
and MT groups but greater than for control infants (P≤.03) (Figures 4, 5 and 6).
Although 100% of infants in the MT group received fortified human milk, versus
38% of the infants in the OT group, the feeding source was not found to be a
significant cofactor in regard to growth or bone mineralization.
Serum and urine markers of bone activity did not differ between groups in Study 1.
However, levels of serum parathyroid hormone (PTH), responsible for regulation
of serum calcium homeostasis, were found to be related inversely to BMC at study
completion (r=–.83, P=.01). In Study 2, PTH levels were greater in infants
receiving PM (P=.03), although the change from study entry to study completion
did not differ statistically and were within normal limits for age. In Study 3, my
colleagues and I found that levels of bone alkaline phosphatase (BAP), a marker
of bone formation, increased approximately 15% from baseline in MT and OT
infants, but decreased approximately 20% from baseline in controls (P<.05). Urine
calcium to creatinine excretion was similar and within normal limits for all groups.
Discussion
My colleagues and I conducted the first longitudinal, randomized studies to

describe the effect of daily PM on growth and bone mass in hospitalized, premature
infants17,18 (also Moyer-Mileur, Ball, Brunstetter & Chan, unpublished data, 2003).
Previous longitudinal studies, using single-photon absorptiometry (SPA) of the

distal radius, confirmed the need for higher intakes of dietary calcium, phosphorus
and vitamin D to improve bone mineralization in hospitalized, premature
infants.19,20 My coinvestigators and I have also used SPA to assess bone
mineralization in premature infants: We found a greater response when a program
of daily PM was provided in addition to an adequate intake of bone nutrients.17
The use of dual-energy X-ray absorptiometry (DXA) in the latter 2 studies allowed
for bedside evaluation of a daily program of PM for site-specific bone

mineralization in premature infants18 (also Moyer-Mileur, Ball, Brunstetter &
Chan, unpublished data, 2003). Results of these 3 studies provide support for the
premise that a program of daily PM enhances bone GD in premature, VLBW
infants17,18 (also Moyer-Mileur, Ball, Brunstetter & Chan, unpublished data, 2003).
Although all indices of bone mass increased for all groups, infants receiving PM
had consistently greater gains in BMC and BMD. This was true whether an OT
or the infant’s mother administered the PM.
Bone growth and development is a dynamic process initiated by an increase in

bone surface area or matrix and completed by the binding of calcium salts into
hydroxyapatite, which fills in to mineralize the bone matrix. Once the bone matrix
is formed, an additional 10 days is required for mineralization.22 An investigator
would expect to see, therefore, an increase in BA before an increase in BMC or BMD
during periods of rapid growth. My colleagues and I suggest that PM maintains
bone GD, while routine care, which limits movement, slows the rate of bone
formation in premature, VLBW infants. The rise in activity of PTH in infants who
received PM in Study 218 may have contributed to bone growth and mineralization,
since PTH levels increase following physical activity in adults,23 while therapeutic
administration of PTH has been noted to stimulate bone formation.24 Increases in
the activity of BAP associated with PM in Study 3 (Moyer-Mileur, Ball, Brunstetter
& Chan, unpublished data, 2003) also indicate enhanced bone formation. The
data found by myself and my colleagues support the positive effect that a daily
program of PM has on forearm growth and mineralization.
Infants in these studies17,18 (also Moyer-Mileur, Ball, Brunstetter & Chan,

unpublished data, 2003) were fed energy and nutrient intakes at recommended
levels.1 Greater rates of weight gain during the studies, despite similar energy and
nutrient intakes between groups, was an unexpected finding. Initial weight loss
after birth and the rate of weight gain from birth to study entry did not differ
between groups. Weight gain accelerated in infants who received OT-administered
PM versus control infants in all 3 trials, although energy and nutrient intakes did
not diminish or differ between groups. Current work by my colleagues and myself
and others to examine the interaction between PM and growth hormones and
mediators and their effect on growth in premature infants may provide answers
to this unexpected outcome.
In 1985, the American Academy of Pediatrics stated that the nutritional goal for
premature infants is to provide optimal nutrition to support growth equivalent to
in-utero gain during the third trimester.25 Nutritional intervention, while helping
to promote adequate weight gain, offers variable effects on postnatal bone
mineralization in premature infants. Multiple factors influence bone growth and
development including diet and PM. Although we have demonstrated the positive
effects of PM on bone mineralization in premature infants, a program of PM may
prove inappropriate for infants who have poor nutrient intake. Specker, Mulligan
and Ho26 reported on the effects of a 1-year program of PM on BMC in older
infants. They found that BMC is related to calcium intake. Randomized PM and
control infants who had moderately high calcium intakes had similar BMCs, but
low calcium intakes in infants in the PM program resulted in lower BMCs. These
investigators speculated that participation in a program of PM during rapid bone
growth may lead to reduced bone accretion in the presence of a low intake of
calcium. Thus, caution should be used before starting a program of PM in infants
who have limited nutrient intake; for example, premature infants fed unfortified
mother’s milk or standard infant formula, or infants who have a chronic illness or
are taking medication that interferes with the delivery of bone nutrients.
Conclusion
In the 3 studies reviewed, my colleagues and I have shown successfully that bone
mass indices increase when premature VLBW infants receive daily PM in addition
to recommended levels of energy and nutrients. More importantly, we have
demonstrated that a program of PM administered by the infant’s own mother
is equally as effective as a therapist-administered program of PM in promoting
greater bone mineralization in premature infants.
Re f e re n c e s
1. Tsang RC, Lucas A, Uauy R, Zlotkin S, eds. Nutritional Needs for Preterm Infants: Scientific
Basis and Practical Guidelines. Calcium, magnesium, phophorus, and vitamin D. Baltimore,
Md: Williams & Williams; 1993:1-10.
2. Koo WWK, Sherman R, Succop P, Ho M, Buckley D, Tsang RC. Serum vitamin D

metabolites in very low birth weight infants with and without rickets and fractures. The
Journal of Pediatrics. 1989;114:1017-1022.
3. Tsang RC, Lucas A, Uauy R, Zlotkin S, eds. Nutritional Needs for Preterm Infants: Scientific
Basis and Practical Guidelines. Baltimore, Md: Williams & Williams; 1993:134-141.
4. Widdowson EM, Southgate DAT, Hay E. Fetal growth and body composition. In: Landblad
BS, ed. Perinatal Nutrition. New York, NY: Academic Press; 1988:4-14.
5. Koo WWK, Massom LR, Walters J. Validation of accuracy and precision of dual energy X-ray
absorptiometry for infants. Journal of Bone and Mineral Research: the Official Journal of the
American Society for Bone and Mineral Research. 1995;10:1111-1115.
6. Rigo J, Nyamugabo K, Picaud JC, Gerard P, Pieltain C, De Curtis M. Reference values of
body composition obtained by dual energy X-ray absorptiometry in preterm and term
neonates. Journal of Pediatric Gastroenterology and Nutrition. 1998;27:184-190.
7. Abrams SA, Schanler RJ, Tsang RC, Garza C. Bone mineralization in former very low birth
weight infants fed either human milk or commercial formula: one-year follow-up observation.
The Journal of Pediatrics. 1989;114:1041-1044.
8. Krug SK. Osteopenia of prematurity. In: Groh-Wargo S, Thompson M, Hovasi Cox J, eds.
Nutritional Care for High-Risk Newborns. 3rd ed. Chicago, Ill: Precept Press; 2002:489-506.
9. Yeh JK, Liu CC, Aloia JF. Effects of exercise and immobilization on bone formation and
resorption in young rats. The American Journal of Physiology. 1993;264:E182-E189.
10. Schulthies L. The mechanical control system of bone in weightless spaceflight and in aging.
Experimental Gerontology. 1991;26:203-214.
11. Pirnay F, Bodeux M, Crielaard JM, Franchimont P. Bone mineral content and physical
activity. International Journal of Sports Medicine. 1987;8:331-335.
12. Pocock NA, Eisman JA, Yeates MG, Sambrook PN, Eberl S. Physical fitness is a major
determinant of femoral neck and lumbar spine bone mineral density. The Journal of Clinical
Investigation. 1986;78:618-621.
13. Davee AM, Rosen CJ, Adler RA. Exercise patterns and trabecular bone density in college
women. Journal of Bone and Mineral Research: the Official Journal of the American Society for
Bone and Mineral Research. 1990;5:245-250.
14. Mazess RB, Whedon GD. Immobilization and bone. Calcified Tissue International.
1983;35:265-267.
15. Lickliter R. The role of sensory stimulation in perinatal development: insights from
comparative research for care of the high-risk infant. Journal of Developmental and Behavioral
Pediatrics: JDBP. 2000;21:437-447.
16. Rodriguez JI, Garcia-Alix A, Palacios J, Paniagua R. Changes in the long bones due to fetal
immobility caused by neuromuscular disease. A radiographic and histological study. Journal
of Bone and Joint Surgery: American Volume. 1988; 70:1052-1060.
17. Moyer-Mileur L, Luetkemeier M, Boomer L, Chan GM. Effect of physical activity on bone
mineralization in premature infants. The Journal of Pediatrics. 1995;127:620-625.
18. Moyer-Mileur LJ, Brunstetter V, McNaught TP, Gill G, Chan GM. Daily physical activity
program increases bone mineralization and growth in preterm very low birth weight infants.
Pediatrics. 2000;106:1088-1092.
19. Chan GM, Mileur L, Hansen JW. Calcium and phosphorus requirements in bone
mineralization of preterm infants. The Journal of Pediatrics. 1988;113:225-229.
20. Greer FR, McCormick A. Bone growth with low bone mineral content in very low birth
weight premature infants. Pediatric Research. 1986;20:925-928.
21. Horsman A, Ryan SW, Congdon PJ, Truscott JG, Simpson M. Bone mineral accretion rate
and calcium intake in preterm infants. Archives of Disease in Childhood. 1989;64(spec no.
7):910-918.
22. Termine JD, Robey PG. Bone matrix proteins and the mineralization process. In: Favus MJ,
ed. Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism. Philadelphia,
Pa: Lippincott-Raven; 1996:24-28.
23. Ljunghall S, Joborn H, Roxin LE, Skarfors ET, Wide LE, Lithell HO. Increase in serum
parathyroid hormone levels after prolonged physical exercise. Medicine and Science in Sports
and Exercise. 1988;20:122-125.
24. Miller SC, Hunziker J, Mecham M, Wronski TJ. Intermittent parathyroid hormone
administration stimulates bone formation in the mandibles of aged ovariectomized rats.
Journal of Dental Research. 1997;76:1471-1476.
25. American Academy of Pediatrics Committee on Nutrition. Nutritional needs of low-birth-

weight infants. Pediatrics. 1985;75:976-986.
26. Specker BL, Mulligan L, Ho M. Longitudinal study of calcium intake, physical activity, and
bone mineral content in infants 6-18 months of age. Journal of Bone and Mineral Research:
the Official Journal of the American Society for Bone and Mineral Research. 1999;14:569-576.
S E C T I O N I V.
EFFECTS OF MASSAGE
O N S L E E P, R E L A X A T I O N
AND WELL-BEING
CHAPTER 11:
MASSAGE THERAPY
A N D S L E E P - WA K E R H Y T H M S
I N T H E N E O N AT E
Sari Goldstein Ferber, PhD

Department of Nursing
University of Haifa, Israel
Sackler School of Medicine
Tel Aviv University, Israel
Ab s t ra c t
The neonate spends most of the perinatal period sleeping. Whereas the
full-term neonate sleeps 70% of the 24-hour period, the preterm infant sleeps
up to 90% of a full day.1 Since self-regulation and growth occur during sleep,
enhancing sleep and assuring undisturbed sleep by the caregiver are of great
importance. In addition, one of the natural and immediate ways to comfort
an infant is by holding, touching, stroking and keeping the infant in close
physical contact with a caregiver. This chapter provides empirical evidence for
the benefits of touch by a caregiver on sleep. The infant’s sleep and adjustment
to the day-night cycle are described, and the effects of massage therapy are
illustrated by empirical data.
184 Section IV. Effects of Massage on Sleep, Relaxation and Well-Being
The neonate’s behavioral states have been considered an expression of

neurobehavioral organization.2 State regulation reflects the infant’s capacity to
control responses and to adjust to the new environment outside the womb. The
night-day, sleep-wake cycle develops before birth. As early as the third trimester
of pregnancy, fetal diurnal rhythms are already in tune with maternal day-night
rhythms.3,4 During the first weeks of life, social time cues gain importance, while
only at a later stage does light become a dominant environmental cue that entrains
the infant’s cycles to the 24-hour period.5-8 Recent studies of full-term Japanese
infants showed that the sleep-wake rhythm emerged approximately 1 month
after birth.9 In contrast, studies of European infants have shown that sleep-wake
rhythms are organized in infants much later, at approximately 3 months to
4 months of age.10,11
Neonates exhibit a greater amount of activity during the night than do older
infants.12-15 Therefore, it is not surprising that considerable maternal distress is
associated with infants’ disrupted night sleep.16 Consequently, practitioners have
sought methods to enhance the development of sleep-wake rhythms in infants.12,17
Mel a t o n i n a n d S o c i a l Cu e s
Melatonin, a naturally occurring hormone released at night and ceased by exposure

to light,18 acts as a sleep regulator and time cue.19 It also plays a role in visual,
cerebrovascular, reproductive, neuroendocrinological and neuroimmunological
functions.20 The melatonin rhythm develops during the first months of postnatal
life, from low basal levels at 6 weeks to 8 weeks to maximum capacity at 12
months.21
Although bright light is an important time cue, social cues are also important for
the entrainment of sleep-wake rhythms. Social cues—including social demands or
tasks, such as regularly timed meals and sleep-wake schedules—have been shown
to entrain sleep-wake rhythms.22
Support for the role of the caregiver in the infant’s modulation of states comes from
a study conducted by Thoman, Davis and Denenberg in which a group of full-term
infants were observed at home at 2 weeks to 5 weeks of age.23 Data were obtained
Chapter 11: Massage Therapy and Sleep-Wake Rhythms in the Neonate 185
in 2 social contexts: when the baby was alone and when the baby was with the
mother. Fewer state-modulated relationships were noted when the babies were with
their mothers than when they were alone. The investigators concluded that the
mother is probably acting as a social cue, influencing the infant’s adaptation to
the day-night environment.
When the infants were with their mothers they were awake most of the time.
Their waking states were more numerous and less sustained and the infants were
more susceptible to environmental influences than when they were in their sleeping
states.24 Thus, the mothers’ activities with their infants appeared to have more
effect on the infant during the daytime.
Massage Therapy Enh a nc e s t h e De ve lop me nt

o f Sle e p-Wa k e Cyc l e s i n Fu l l - Ter m Infants
We conducted a study on the effect of massage therapy on sleep-wake cycles in

full-term infants.24 The objective was to determine if mothers who provided
massage therapy as a bedtime routine could affect their infants’ sleep-wake cycles.
A motion-sensing device measured the rest-activity cycles of these infants before
and after 14 days of massage therapy and at 6 weeks and 8 weeks of age. Levels
of melatonin were assessed in urine samples at 6 weeks, 8 weeks and 12 weeks.
Mothers in the experimental group were instructed to prepare their babies for
sleep following the evening meal and bath, between 8:00 pm and 9:00 pm and to
provide 30 minutes daily massage therapy for their infants for 14 days, starting on
Day 10 of life. These mothers were instructed to relax, sit comfortably on a sofa
and hold their babies in their laps or lay them on the sofa next to them. Mothers
were then instructed to roll their infants on their sides and perform massage
therapy in a rhythm of 3 movements per second. This massage therapy involved
touching the infant’s head with one hand and lightly stroking its back in a circular
motion under the bedclothes with the other hand. Mothers in the control group
performed their usual bedtime activity, but without the massage therapy.
The massaged infants, versus the control infants, achieved more favorable
adjustments of their rest-activity cycles at the age of 8 weeks and a higher nocturnal
melatonin production at the age of 12 weeks. At 8 weeks, melatonin levels in

the control infants reached a peak of activity at about 12:00 midnight and again
at approximately 12:00 noon, whereas in the massaged infants, a major peak
was reached early in the morning, with a secondary peak in the late afternoon
(Figure 1). As indicated in Figure 2, nocturnal melatonin levels at 12 weeks
were significantly higher in the massaged infants (1346 nanog/night versus
823 nanog/night, respectively; P<.05). Total movement was not affected by
treatment. The shift in melatonin peaks towards the early-morning hours in the
massaged infants most likely represents an adjustment to the nocturnal period
at an age earlier than in the control infants. We concluded that massage therapy
given during infants’ bedtime routines could be a strong social cue that can affect
the development of sleep-wake cycles.
Figure 1. Rest-activity cycles during night and day comparing treatment and control
groups at age 8 weeks. Treated subjects displayed delayed primary and secondary peaks
of melatonin activity (P<.05).
Figure 2. Comparison of levels of 6-sulphatoxymelatonin secretion in treatment and

control subjects during the night at age 12 weeks. Levels in the treated subjects were
significantly higher (P<.05).
I wish to thank Professors N. Zisapel, A. Weller and P. Lavie and Dr. M. Laudon
for their assistance in preparing this chapter. Special thanks also to the Sleep
Laboratory Library Team at the Technion Institute, Haifa, Israel.
Ref ere n c e s
1. Hutt C, von Bernuth H, Lenard HG, Hutt SJ, Prechtl HF. Habituation in relation to state
in the human neonate. Nature. 1968;220:618-620.
2. Feldman R, Weller A, Sirota L, Eidelman AI. Skin-to-skin contact (Kangaroo care) promotes
self-regulation in premature infants: sleep-wake cyclicity, arousal modulation, and sustained
exploration. Developmental Psychology. 2002;38:194-207.
3. Brazelton TB. Saving the bathwater. Child Development. 1990;61:1661-1671.
4. Rivkees SA, Hao H. Developing circadian rhythmicity. Seminars in Perinatology.
2000;24:232-242.
5. Cavallo A. The pineal gland in human beings: relevance to pediatrics. The Journal of
Pediatrics. 1993;123:843-851.
6. Kennaway DJ, Stamp GE, Goble FC. Development of melatonin production in infants and
the impact of prematurity. The Journal of Clinical Endocrinology and Metabolism.
1992;75:367-369.
7. Lohr B, Siegmund R. Ultradian and circadian rhythms of sleep-wake and food-intake

behavior during early infancy. Chronobiology International. 1999;16:129-148.
8. Anders TF. Infant sleep, nighttime relationships, and attachment. Psychiatry. 1994;57:11-21.
9. Shimada M, Takahashi K, Segawa M, Higurashi M, Samejim M, Horiuchi K. Emerging
and entraining patterns of the sleep-wake rhythm in pre-term and term infants. Brain
Development. 1999;7:468-473.
10. Wolfson AR. Sleeping patterns of children and adolescents: developmental trends, disruptions
and adaptations. Child and Adolescent Psychiatric Clinics of North America. 1996;5:549-568.
11. Menna-Barreto L, Benedito-Silva AA, Marques N, de Andrade MM, Louzada F. Ultradian
components of the sleep-wake cycle in babies. Chronobiology International. 1993;10:
103-108.
12. Weissbluth M, Weissbluth L. Colic, sleep inertia, melatonin and circannual rhythms. Medical
Hypotheses. 1992;38:224-228.
13. McGraw K, Hoffmann R, Harker C, Herman JH. The development of circadian rhythms
in a human infant. Sleep. 1999;22:303-310.
14. Mirmiran M, Lunshof S. Perinatal development of human circadian rhythms. Progress in
Brain Research. 1996;111:217-226.
15. Sadeh A, Lavie P, Scher A, Tirosh E, Epstein R. Actigraphic home-monitoring sleep-disturbed
and control infants and young children: a new method for pediatric assessment of sleep-wake
patterns. Pediatrics. 1991;87:494-499.
16. Guilleminault C, Leger D, Pelayo R, Gould S, Hayes B, Miles L. Development of circadian
rhythmicity of temperature in full-term normal infants. Neurophysiologie Clinique - Clinical
Neurophysiology. 1996;26:21-28.
17. Nishihara K, Horiuchi S, Eto H, Uchida S. Mothers’ wakefulness at night in the post-partum
period is related to their infants’ circadian sleep-wake rhythm. Psychiatry and Clinical
Neurosciences. 2000;54:305-306.
18. Renfrew MJ, Lang S, Martin L, Woolridge M. Interventions for influencing sleep patterns
in exclusively breastfed infants. Cochrane Database of Systematic Reviews. 2000;No.
2:CD000113.
19. Sivan Y, Laudon M, Tauman R, Zisapel N. Melatonin production in healthy infants:

evidence for seasonal variations. Pediatric Research. 2001;49:63-68.
20. Sandyk R. Melatonin and maturation of REM sleep. The International Journal of
Neuroscience. 1992;63:105-114.
21. Masana MI, Dubocovich ML. Melatonin receptor signaling: finding the path through the
dark. Science’s STKE [Electronic Resource]: Signal Transduction Knowledge Environment.
November 6, 2001:PE39.
22. Kennaway DJ, Goble FC, Stamp GE. Factors influencing the development of melatonin
rhythmicity in humans. The Journal of Clinical Endocrinology and Metabolism. 1996;81:
1525-1532.
23. Ehlers CL, Frank E, Kupfer DJ. Social zeitgebers and biological rhythms. A unified approach
to understanding the etiology of depression. Archives of General Psychiatry. 1988;45:948-952.
24. Thoman EB, Davis DH, Denenberg VH. The sleeping and waking states of infants:
correlations across time and person. Physiology & Behavior. 1987;41:531-537.
25. Ferber SG, Laudon M, Kuint J, Weller A, Zisapel N. Massage therapy by mothers enhances
the adjustment of circadian rhythms to the nocturnal period in full-term infants. Journal of
Developmental and Behavioral Pediatrics: JDBP. 2002;23:410-415.
CHAPTER 12:
M A S S A G E , R E L A X AT I O N
AND WELL-BEING:
A POSSIBLE ROLE
FOR OXYTOCIN AS AN
I N T E G R AT I V E P R I N C I P L E ?
Kerstin Uvnäs-Moberg, MD
Swedish University of Agricultural Science
Department of Animal Physiology
Karolinska Institute, Stockholm
Abstract
Massage has been shown to decrease cortisol and anxiety levels. Reduced anxiety
is accompanied by an increased sense of well-being. This chapter examines the
effects of massage on promoting relaxation, on enhanced sense of well-being and
positive social interaction. Oxytocin—a pituitary hormone that stimulates the
contraction of uterine muscle and the secretion of breast milk—may play a pivotal
role in producing these responses to massage. Studies are reviewed that show
increased oxytocin levels in animals, as well as in humans, in response to physical
contact between mother and child, such as skin-to-skin contact during
breastfeeding. Research has also exhibited decreased gastrin levels in response
to massage, suggesting a possible marker of increased social interests and skills.
Introd uc t i o n
Massage has been used since ancient times to relax muscles, cure disease and
promote general well-being. In spite of this, few studies exist that document the
positive effects of massage in a controlled scientific manner. In addition, there
are only a few studies that attempt to delineate the mechanisms underlying the
beneficial effects of massage. Field and colleagues,1 who represent an exception
in this regard, have performed extensive clinical studies on the effects of massage,
ranging from studies on premature infants to the effects on patients who have
various psychiatric disorders (also see chapter by Field et al, this volume). Most
of these studies have demonstrated that massage offers undeniable beneficial effects.
In addition to demonstrating the alleviation of specific symptoms, they have
typically found decreased cortisol levels. Anxiety levels have also been reduced,
and the resulting sense of well-being has increased.1-3 Massage has also been noted
to increase growth hormone, with positive effects on cellular growth.4
One of the aims of this chapter is not to summarize the clinical effects of massage,
but rather to suggest that massage, in part, exerts its beneficial effects by activating
a psychophysiological system that acts in opposition to the “fight-or-flight” system
and which therefore promotes relaxation, well-being and positive social interaction.
Oxytocin is suggested to be a major organizing principle of these effects at the
hypothalamic level.
The “Fight-or-Flight” Response
The “fight-or-flight” response, described by Cannon in 1932, is induced by acute

stress, which may be of psychological (perception of fear or threat) or physiological
(response to pain or trauma to the body) origin.5 This response is characterized
by behavioral arousal, anxiety, defensive aggression and a physiological pattern
consistent with fighting or fleeing. The corresponding neuroendocrine pattern
involves activation of the hypophyseal-pituitary-adrenal (HPA) axis and the
sympathetic nervous system (SNS). Hypothalamic corticotropin-releasing factor
and vasopressin, together with the brain-stem noradrenaline (NA) system, play
important regulatory roles.
Chapter 12: Massage, Relaxation and Well-Being: A Possible Role for Oxytocin 193
The hypothesis presented here is that there is an analogous unified pattern

characterized by well-being, calm and positive social interactions. The corresponding
physiological pattern consists of the relaxation of muscles, decreased cortisol levels
and cardiovascular (CV) activity, and enhanced activity in the gastrointestinal (GI)
tract that promotes digestion. This pattern, which will be referred to as “relaxation
and well-being,” can be triggered by environmental and psychological calm as
well as by calming physiological stimuli, such as touch and warmth. The vagal
parasympathetic nervous system (PNS) (the “physiological slowing system”) is
activated and sympathoadrenal activity (the “fight or flight system”) is shut down.
At a central level, hypothalamic oxytocin plays an important integrative role.6-8
In contrast to the rapidity with which the fight-or-flight response develops, the
state of relaxation and well-being develops slowly. The subjective signs of the latter
are subtle and sometimes more easily defined by their absence. Yet, relaxation and
well-being have distinct psychophysiological patterns, and their expressions can be
demonstrated experimentally in several behavioral and physiological model systems.
Pulse rate and blood pressure (BP) are kept at a low, healthy and balanced level,
and the GI tract, which is controlled by the vagus nerve, is activated, promoting
digestion and the storing of nutrients. Energy would rather be used for growth and
restorative processes than for muscular or thermogenic activity. Behaviorally,
reduced arousal and the development of calm prevail. Subjectively, this might be
related to a sense of well-being and relaxation. This state should not, however, be
confused with euphoria, which is a more intense feeling of joy and reward.6-8
Guido Reni’s painting of Madonna and Child (Figure 1) captures some of the
most significant features of the state of relaxation and well-being. The painting
illustrates that this state contains both an individual and an interactive component:
The Madonna appears relaxed, calm, content, happy, peaceful, warm, open,
generous, empathic and friendly. In the interaction with the child, she displays
closeness, trust, loyalty, giving and receiving, as well as love. Individual boundaries
are erased, and a sense of unity prevails. The Madonna and Child depicts just one
example of this state of relaxation and well-being, which can be experienced in a
variety of situations, independent of gender and age. Such situations range from
the well-being and physiological relaxation induced by a hot bath, by being part
of a social group or by receiving massage.
Figure 1. The Madonna with child as an expression of the individual and the
interactional aspects of eustasis (Madonna and Child by Guido Reni, 1575-1642).
relaxed closeness
calm trust
content loyalty
happy giving
peaceful receiving
warm love
open unity
generous
empathic
friendly
O x y t o c i n a s a n Age nt Be h i nd Re l axation
and Well-Be i n g : t h e Ev i d e nc e
Since we are suggesting that oxytocin is an important integrating factor in the

mediation of relaxation and well-being, oxytocin, as well as its physiological effects,
will be described in greater detail. Oxytocin is produced in the paraventricular
nucleus (PVN) and in the supraoptic nulceus (SON) of the hypothalamus.
Magnocellular oxytocinergic neurons in these nuclei project to the posterior
pituitary, from which oxytocin can be secreted to exert its well-known hormonal
effects during labor and lactation. Parvocellular oxytocinergic neurons of the PVN,
on the other hand, ramify within the brain to reach limbic, medullary and spinal
areas: For example, oxytocinergic fibers reach the amygdala, the nucleus tractus
solitarius (NTS), the vagal motor nucleus (DMX), the locus ceruleus (LC) and
the raphe nuclei of the brain stem.9 Oxytocin, therefore, acts as both a hormone
and a neuropeptide, and the spectrum of effects of oxytocin is much broader than
thought previously10: For example, it has been shown that oxytocin stimulates
various interactive behaviors, such as maternal, sexual and other social behaviors.11,12
Oxytocin may also produce anxiolytic-like or sedative effects and increase the pain
threshold. It also influences CV and GI functions.13-18 In response to repeated
administrations, oxytocin induces long-lasting antistress and growth-promoting

effects. In several studies, 5 daily injections of oxytocin induced effects lasting for
up to 3 weeks, including anxiolytic-like actions, the lowering of BP, the elevation of
pain thresholds and a decrease in plasma corticosterone levels.19-22 Learning deficits
due to high stress levels are improved markedly by pretreatment with oxytocin. It
has also been shown to possess antidepressant-like properties in animal models.23-25
Lastly, this treatment regimen stimulates digestion and anabolic processes, as well
as weight gain, growth and healing.26,27
Results from neurophysiological and pharmacological experiments as well as results

from autoradiography, suggest that the long-lasting effects of oxytocin are related
to adaptive changes in central neurotransmitter systems.20 Increased opioid-like
activity is the basis behind the prolonged elevation of pain thresholds.28 An
enhanced α2-adrenoceptor function, for example, in the amygdala, hypothalamus,
LC and NTS, is related to the long-term antistress-like effects of oxytocin.20,28-30
The NA system, emanating in the LC, which is related strongly to mechanisms
of arousal, is inhibited by α2-adrenoceptor activation, as is the function in the
SNS, whereas parasympathetic tone is increased.31 Consequently, treatment with
oxytocin results in an attenuation of arousal and stress levels. In addition, energy
conservation will be stimulated.
For the purposes of this chapter, it is important to note that oxytocin is released
not only in response to suckling during breastfeeding and in response to labor,
but it is also released by nonnoxious stimulation—such as by touch, warmth and
stroking—applied to other parts of the body.14,18,32 Levels of oxytocin rise in plasma
as well as in cerebrospinal fluid (CSF) in response to these stimuli. A release of
oxytocin in the amygdala has been demonstrated after suckling in sheep33 and
rats,34 which suggests that oxytocin is released from nerve terminals in the specific
brain regions receiving oxytocinergic nerve projections.
Physiological Mechanisms in the Induction of

Re la x a t i o n a nd We l l - Be i ng i n Re sp onse to Ph ysical
C o n t a c t Be t we e n Mo t h e r a nd C h ild
In their most basic forms, relaxation and well-being are induced by warmth, touch
and light pressure. This may occur in noninteractive situations, such as during
ingestion of food and exposure to warmth, as well as in interactive situations,

such as during closeness, breastfeeding or sexual interactions.35 The sensory fibers
involved include thick myelinated fibers of the Aβ type as well as a subpopulation
of slowly conducting C-fibers.36 In addition, some cutaneous afferents originating
from the ventral side bypass the spinal cord to reach the vagal sensory area of
the NTS via the nodose ganglion.37 This latter type of sensory fiber should be of
particular importance for the induction of well-being and relaxation in situations
of close individual physical contact. It should be noted that the effects discussed
here are not related to the cortical somatotopic representation of sensory afferents,
but rather to effects exerted at limbic or lower brainstem levels.
To elaborate the processes by which touch, warmth and stroking and oxytocin may
take part in inducing relaxation and well-being, their roles in the mother-infant
interaction are described. This is a situation during which mother and child, via
close physical contact, induce relaxation and well-being in each other. When the
mother and child are allowed skin-to-skin contact after birth, the newborn baby
exhibits a spontaneous breast-seeking behavior. Before he/she starts suckling, the
newborn infant massages the mother’s breasts with his/her hands.38 During this
period, a pulsatile release of oxytocin is induced in the mother. Interestingly, there
is a strong positive correlation between the maternal rise of oxytocin levels and the
number of infant hand movements, suggesting that it is the massage that induces
the release of oxytocin.39,40 Circulating oxytocin released from the mother’s posterior
pituitary during this period not only causes milk ejection, but it also induces
vasodilation in the skin of the mother’s chest.41 The mother experiences this
vasodilation as warmth, which enhances her sense of relaxation and well-being
and her interactions with her child. After birth, the child, in turn, responds to the
physical contact with the mother by becoming calmer, as expressed by less crying.42
The child also becomes more relaxed physically, which is evidenced by a higher foot
temperature as a result of decreased peripheral vasoconstriction.43 Interestingly,
there is a strong positive relationship between maternal breast temperature and the
baby’s foot temperature: The warmer the mother’s breasts, the warmer the baby’s
feet, which suggests that maternal temperature influences the sympathetic nervous
tone of the baby. In this way, the mother and child in our example are interlocked
in a feed-forward process of well-being and relaxation.
At a purely mechanistic level, simple hormonal neurogenic reflexes, as well as

more-complex central mechanisms, take part in the mediation and expression of
well-being and relaxation when induced by ventral physical contact between the
mother and the newborn. In addition to the ordinary sensory fibers that mediate the
senses of touch, light pressure and warmth, the cutaneous vagal afferents, which
project to the NTS, are activated. At the level of the NTS, sensory stimuli induce a
physiological relaxing effect by decreasing sympathetic nerve activity and increasing
parasympathetic tone. This results in a lowered BP, increased peripheral circulation
and activation of the endocrine system of the GI tract. In a second step, NA
pathways that project from the NTS to the hypothalamus are activated. As a
consequence, the HPA axis is inhibited, as evidenced by a decrease in plasma cortisol
levels. In addition, the secretion of oxytocin is stimulated not only from the pituitary
but also from the nerves that project to different areas of the brain. Oxytocin
has been shown to exert anxiolytic-like and sedative effects by way of central
mechanisms, in part by actions in the amygdala and in the LC. This may, therefore,
mediate the calming effect of physical contact. It also elevates the pain threshold by
an action in the periaqueductal gray and in the spinal cord (Figures 2A, 2B and 2C).
Figure 2A. Schematic illustration of nervous system connections between the skin and
the brain: A) vagal afferents.
NA = noradrenaline; NTS = nucleus tractus solitarius

Figures 2B and 2C. Schematic illustrations of nervous system connections between

the skin and the brain: B) consequences of activation of NA mechanisms in the NTS;
and C) consequences of activation of oxytocinergic neurons originating in the
hypothalamus.
NA = noradrenaline; NTS = nucleus tractus solitarius

Each breastfeeding session has been shown to be related to an induction of calm,

decreases of BP and cortisol levels and to an influence of the levels of GI hormones
in the mother.8 In the newborn, suckling is accompanied by a release of GI
hormones, such as insulin, and by the expressions of mental calm and increased
pain thresholds.44 Interestingly, premature babies who are placed in skin-to-skin
contact with their mothers have not only exhibited lower levels of cortisol but
have also shown lower levels of the vagally controlled GI hormones gastrin and
cholecystokinin (CCK), thereby suggesting that vagal nerve activity is influenced
by ventral physical contact.45 The implications of this finding are discussed in more
detail later in this chapter.
Long-Ter m Effects of Ea rl y Sk i n-to-Sk in C ontact

an d Bre a s t f e e d i ng
Many different sources of information have shown that early skin-to-skin contact
between mothers and infants offers long-term benefits.46 If allowed skin-to-skin
contact after birth, mothers interact more frequently with their babies and have
more milk and lower gastrin levels 4 days later.47 In addition, they present a more
pulsatile pattern of oxytocin. The number of pulses is related to the sociability of
the mothers and how much milk they give their babies.39,48 In addition, continued
breastfeeding is known to be related to sustained beneficial psychological and
physiological effects.8 According to personality trait scales, these mothers are more
calm in the sense that they prefer a calm and regular lifestyle.8 They have also
proved to be more interested in social interactions,49,50 have lower BPs and react
with lesser levels of cortisol in response to physical stress.51
Taken together, these data indicate that mothers who are breastfeeding and who
are exposed to repeated releases of oxytocin develop a psychophysiological pattern
characterized by increased social competence, as well as by relaxation and calm—
exactly the effect pattern caused by repeated administrations of oxytocin in rats.15
Although not yet described in the literature, it is possible that endogenous
“opioidergic,” as well as α2-receptor, function has been increased as a result of
repeated treatments with oxytocin. In support of a role for oxytocin in creating
long-term antistress effects in mothers who are breastfeeding are the findings of
positive correlations between plasma oxytocin levels and calm personality traits,
socialization and social dependency in individual mothers.48-50
Effect s o f Ma s s a g e : A Po s s i b l e Ro l e for O x ytocin?
Clinical data suggest that massage given to humans produces many of the
physiological signs characteristic of relaxation and well-being, as seen in mothers
who are breastfeeding.8 Furthermore, experimental data obtained from rats
suggest a direct link between enhanced levels of oxytocin due to massage and
the psychophysiological state of relaxation and well-being.51 Each rat (both males
and females) was stroked on their ventral side for 5 minutes at a rate of 40 strokes
per minute, the frequency found to be the most effective.51 This treatment lowers
BP, causes sedation, elevates pain threshold and influences the release of GI
hormones.18,26,52 In support of a role for oxytocin in the coordination of the
massage-induced effects, levels of oxytocin were enhanced in the circulation, as
well as in the cerebrospinal fluid, in response to this treatment.51 Furthermore,
an oxytocin antagonist blocked some of the massage-induced effects, such as an
elevated pain threshold.18 Interestingly, a sustained anxiolytic-like effect enhanced
nociceptive threshold, lowered BP and decreased fasting levels of the vagally
controlled hormone gastrin in response to repeated massage-like treatments.
Recent experiments also showed that rats learned conditioned avoidance more
quickly and were more interactive socially after being given this massage-like
treatment.53 The rats treated with the massage-like stroking also gained more
weight than did the control rats.53 These effects on the rats are, in fact, exactly the
same as those described previously, which occur in response to repeated injections
of oxytocin.15 Also, the administration of an oxytocin antagonist counteracts most
of these effects. Taken together, these data support the idea that oxytocin released
by the massage-like stroking may be involved in the effects induced by this
treatment in adults.8
The fact that gastrin levels decreased in response to the massage-like treatment
warrants some comments.8 The hormone gastrin is well-known for its role as
a stimulator of secretion.13 It is released in response to the presence of food
constituents in the stomach and by vagal nerve activity. In fact, the vagal nerve
serves dual roles, as it may stimulate as well as inhibit gastrin release. The latter
“inhibitory” effect is mediated via a cholinergic mechanism.13 Oxytocinergic
neurons project from the paraventricular nucleus to the vagal motor nucleus9
and intracerebral administration of oxytocin may influence the release of gastrin,
suggesting that oxytocin may influence the levels of gastrin via central
mechanisms.54 As mentioned previously, levels of gastrin are lowered not only

in response to the massage-like treatment, but also by other types of nonnoxious
stimulation of somatosensory afferents, such as by warm temperature and touch.13
The release of gastrin caused by these stimuli is likely to involve the central
oxytocinergic pathway described previously.8
Eff e c t s o f Ma s s a ge o n Ho r m o ne Le ve ls in Humans
In the studies performed by Field and colleagues, massage was delivered to humans
of different ages who had different types of disease.1-3 In addition to a reduction
of the intensity of specific symptoms found in these studies, more general findings
were that massage reduced the levels of cortisol, it decreased BP and it induced an
anxiolytic-like effect. All of these effects are consistent with the hypothesis presented
previously in this chapter that nonnoxious sensory stimulation promotes social
interaction and reduces the levels of anxiety and stress.
We have suggested that oxytocin plays an integrative role in this system.14 Very
few studies, however, have actually demonstrated a release of oxytocin in response
to massage in humans. The study mentioned previously in this chapter, in which
infant massage released oxytocin in mothers, represents an exception in this
regard,40 as does a study by Turner and colleagues, who also demonstrated the
release of oxytocin in some adults following massage.55 In a recent study undertaken
with several of my colleagues, a clear rise was observed in the levels of oxytocin
in 10 humans following massage that involved stroking of the shoulders, chest and
the abdominal area (Uvnas-Moberg et al, unpublished data). The response—values
of oxytocin that were approximately doubled (Figure 3)—which was transient,
appeared 10 minutes or 20 minutes after the start of the massage. In another recent
study performed with 30 humans—in which no stroking was applied, but muscles
were relaxed by applying light pressure—my colleagues and I also observed a
significant release of oxytocin in the massage therapist, particularly at the beginning
of treatment (Uvnas-Moberg et al, unpublished data). Those who received massage
in this study displayed clear declines in the levels of GI hormones, especially
gastrin, but also in the levels of cholecystokinin (Figure 4). Thus, results of this
study indicate that the effects of massage on hormone release are bidirectional,
since both the massage therapist and those receiving massage were affected. In
addition, these results demonstrate that oxytocin is released following various types
of massage. In the studies described previously, the rise in the levels of oxytocin
was observed in response to stroking movements and following touch and light
pressure.40,55
Figure 3. Levels of oxytocin in 10 humans before and after receiving stroking and
light-pressure massage.57
Before Massage
After Massage
Figure 4. Levels of A) oxytocin and B) gastrin in 30 humans receiving massage

consisting of holding and light pressure, but no stroking.58
The finding of elevated levels of oxytocin in response to massage supports the

hypothesis that oxytocinergic mechanisms may be involved in some of the effects
induced by massage. It should be noted, however, that the stress-relieving effects
of oxytocin are exerted by release of this hormone within the central nervous
system. Therefore, ideally, oxytocin should be measured in specific brain regions
to determine the relevance of these effects. However, no such studies have been
performed to date, although oxytocin has been shown to be released into the
cerebrospinal fluid by massage-like stroking in rats.15 In addition, a release of
oxytocin into the amygdala of lactating sheep33 and rats34 has been demonstrated
in response to suckling. Since oxytocin that is released in the brain decreases
activity in the HPA axis, decreased levels of cortisol may be used as an indication
of the oxytocin-induced stress-relieving effect.
Interestingly, the decreased levels of gastrin observed following massage

(Uvnas-Moberg et al, unpublished data) may be used as an indirect measurement
of oxytocin release in the DMX area, which controls vagal nerve activity, but may
also reflect oxytocin-mediated stimulation of social interaction. As noted previously,
the levels of gastrin and CCK can be decreased following stimulation of sensory
nerves in rats and humans.45 Levels of gastrin and CCK in rats may be lowered
by touch, warmth and stroking via a cholinergic inhibitory vagal mechanism.56
When premature babies are placed in the “kangaroo” position—ventral skin-to-skin
contact with their mothers—levels of CCK fall.45 During feeding, the response is
facilitated.45 Levels of gastrin decrease only 4 days after birth in women who are
breastfeeding who have also been allowed to have skin-to-skin contact with their
babies during the sensitive period immediately after birth.46 These women also
have more-pronounced pulsatile oxytocin release in relation to their breastfeeding.
In addition, these women have been noted to increase their social interactions
with their children.39,49 Since oxytocin decreases gastrin levels and increases social
competence, low fasting gastrin and/or CCK levels may be an indirect reflection
of high social interests/skills. Interestingly, such a relationship is supported by, and
may explain the unexpected finding of, a strong negative (ie, inverse) correlation
between the personality trait “socialization” (as measured by the Karolinska scale
of personality) and levels of gastrin in 2 independent studies involving lactating48
and nonlactating women.57
As discussed previously, a subset of sensory nerves, which join the sensory fibers
of the vagal nerve, can stimulate the CNS from the skin of the chest and the
abdominal area.41 Ventral physical contact is followed by an immediate state of

relaxation and well-being, often in combination with increased social activity.
Levels of gastrin and CCK, the vagally controlled hormones, decrease. This effect is
caused by activation of inhibitory vagal nerve fibers. Contrastingly, the stimulatory
vagal fibers are activated during food intake. In this situation, the levels of gastrin
and CCK rise to promote the digestion of food and subsequent anabolic
metabolism. From a simplistic physiological point of view, it might be assumed that
low levels of gastrin, as a general trait, reflect a stronger mental representation of
the skin and “external world” and a consequent activation of the inhibitory vagal
fibers. Contrastingly, individuals who have higher basal levels of gastrin might be
more influenced by a stronger representation of the GI tract and an “internal
world” that might be connected to a higher tone in the stimulatory vagal fibers.
The former, externally oriented group of individuals is likely to score higher in the
personality trait “socialization” than is the latter, more internally oriented group
of individuals. Interestingly, massage may modify these patterns.
Su mm a r y
Hormone levels are used to evaluate the effects of massage on humans. Cortisol
levels in humans have been noted to decrease following massage and are assumed to
reflect an antistress effect caused by the massage. Increased oxytocin levels following
massage may be related to antistress-like effects and sociability, whereas lowered lev-
els of gastrin following massage may be a marker of a massage-induced increase in
social interests and skills.
Re f e re n c e s
1. Field T. Massage therapy for infants and children. Journal of Developmental and Behavioral
Pediatrics: JDBP. 1995;16:105-111.
2. Field T, Hernandez-Reif M, Hart S, Theakston H, Schanberg S, Kuhn C. Pregnant women
benefit from massage therapy. Journal of Psychosomatic Obstetrics and Gynaecology. 1999;20:
31-38.
3. Hernandez-Reif M, Field T, Diego M, Beutler J. Evidence-based medicine and massage.
Pediatrics. 2001;108:1053.
4. Schanberg S, Field TM. Maternal deprivation and supplemental stimulation. In: Field TM,
McCabe PM, Schneiderman N, eds. Stress and Coping: Across Development. Hillsdale, NJ:
Lawrence Erlbaum Associates; 1988.
5. Cannon WB. The Wisdom of the Body. New York, NY: Norton; 1932.
6. Uvnäs-Moberg K. Oxytocin-linked antistress effects — the relaxation and growth response.
Acta Physiologica Scandinavica Supplementum. 1997;161(suppl 640):38-42.
7. Uvnäs-Moberg K. Antistress pattern induced by oxytocin. News in Physiological Sciences: An

International Journal of Physiology Produced Jointly by the International Union of Physiological
Sciences and the American Physiological Society. 1998;13:22-25.
8. Uvnäs-Moberg K. Oxytocin may mediate the benefits of positive social interaction and
emotions. Psychoneuroendocrinology. 1998;23:819-835.
9. Sofroniew MV. Vasopressin and oxytocin in the mammalian brain and spinal cord. Trends
in Neurosciences. 1983;6:467-472.
10. Richard P, Moos F, Freund-Mercier MJ. Central effects of oxytocin. Physiological Review.
1991;71:331-370.
11. Insel TR. Oxytocin — a neuropeptide for affiliation: evidence from behavioral, receptor
autoradiographic, and comparative studies. Psychoneuroendocinology. 1992;17:3-35.
12. Carter CS. Neuroendocrine perspectives on social attachment and love.
Psychoneuroendocrinology. 1998;23:779-818.
13. Uvnäs-Moberg K, Lundeberg T, Bruzelius G, Alster P. Vagally mediated release of gastrin
and cholecystokinin following sensory stimulation. Acta Physiologica Scandinavica.
1992;146:349-356.
14. Uvnäs-Moberg K, Bruzelius G, Alster P, Lundeberg T. The antinociceptive effect of
non-noxious sensory stimulation is mediated partly through oxytocinergic mechanisms. Acta
Physiologica Scandinavica. 1993;149:199-204.
15. Uvnäs-Moberg K, Ahlenius S, Hillegaart V, Alster P. High doses of oxytocin cause sedation
and low doses cause an anxiolytic-like effect in male rats. Pharmacology, Biochemistry, and
Behavior. 1994;49:101-106.
16. Windle RJ, Shanks N, Lightman SL, Ingram CD. Central oxytocin administration reduces
stress-induced corticosterone release and anxiety behavior in rats. Endocrinology.
1997;138:2829-2834.
17. McCarthy MM, Altemus M. Central nervous system actions of oxytocin and modulation
of behavior in humans. Molecular Medicine Today. 1997;3:269-275.
18. Ågren G, Lundeberg T, Uvnäs-Moberg K, Sato A. The oxytocin antagonist 1-deamino-2-D-

Tyr-(Oet)-4-Thr-8-Orn-oxytocin reverses the increase in the withdrawal response latency
to thermal, but not mechanical nociceptive stimuli following oxytocin administration or
massage-like stroking in rats. Neuroscience Letters. 1995;187:49-52.
19. Petersson M, Alster P, Lundeberg T, Uvnäs-Moberg K. Oxytocin causes a long-term decrease
of blood pressure in female and male rats. Physiology & Behavior. 1996;60:1311-1315.
20. Petersson M, Alster P, Lundeberg T, Uvnäs-Moberg K. Oxytocin increases nociceptive
thresholds in a long-term perspective in female and male rats. Neuroscience Letters.
1996;212:87-90.
21. Petersson M, Ahlenius S, Wiberg U, Alster P, Uvnäs-Moberg K. Steroid dependent effects
of oxytocin on spontaneous motor activity in female rats. Brain Research Bulletin.
1998;45:301-305.
22. Petersson M, Hulting AL, Uvnäs-Moberg K. Oxytocin causes a sustained decrease in plasma
levels of corticosterone in rats. Neuroscience Letters. 1999;264:41-44.
23. Uvnäs-Moberg K, Bjokstrand E, Hillegaart V, Ahlenius S. Oxytocin as a possible mediator
of SSRI-induced antidepressant effects. Psychopharmacology. 1999;142:95-101.
24. Uvnäs-Moberg K, Bjorkstrand E, Salmi P, Johansson C, Astrand M, Ahlenius S. Endocrine
and behavioral traits in low-avoidance Sprague-Dawley rats. Regulatory Peptides. 1999;80:
75-82.
25. Arletti R, Bertolini A. Oxytocin as an antidepressant in two animal models of depression.
Life Sciences. 1987;41:1725-1730.
26. Uvnäs-Moberg K, Alster P, Petersson M. Dissociation of oxytocin effects on body weight in

two variants of female Sprague-Dawley rats. Integrative Physiological and Behavioral Science:
The Official Journal of the Pavlovian Society. 1996;31:44-55.
27. Petersson M, Lundeberg T, Sohlström A, Wiberg U, Uvnäs-Moberg K. Oxytocin increases
the survival of musculocutaneous flaps. Naunyn-Schmiedeberg’s Archives of Pharmacology.
1998;357:701-704.
28. Petersson M, Uvnäs-Moberg K, Erhardt S, Engberg G. Oxytocin increases locus coeruleus

alpha 2-adrenoreceptor responsiveness in rats. Neuroscience Letters. 1998;255:115-118.
29. Petersson M, Lundeberg T, Uvnäs-Moberg K. Oxytocin enhances the effects of clonidine
on blood pressure and locomotor activity in rats. Journal of the Autonomic Nervous System.
1999;78:49-56.
30. Díaz-Cabiale Z, Petersson M, Narváez JA, Uvnäs-Moberg K, Fuxe K. Systemic oxytocin

treatment modulates alpha2-adrenoceptors in telencephalic and diencephalic regions of the
rat. Brain Research. 2000;887:421-425.
31. Rajkowski J, Kubiak P, Ivanova S, Aston Jones G. State related activity, reactivity of locus
ceruleus neurons in behaving monkeys. In: Richard P, Moos F, Freund-Mercier M-J. Central
effects of oxytocin. Physiological Reviews. 1991;71:331-370.
32. Stock S, Uvnäs-Moberg K. Increased plasma levels of oxytocin in response to afferent
electrical stimulation of the sciatic and vagal nerves and in response to touch and pinch in
anaesthetized rats. Acta Physiologica Scandinavica. 1988;132:29-34.
33. Kendrick KM, Keverne EB, Baldwin BA, Sharman DF. Cerebrospinal fluid levels of
acetylcholinesterase, monamines and oxytocin during labour, parturition, vaginocervical
stimulation, lamb separation and suckling in sheep. Neuroendocrinology. 1986;44:149-156.
34. Neumann ID, Krömer SA, Toschi N, Ebner K. Brain oxytocin inhibits the (re)activity of the
hypothalamo-pituitary-adrenal axis in male rats: involvement of hypothalamic and limbic
brain regions. Regulatory Peptides. 2000;96:31-38.
35. Uvnäs-Moberg K. Physiological and endrocine effects of social contact. Annals of the New
York Academy of Sciences. 1997;807:146-163.
36. Vallbo AB, Olausson H, Wessberg J. Unmyelinated afferents constitute a second system
coding tactile stimuli of the human hairy skin. Journal of Neurophysiology. 1999;81:
2753-2763.
37. Eriksson M, Lundeberg T, Uvnäs-Moberg K. Studies on cutaneous blood flow in the
mammary gland of lactating rats. Acta Physiologica Scandinavica. 1996;158:1-6.
38. Widström AM, Ransjö-Arvidson AB, Christensson K, Matthiesen AS, Winberg J,
Unväs-Moberg K. Gastric suction in healthy newborn infants. Effects on circulation and
developing feeding behaviour. Acta Paediatrica Scandinavica. 1987;76:566-572.
39. Nissen E, Uvnäs-Moberg K, Svensson K, Stock S, Widström AM, Winberg J. Different
patterns of oxytocin, prolactin but not cortisol release during breastfeeding in women
delivered by caesarean section or by the vaginal route. Early Human Development.
1996;45:103-118.
40. Matthiesen AS, Ransjö-Arvidson AB, Nissen E, Uvnäs-Moberg K. Postpartum maternal

oxytocin release by newborns: effects of infant hand massage and sucking. Birth (Berkeley,
CA.). 2001;28:13-19.
41. Eriksson M, Lindh B, Uvnäs-Moberg K, Hökfelt T. Distribution and origin of
peptide-containing nerve fibers in the rat and human mammary gland. Neuroscience.
1996;70:227-245.
42. Christensson K, Cabrera T, Christensson E, Unväs-Moberg K, Winberg J. Separation distress

call in the human neonate in the absence of material body contact. Acta Paediatrica.
1995;84:468-473.
43. Bystrova K, Widström AM, Matthiesen AS, et al. Skin-to-skin contact may reduce negative
consequences of ‘the stress of being born’: a study on temperature in newborn infants,
subjected to different ward routines in St. Petersburg. Acta Paediatrica. 2003;92:320-326.
44. Uvnäs-Moberg K. Neuroendocrinology of the mother-child interaction. Trends in

Endocrinology and Metabolism. 1996;7:126-131.
45. Törnhage CJ, Serenius F, Uvnäs-Moberg K, Lindberg T. Plasma somatostatin and

cholecystokinin levels in response to feeding in preterm infants. Journal of Pediatric
Gastroenterology and Nutrition. 1998;27:199-205.
46. Klaus MH, Klaus PH. The Amazing Newborn. Reading, Mass: Addison-Wesley Publishing
Company; 1985.
47. Widström AM, Wahlberg V, Matthiesen AS, et al. Short-term effects of early suckling and
touch of the nipple on maternal behavior. Early Human Development. 1990;21:153-163.
48. Uvnäs-Moberg K, Widström AM, Nissen E, Björvell H. Personality traits in women 4 days
postpartum and their correlation with plasma levels of oxytocin and prolactin. Journal of
Psychosomatic Obstetrics and Gynaecology. 1990;11:261-273.
49. Nissen E, Gustavsson P, Widström AM, Uvnäs-Moberg K. Oxytocin, prolactin, milk
production and their relationship with personality traits in women after vaginal delivery
or Cesarean section. Journal of Psychosomatic Obstetrics and Gynaecology. 1998;19:49-58.
50. Altemus M, Redwine LS, Leong YM, Frye CA, Porges SW, Carter CS. Responses to
laboratory psychosocial stress in postpartum women. Psychosomatic Medicine. 2001;63:
814-821.
51. Kurosawa M, Lundeberg T, Ågren G, Lund I, Uvnäs-Moberg K. Massage-like stroking of

the abdomen lowers blood pressure in anesthetized rats: influence of oxytocin. Journal of the
Autonomic Nervous System. 1995;56:26-30.
52. Lund I, Lundeberg T, Kurosawa M, Uvnäs-Moberg K. Sensory stimulation (massage) reduces

blood pressure in unanaesthetized rats. Journal of the Autonomic Nervous System. 1999;78:
30-37.
53. Lund I, Yu LC, Uvnäs-Moberg K, et al. Repeated massage-like stimulation induces long-term
effects on nociception: contribution of oxytocinergic mechanisms. The European Journal of
54. Björkstrand E, Ahlenius S, Smedh U, Uvnäs-Moberg K. The oxytocin receptor antagonist
1-deamino-2-D-Tyr-(OEt)-4-Thr-8-Orn-oxytocin inhibits effects of the 5-HT1A receptor
agonist 8-OH-DPAT on plasma levels of insulin, cholecystokinin and somatostatin.
Regulatory Peptides. 1996;63:47-52.
55. Turner RA, Altemus M, Enos T, Cooper B, McGuiness T. Preliminary research on plasma
oxytocin in normal cycling women: investigating emotion and interpersonal distress.
Psychiatry. 1999;62:97-113.
56. Uvnäs-Moberg K, Alster P, Hillegaart V, Ahlenius S. Oxytocin reduces exploratory motor

behaviour and shifts the activity towards the centre of the arena in male rats. Acta Physiologica
Scandinavica. 1992;145:429-430.
57. Uvnäs-Moberg K, Arn I, Theorell T, Jonsson CO. Personality traits in a group of individuals
with functional disorders of the gastrointestinal tract and their correlation with gastrin,
somatostatin and oxytocin levels. Journal of Psychosomatic Research. 1991;35:515-523.
S E C T I O N V.
MASSAGE THERAPY
FOR ORPHANS AND
P E D I AT R I C P RO B L E M S
CHAPTER 13:
MASSAGE THERAPY
WITH PRETERM INFANTS
AND CHILDREN WITH
CHRONIC ILLNESSES
Herminia L. Cifra, MD, MSc, MHA, FPPS1,2

Melanie N. Sancho, MD, FPPS2
1 Department of Pediatrics
University of the Philippines
College of Medicine
2 Philippine Children’s Medical Center
Ab s t ra c t
Research on massage in the Philippines has revealed its positive effects on many
functions in infants and children. These findings have led to widespread support
of massage therapy in the local pediatric community. This chapter focuses on the
results of studies from research in the Philippines on the effects of massage therapy
on infant behavior, weight gain, arterial oxygen tension, pain reduction, stress
hormones and immunoglobulins, asthma and labor. Massage therapy research
on neglected and abused children and burn management is also reviewed.
212 Section V. Massage Therapy for Orphans and Pediatric Problems
Introd uc t i o n
Touch therapy is quickly becoming an integral part of modern pediatric practice.

As pediatricians, we have always been cognizant that tactile stimulation exists
throughout life, but that it appears to be most intense and crucial during infancy.
Several investigators have cited tactile stimulation as playing a very significant
role in the growth and development of the infant. For example, Ribble theorized
that the infant has inborn “stimulus hunger” for various types of stimulation,
including gentle touch, body contact, body positioning and sound.1 Without
regularly available maternal stimulation, the premature infant may consequently
be expected to suffer from stimulus deprivation. This consequence eventually
affects the neurodevelopmental outcome of these babies. My colleagues and I have
investigated the notion that massage therapy, as a medical intervention, would
have both physiological and psychological effects on the infant.2 In studies we have
conducted, massage therapy was found to affect metabolism, intestinal motility and
glandular, biochemical and muscular function.3 Psychologically, massage promotes
relaxation in infants. As a result of these findings, massage has gained significant
support and popularity in our local pediatric community.
As noted by leading neonatologists and gastroenterologists who advocate

breastfeeding and essential nutrition for optimal growth and development,
breastfeeding needs to be augmented by other kinds of touch, whether through
kangaroo care, grooming or massage therapy.4 Thus, massage therapy has a definite
function in a child’s nutrition. This is achieved by promoting a relaxed state that is
beneficial to preterm infants. Enhancing the level of relaxation of preterm infants
during feedings results in greater weight gain, thereby shortening the length of
hospital stays.
The massaging of infants provides a gentle and positive stimulus for newborns.
This touch therapy is a significant way to maintain and enhance the attachment
process between parents and baby. One of the major goals and concerns of
pediatricians in the Philippines has been to heighten the neurodevelopmental
outcome of children. The value of touch has been supported by several clinically
significant studies that highlight the benefits of this relatively inexpensive medical
intervention. In many countries, therapeutic massage has been supported by
numerous scientific studies, including those conducted at the Touch Research
Institute of the Philippine Children’s Medical Center. In these countries, massage
Chapter 13: Massage Therapy With Preterm Infants and Children With Chronic Illnesses 213
therapy is no longer an alternative medicine, but rather is now part of routine

health care and mainstream medicine. The increasing number of extremely
immature and sensitive—but viable—newborns constitutes a growing challenge
to provide an adequately appropriate extrauterine environment during and
immediately following hospitalization. Numerous studies have explored the
benefits of providing tactile stimulation for these preterm newborns.5 This
intervention appears to facilitate the recovery of small, sick babies who find
themselves in incubators.
Pe r f o r ma n c e o n t h e Bra z e l t o n Ne onatal Be h avioral

A ss e ss me n t S c a l e
In her award-winning paper on perinatal care, Pablo demonstrated clearly how

massage improved performance on the Brazelton Neonatal Behavioral Assessment
Scale, especially for those newborns who were not delivered spontaneously and
thus did not experience the massage of normal delivery through the birth canal.6
These babies delivered by cesarean section (C-section) who were massaged also
experienced a greater incidence of breastfeeding and rooming-in.7
Weight Gain
In an initial study undertaken by Jinon et al, preterm babies exhibited a 45%

greater weight gain following 15-minute massages, 3 times per day, for 10 days
(Table 1).8 These babies also were discharged earlier than those who were not given
massage. Further, the massaged infants were observed to be more active during the
stimulation sessions than during nonstimulation sessions. Following publication
of these findings, additional researchers were inspired to explore different aspects
of health and development in preterm infants.9
Ar t e r i a l O x y ge n Te ns i o n
Tirador et al demonstrated the beneficial effects of massage therapy on intubated

infants.10 He reported the lowering of arterial oxygen (O2) tension among stable
babies who were given 15-minute massages, once daily, for 3 consecutive days.
The functional, inspired O2 required to maintain optimal O2 saturation in
“graduates” of the neonatal intensive-care unit (NICU) was reduced notably.
The study also revealed that massaged babies experienced significantly higher
transcutaneous O2 levels and a lower fractional concentration of O2 in inspired
gas and pressure requirements compared with control babies who did not receive
massage therapy (Table 2).
Table 1. Mean and Standard Deviations for the Formula Intake and Weight Gain
Massage Group Control
(n = 20) (n = 20)
Measures M SD M SD P Value
Average daily weight gain

40 4.2 36 3.7 NS
prior to study (3 days) (g)
Average daily weight gain

89.7 62.4 61.4 51.8 <.5
during study (g)
Number of feeds/day 12 0 12 0 NS
Average Fluid Intake

Cc/k/day 17.5 11.5 176.1 12.5 NS
Calories/kg/day 123 11.3 122.6 11.4 NS
Calories/oz 20 0 20 0 NS
NS = not significant
Table 2. Effect of Massage Therapy on Intubated Infants: Comparison of

Transcutaneous PO2 Levels11
Experimental Control P Value
(n = 15) (n = 15)
1st TcPO2 37.4+/-6.4 38.7%+/-11.8 .7085
2nd TcPO2 40.6%+/-4.5 36.3%+/-7.2 .0756
Mean change 3.2+/-8.4 -2.4+/-14 .223486

Pa i n Re d uc t i o n
As early as the moment of birth, the infant can already feel pain. Early studies of
neonatal development suggested that the neonate’s response to pain was decorticate
in nature, and that the perception or localization of pain was not present.11,12
This supposition was believed for many years and was applied particularly to
preterm infants who were thought to be incapable of feeling pain since they
lacked the fundamental neuronal pathways to transmit messages regarding painful
experiences. However, current evidence suggests that pain pathways and the
neurotransmitters involved in the transmission of impulses relating to noxious
stimuli are well developed at an early stage in gestation and that infants respond
to noxious stimuli in a variety of ways.13
Infants in the NICU are handled constantly, and the tactile stimuli they receive
are often painful and uncomfortable. Unfortunately, there are innumerable ways
in which a neonate experiences pain in a modern NICU: Pain-producing
interventions—from intravenous lines to nasogastric tubes, from nasal catheters
to indwelling cannulas, and from intubations to ventilations—cannot be avoided
completely if seriously ill babies are to survive. It would be beneficial for the
infants, of course, if the staff in the NICU was trained in massage therapy
techniques and allotted adequate time for this intervention.
Infants in the NICU can benefit from tactile stimuli

In 1998, at the University of the Philippines–Philippine General Hospital, NICU

workers provided massages during the heelstick procedure. Now, when babies are
touched in this unit, they do not invariably feel the infliction of pain.14 A study by
Dimaano-Aliwalas and Cifra showed that massage applied before, during and after a
heelstick reduced the infant’s response to pain, including a longer latency to crying,
a shorter duration of crying and a faster return to baseline (Table 3).15 The effect of
the massage was comparable to that of sucrose in pain reduction. In a similar study
by Jarumahum and Cifra, massage proved as effective as sucrose in reducing pain
following heelsticks.16 The authors noted that massage does not have the known
disadvantages of labile sucrose levels inimical to brain and adrenal function.
Table 3. Effects of Massage Therapy and Sucrose on Reducing Pain Following Heel
Sticks in Infants16
Group 1 Group 2 Group 3 Kruskal-Wallis
Measures Massage Sucrose Control 1-Way Anova
Median n = 17 n = 17 n = 17 Results
Before Tx
Heart Rate (bts/min) 135 139 130 NS

Respiratory Rate (brt/min) 44 44 46 NS
O2 Saturation (%) 0 0 0 NS
NIPS 97 98 97 NS
After Tx
Cry Duration (min) 2.67 2.50 4.00 P<.05
Cry Latency (min) 3.00 3.00 1.00 P<.05
Time to Baseline (min) 3.00 3.00 5.00 P<.05

NIPS 1.00 2.00 3.00 NS
Dimaano-Aliwalas et al, UP-PGH 1998

NS = Not significant
In research similar to that done on pain in neonates, Senturias et al showed that

delivering massage to just one body site—without whole body massage—reduced
the perception of pain in that area as well as the perception of general pain
body-wide after a painful intramuscular injection.17 This study involved
pediatric patients receiving monthly penicillin injections at a clinic for rheumatic
heart-disease patients. Of the subjects who received massage, 83% reported that
they experienced decreased levels of pain compared with those who did not receive
massage over the injection site. Some of the study’s participants, in fact, requested
a massage prior to subsequent injections.
St re s s Ho r m o ne s a nd Im m u no gl obulins
The response to stress in humans involves increases in the concentrations of

catecholamines, including norepinephrine, epinephrine, adrenocorticotropic
hormone (ACTH), cortisol, and thyroid hormones, and a concomitant decrease in
the levels of serum serotonin. The levels of these stress hormones have reportedly
decreased in several studies investigating massage therapy.5,18,19
In a study by Serafica and Cifra involving critically ill infants and children,
massage therapy reduced the levels of many stress hormones, including ACTH,
cortisol, triiodothyronine (T3) and thyroxine (T4) (Table 4).20 It also increased
the levels of IgG, IgA and IgM in ill infants (median age, 4.1 months).
In these studies, which were conducted in the NICU and the pediatric ICU,
other positive effects were noted anecdotally: better mother-infant bonding and
nurse-patient-family interactions and greater family involvement in childcare.
In addition, the incidence of breastfeeding was greater and a better acceptance
of rooming-in practices was observed. Nurses also reported feeling greater job
satisfaction. Parents who were initially tired and hesitant to touch their babies
subsequently became relaxed and happy as a direct result of touching their infants,
as did surrogate parents and the extended family of grandparents, uncles, aunts
and older brothers and sisters who often substituted for the parents in applying
massage.
Table 4. Effects of Massage on Serum Stress Hormones and Immunoglobulins in PICU

Patients22
Massage Group ISD Control Group ISD P-Value

(n = 20) (n = 20)
Sex Ratio 9:11 10:10
Age 3.99+/-5.71 4.30 (+/-4.73)
Cortisol Baseline 29.17 (+/-13.03) 19.93 (+/-9.87)
After 1 week 15.26 (+/-8.3) 16.98 (+/-8.76) <.01
Normal 6-28 ug/dL
T3 Baseline 1.11 (+/-0.41) 1.06 (+/-0.36)
After 1 week 0.74 (+/-0.31) 0.90 (+/-0.33) .013
Normal 1.16-4.00 nmol/L
T4 Baseline 89.82 (+/-31.72) 89.14 (+/-22.86)
After 1 week 67.81 (+/-32.56) 87.05 (+/-21.2) <.01
Normal 3.2 - 12.6 ug/L
IgG Baseline 146.45 (+/-62.62) 110.02 (+/-52.13)
After 1 week 213.9 (+/-82.22) 148.86 (+/-60.43) .003
Normal 176-601 mg/dL
IgA Baseline 110.39 (+/-68.47) 94.88 (+/-32.91)
After 1 week 142.62(+/-86.62) 108.96 (+/-40.36) .0047

Normal 44-840 mg/L
IgM Baseline 217.48 (+/-101.57) 192.71 (+/-50.77)
After 1 week 310 (+/-104.60) 194.20 (+/- 50.77) .004
Normal 170-1050 mg/dL
P value ≤.05
He a li n g
Moderate pressure massage can enhance the body’s energies to promote healing. In
addition to reducing the levels of certain types of pain, particularly acute pain, it
can alleviate the anxiety that often accompanies pain as well as alter the perception
of pain. The “gate control theory” of pain, a potential explanation for the alleviation
of pain,12 states that inhibitory interneurons that block input from the classic
nociceptors to the spinal cord are activated by mechanisms that decrease stress.
A st hma
Massage therapy has been shown to improve pulmonary function in children.21

Similar data were reported by Balanag et al22 regarding the improvement of
pulmonary function in asthmatic children (Table 5). One notable finding is the
improvement in FEF 25%-75%, which is a sensitive measure of airflow through
small airways. Attitudes toward asthma also improved, which is an important
achievement in a chronic disease like asthma.
Table 5. The Effect of Massage Therapy on Pulmonary Function Test and Patient
Satisfaction of Asthmatic Patients22
Massage Group Control Group

(n = 15) (n = 15)
Pre Post Diff P-Value Pre Post Diff P-Value
Pain Assessment
0.64 0.48 0.16 <.00001* 0.63 0.56 0.06 <.05†
Scale (Guilford, 1990)
FVC 52.1 61.7 9.6 .03* 58.5 64.7 6.7 .01*
FEV 57.2 61.7 9.6 .03* 64.6 71.7 7.26 .004*
PEF 47.6 67.3 19.7 <.00001* 56.1 63.1 7.06 .001*
FEF (25%-75%) 86.2 100.8 14.6 .02 * 98.9 102.7 3.8 <.05†
*Significant; †Not Significant; P-value = <.05; FVC= Forced Vital Capacity; FEV= Forced Expiratory Volume;
PEV= Peak Expiratory Velocity; FEF 25%-75%= Forced Expiratory Flow at 25%-75% of FVC; (% of Predicted)
Labor
Support during labor is centuries old. According to Klaus and colleagues,23 the
challenge is to use obstetric technology only when necessary, relying instead on
the practice of continuous support during labor to help the birth process follow its
natural course. This concept maximizes the importance of the doula (a woman who
gives care to a pregnant woman during labor). Studies worldwide have revealed that
support given by a doula reduces the use of epidural anesthesia by 60% and the
rate of C-sections by 30%.24 This intervention has been shown to play a significant
role in unwanted teenage pregnancies: Sarte and Cifra found that unmarried,
teenage mothers who were given care by a doula during labor experienced reduced
levels of anxiety and pain, shorter lengths of labor and less need for medication
(Table 6).25 Reduced uses of forceps and oxytocin were also noted.
Table 6. Responses to Questionnaires Within 24 hrs of Birth by Unmarried, Teenage

Mothers Given Care by Doula (Expressed as Mean ISE)25
Variables n Support n Control P-Value
State Anxiety Scores 45 28.2 (.85) 45 37.8 (1.10) .00001*
Present Pain Index 45 2.4 45 3.8 .05*
Pain at time of questioning:

moderate 45 9.0 (20.7) 45 15.0 (43.4) .05*
severe 45 20.0 (3.3) 45 11.0 (24) .00001*
Mother’s perception of labor:

very difficult 45 16.0 (34.8) 45 26.0 (59.4) .002*
coped well 45 24.0 (58.7) 45 11.0 (24) .00001*

felt very tense 45 14.0 (30.4) 45 27.0 (59) .00001*
much worse
45 17.0 (77) 45 28.0 (61.6) .003*
than imagined
*Statistically significant
Infants whose mothers received support from a doula have been found to receive
more care and touch in the first 20 minutes after birth.26 The incidences of
breastfeeding and demand-feeding were also higher in the group cared for by
a doula. Since health problems in these infants also appeared to decrease, the
pediatricians at the Philippine General Hospital are actively promoting the use
of doulas to their obstetricians.
Fu t ure Di re c t i o ns
Several studies involving touch and massage are in progress or are in the planning
stages. Ongoing studies are being conducted on neglected and sex-abused children.
It is expected that touch therapy will decrease victims’ anxiety, depression and
aversion to touch. It may also hasten their ability to establish new relationships.
In addition, children with autism are expected to benefit from massage therapy:
Massage therapy has improved relationships between these special pediatric patients
and their caregivers as well as their social environments.27
Touch therapy is being used currently to hasten the return of sensory perception
and responsiveness in comatose children (unpublished data). Massage may, in fact,
enhance circulation and improve mental functioning. Levels of cortisol have
decreased notably in comatose patients during massage therapy. This benefit
has been accompanied by heightened alertness, based on electroencephalograms
showing a decrease in alpha brain waves and increases in beta and theta brain
waves—alterations consistent with heightened alertness. These measures, it appears,
will continue to be used on comatose children.
Massage therapy has been shown to reduce the incidence and severity of headache
and abdominal colic.28 Infant colic is a very common complaint presented by
parents. Massage of the infant during episodes of colic and prior to bedtime may
enhance food absorption, increase positive feeding interactions with caregivers,
result in more-organized sleep/wake behaviors and reduce irritability and stress
behaviors.
Research shows that the management of burns may improve with massage by
raising the threshold of pain.29 Extensive physical handling of the burn patient
causes severe-to-excruciating pain. Ongoing research is assessing the levels of

depression, anxiety and pain before and after skin debridement as well as how these
parameters are reduced if massage is conducted before the skin-brushing procedure.
Pediatric patients with hematological conditions who undergo painful lifetime

procedures, as well as who suffer the side effects and complications of their
prescribed treatments, are another group of patients that may benefit from touch
therapy.30 In fact, massage therapy is now being tested to determine its benefits on
those who have experienced frequent painful procedures as well as on its effects on
alleviating depression. Expected benefits include lower stress levels during invasive
procedures and reduced feelings of helplessness among parents.31
Massage therapy has been shown to alter the levels of stress hormones, such as
cortisol,32 and growth hormones, gastrin and insulin, in urine, saliva and blood.33
Gastrointestinal (GI) hormones, such as gastrin and insulin, are essential for the
process of food absorption, which, when enhanced, leads to weight gain. Previous
findings suggest that massage appears to increase vagal activity. This, in turn,
stimulates the production and release of the GI hormones.34 Studies conducted
at the Philippine General Hospital determined how touch can alter the level of
these hormones.35
Also being investigated is the best way to apply touch. Investigations are assessing
whether whole-body touch, touch of areas specifically related to the condition,
or procedure-related touch are indicated in specific situations. Studies are also
evaluating the use of oils and other materials as mediators to enhance the
applications of touch. Results of these various studies will soon be disseminated
to health workers and massage therapists in order to deliver this cost-effective
intervention most effectively.
Conclusions
Massage therapy has opened up many avenues in health maintenance, the

promotion of health, the prevention of disease and even cure. Such diverse
applications of touch therapy have encouraged researchers to investigate further
and support such applications with clinical trials. Touch therapy has already been
shown to be critical to physiological and social development.36
Modern endocrinology, neurology, gastroenterology, immunology and sociology

laboratories have supported this work. What were once just conjectures and
anecdotal observations have now been substantiated pathophysiologically. Thus,
modern scientific evidence is revealing what our ancestors have been telling us for
a long time—touch maintains and even enhances health.
This research was supported by the National Institute of Health of the Philippines,
and by Johnson & Johnson Baby Company of the Philippines.
Ref ere n c e s
of the tactile stimulation as a form of early intervention: a quantitative evaluation. Journal
of Developmental and Behavioral Pediatrics. 1987;8:68-76.
2. White J, Labarba R. The effects of tactile and kinesthetic stimulation on neonatal
development in the premature infant. Developmental Pscyhobiology. 1976;9:569-577.
3. Kuhn C, Schanberg S, Field T, et al. Tactile/kinesthetic stimulation effects on sympathetic
and adrenocortical function in preterm infants. Journal of Pediatrics. 1991;119:434-440.
4. Scholz K, Samuels C. Neonatal bathing and massage intervention with fathers, behavioral
effects 12 weeks after birth of the first baby: the Sunrayasia Australia Intervention Project.
International Journal of Behavior Development. 1992;15:67-81.
5. Schanberg S, Field T. Sensory deprivation stress and supplemental stimulation in the rat pup
and preterm human neonate. Child Development. 1987; 5-8:1431-1447.
6. Klaus MH, Kennell JH, Plumb N, Zuehlke S. Human maternal behavior at the first contact
with her young. Pediatrics. 1970;46:187-192.
7. Righard L, Alade MO. Effects of delivery room routines on success of first breast-feed.
Lancet. 1990;77:1105-1107.
8. Jinon S, Yabes-Almirante C, de Luna MB, eds. The effect infant massage on growth of
pre-term infants. Increasingly Safe and Successful Pregnancies. Netherlands: Elsevier Science BV,
1996:264-269.
9. Wheeden A, Scafidi FA, Field T, Ironson G, Valdeon C, Bandstra E. Massage effects on
cocaine-exposed pre-term neonates. Journal of Developmental and Behavioral Pediatrics.
1994;14:318-322.
10. Tirador MH, Bote JR, Cifra H, Hernandez EA. The effect of massage therapy on arterial
oxygen and carbon dioxide tension of patients on continuous positive airway pressure
(CPAP). J&J Phil. Clinical Reprints. 1998.
11. Anand KJS, Hickey PR. Pain and its effects in the human neonate and fetus. The New
England Journal of Medicine. 1987;19:1321-1329.
12. Bogdeck N. Understanding pain pathways. Medical Progress. March 1989;38-48.
13. Rushfort JA, et al. Behavioral response to pain in healthy neonates. Archives of Diseases
in Childhood. 1994;70:174-176.
14. Tobias JD, Rasmussen GE. Pain management and sedation in pediatric intensive care unit.
Pediatric Clinics of North America. 1994;41:1269-1292.
15. Dimaano-Aliwalas, Cifra HL. Massage therapy as a modality of pain relief among neonates.
UP-PGH, Johnson & Johnson Clinical Reprints. 1998
16. Jarumahum J, Cifra HL. Sucrose and massage: modalities to pain relief in routine neonatal
procedures. UP-PGH, Johnson & Johnson Clinical Reprints. 1997.
17. Senturias JSN, Cifra HL, Ortiz EE, Angtuaco L, Camagay I. The use of touch therapy as
a modality to decrease the pain of intramascular injection of benzathine penicillin G among
pediatric patients. UP-PGH, Johnson & Johnson Clinical Reprints. 1998.
18. Uvnas-Moberg K, Windberg J. Release of AI hormones in mother and infant by sensory
stimulation. Aeta Paediatrica Scandinavica. 1987;76:851-860.
19. Acolet D, et al. Changes in plasma cortisol and catecholamine concentrations in response
to massage in preterm infants. Archives of Disease in Childhood. 1993;68:29-31.
20. Serafica EM, Cifra HL. The effect of massage on the level of serum stress hormones and
serum immunoglobulins in critically ill infants and children. UP-PGH, Johnson & Johnson
Clinical Reprints. 1999.
21. Field T, Henteleff T, Hernandez-Reif M, et al. Children with asthma have improved
pulmonary functions after massage therapy. The Journal of Pediatrics. 1998;132:854-858.
22. Balanag E, Alviado J, Cifra HL. The effect of massage therapy on pulmonary function test
and patient satisfaction of asthmatic patients. Pediatric Capsule. Vol 1 No. 1;2002.
23. Klaus MH, Kennell JH, Robertson SS. The effects of social support during parturition
on maternal and infant morbidity. BMJ (Clinical Research Ed). 1986;293:585-587.
24. Sosa R, Kennell JH, Robertsomn S, Urrutia J. The effect of a supportive companion on
perinatal problems, length of labor and mother-infant interaction. The New England Journal
of Medicine. 1980;303:597-600.
25. Sarte CL, Cifra HL. Doula support in unwanted pregnancy-effects on progress and
perceptions of labor. UP-PGH, Johnson & Johnson Clinical Reprints. 1995.
26. Hofmeyer GJ, Nikodem VC, Wolman WL. Companionship to modify the clinical birth
environment: effects on progress and perceptions of labour and breastfeeding. British Journal
of Obstetrics and Gynecology. 1991;98:756-764.
27. Field T, Lasko D, Mundy P, Henteleff T, Talpins S, Dowling M. Autistic children’s
attentiveness and responsitivity improved after touch therapy. Journal of Autism and
Developmental Disorders. 1996;27:329-334.
28. Lester BM, Boukydis CFZ, Garcia-Coll CT, Hole WT. Colic for developmentalists. Infant
Mental Health Journal. 1990;11:4:321-333.
29. Field T, Peck M, Krugman S, et al. Burn injuries benefit from massage therapy. Journal of
Burn Care and Rehabilitation. 1998;19:241-244.
30. Field T, Cullen C, Diego M, et al. Leukemia immune changes following massage therapy.
2001;5:271-274.
31. Ferrel-Torry AT, Glick OJ. The use of therapeutic massage as a nursing intervention to
modify anxiety and the perception of cancer pain. Review Cancer Nursing. 1993;16:93-101.
32. Acolet D, Modi N, Giannakoulopoulos X, et al. Changes in plasma cortisol and

catecholamine concentrations in response to massage in preterm infants. Archives of Diseases
in Childhood. 1993;68:29-31.
33. Van Wyk JJ, Underwood LE. Growth hormone, somatomedins and growth failure. Hospital
Practice. 1978;13:57-67.
34. Unvas-Moberg K, Widstrom AM, Marchini G, Windberg J. Release of GI hormone in

mother’s infants by sensory stimulation. Acta Paediatrica Scandinavica. 1987;76:851-860.
35. Cobarrubias PD. The role of infant massage therapy in stimulating secretion of insulin
and growth hormone among malnourished patients admitted at the intensive care of the
Philippine children’s medical center. Pediatric Capsule. Vol 2 No. 2, January-March 2003,
36-41.
36. Spense JE, Olsem MA. The therapeutic touch 1975-1995; an integrated review of the
literature. Scand J Caring Sci. 1975;11:183-190.
CHAPTER 14:
MASSAGE THERAPY EFFECTS
ON ILLNESS SYMPTOMS
IN INFANTS LIVING IN
ECUADORIAN ORPHANAGES
Vonda K. Jump, PhD

Early Intervention Research Institute
Utah State University
Ab s t ra c t
Studies have documented the positive effects of touch on immune function

in animals. Clinical trials have also confirmed the positive effects of massage
therapy on the health status of children and adolescents who have various illnesses.
This chapter focuses on the health issues of children living in orphanages and
reports the results of a pilot study that compared the effects of massage therapy
versus play sessions on illnesses in infants residing in 2 orphanages in Quito,
Ecuador. Results of this study suggested that infant massage reduced the number
of days children in orphanages exhibited illness symptoms. Infant massage also
had a positive effect on the symptoms they experienced.
Introd uc t i o n
“Evidence suggests that touch is involved in immune system function, although

it is likely that the relationships are indirect,” as was noted by Reite.1 Over the
past few decades, several researchers have investigated the relationship between
touch and immune function.2-6 However, since manipulating the use of touch in
humans would be considered unethical, researchers must rely on animal studies
to inform practice.
Research involving young, nonhuman primates has indicated that being separated
from their mothers has a negative impact on their immune systems.2,3 An increase
in cortisol levels, as well as changes in other behavioral and physiological indicators,
accompanies this separation in nonhuman primates, which, in turn, negatively
affects the immune system.4 Some research has indicated prolonged negative effects
of maternal separation on the immune system of nonhuman primates.5 Worlein
and Laudenslager asserted that interactions between mother figures and their
peers are necessary for normal immunocompetence.4 They also stated that infant
monkeys reared in nurseries have different behavioral and immune tendencies
than young monkeys raised with their mothers.4 Even when infant monkeys were
raised by humans and were allowed to socialize with other monkeys, their immune
systems were compromised, although they did not experience behavioral deficits.6
These findings may transfer to the human population. Infants who are separated
from their parents, such as those living in orphanages where emotional and physical
care are often neglectful, may experience behavioral and health consequences.
Children adopted into the USA often arrive with very incomplete medical histories.
More complete health-status-related information often becomes available about
these children after their arrival into the USA. Johnson and colleagues found
that only 15% of children adopted from Romania were physically healthy and/or
developmentally normal when they were examined in the USA.8 Similarly, Albers
and colleagues reported that many children adopted into the USA from the
former Soviet Union and other eastern European countries were diagnosed with a
variety of conditions, the most common (51%) being intestinal pathogens.9 These
were retrospective studies that occurred after adoptions by families in the USA.
Even when health-status information exists from the native country, Albers and
colleagues stated that it tends to be sporadic and often incorrect.9 Although only
scant information is available regarding the health status of children housed in
Chapter 14: Massage Therapy Effects on Illness Symptoms in Infants 229
orphanages, existing retrospective work7-9 indicates that a substantial number of

infants and young children arrived in the USA with health issues, indicating that
their health was probably compromised while in the orphanage environment.
Since children being raised in overseas orphanages appear to suffer from a variety
of health-related problems, developing low-cost interventions to address this
problem is necessary.
Studies on human immunodeficiency virus (HIV)-positive adolescents10 and

adults11 suggest positive effects of massage therapy on the immune system. This
is probably attributable to decreased cortisol levels. Cortisol affects the immune
system negatively in part by killing immune cells such as natural killer cells.
Massage therapy has also been shown to positively affect the health status of
premature infants,12-14 children with diabetes15 and children with asthma.16 Because
children in overseas orphanages appear to suffer from compromised health, and
massage seems to influence health status positively, my colleagues and I assessed
the effects of massage on the health of infants being raised in these institutions.
We expected that infant massage, as a low-cost intervention, would facilitate better
health in these children while they were still living in the orphanages.
Cur re n t St u d y
METHOD/PARTICIPANTS
In 2 orphanage nurseries in Quito, Ecuador, 37 infants, ranging in age from 10

months to 11 months, were randomly assigned to a massage therapy group or a
play control group. No congenital birth defects were apparent in any of the infants
at the time of assignment. Of these 37 infants, data are presented on 30 infants.
Seven infants at Site 1 were dropped from the study over the course of the study:
Four infants (one “massage therapy” and 3 “play group”) left the nursery to live in
another section of the orphanage, one play group infant was adopted during the
course of the intervention, one massage therapy group infant was reunited with
her family and one massage therapy group infant died due to pneumonia after
assignment to a group but before the intervention began. Demographic data on
the 30 infants remaining in the study are shown in Table 1.
Table 1. Demographic Information on Infants
Group
Massage Therapy Play Control
(n = 14) (n = 16)
Variable Mean SD(±) Mean SD(±) P Value
Age of infants (mo) 10.6 8.2 10.4 5.9 .90
Age at entry to orphanage (mo) 3.1 8.4 2.4 5.9 .77
Infants entering orphanage

10 10
in first 2 weeks of life (n)
Bayley Mental Score (pretest) 67.4 12.7 72.9 13.10 .26
Bayley Motor Score (pretest) 76.4 15.87 81.6 15.09 .36
Percent female 29 63 .07
Infants found abandoned (n) 7 4 .16
INTERVENTION
Infants in the massage therapy group (n = 14) received one, 15-minute massage
daily, usually in the morning, delivered by orphanage volunteers or staff, all of
whom were trained in infant massage techniques endorsed by the American
Association of Infant Massage. Contrastingly, volunteers interacted with infants in
the play control group (n = 16) for 15 minutes daily in individual play sessions.
The first day of the intervention for the infants began after assignment to a group
and the collection of pretest information: As such, these infants began the study on
different days. Orphanage volunteers were instructed to record daily observations,
usually each morning, of all visible symptoms of illness experienced by the infants.
Symptoms noted included the following: upper respiratory symptoms, including
clear-runny or snotty nose, yellow-runny or snotty nose, congestion, cough and

crusty eyes or nose; symptoms of infection, including diarrhea and vomiting; and
general symptoms of illness, including whining or fussiness that was not considered
normal. Symptoms exhibited later in the day were also recorded in each infant’s
file. Although volunteers were provided with thermometers, no fevers were noted
during the first 2 weeks of the study. The orphanage’s nurse indicated that fevers
were quite infrequent in these children. Due to the amount of time and effort
invested in pursuing data on fevers—with little likely return—it was decided to
discontinue recording temperatures.
The average duration of this experimental intervention was 53 days. The massage
therapy group infants received massage more often than the play control group
infants received play sessions during those 53 days (mean, 35.7 days versus 20.4
days, respectively; P<.001). The same was true during Day 15 to Day 45 of the
study period (mean, 20.7 days massage therapy sessions versus 13.7 days play
sessions, respectively; P<.001). This chapter reports data from Day 15 through
Day 45 of the study period. The first 14 days served as a baseline to establish the
typical infant-health status before using infant-illness data to determine whether
massage was correlated positively with a decrease in the incidence of illness.
SITE DESCRIPTIONS
Site 1 was a Catholic orphanage that, at the time, had 31 infants living in the
nursery. Infants were either brought in by their parents or found abandoned and
brought in by the police. The sanitary conditions were optimal, with an emphasis
placed on cleanliness in most areas. The floors in the nursery area, which included
6 rooms and one hallway, were mopped twice daily. Infants’ clothing was washed
on a daily basis, as were their crib sheets; a necessity, since the infants’ cloth diapers
often soiled both clothing and bed sheets. However, the toys that the infants
played with daily were washed on only a monthly basis. Normally, 17 to 26 infants
older than 6 months of age played in one large room where they were supervised
by 3 to 4 volunteers. On days when the volunteers were not in the orphanage
(approximately 1 of every 14 days), infants were often left to play unsupervised
in the large room, although nuns and orphanage staff were in the other rooms
in the nursery area. Infants younger than 6 months of age were kept in another
room and were taken out of their cribs for diaper changes and baths, but not for
individual stimulation.
Site 2 was a private orphanage that, at the time, had 8 infants and young children
in its care. Infants in this orphanage were brought in by their parents. Sanitary
conditions were similar to those found at Site 1. Normally, 2 volunteers supervised
6 to 8 infants and young children during their “playtime.” On days that volunteers
were not present, the children over the age of 6 months typically played by
themselves, although at least 2 staff members were always present. Similar to Site 1,
infants less than 6 months of age were normally taken out of their cribs for diaper
changes and baths, but not for individual stimulation.
Result s
Although both groups of infants experienced similar play or crib-based conditions

on a daily basis, children in the massage therapy group experienced more days
of wellness and fewer days with symptoms of illness than did infants in the play
control group. These results were based on data recorded on Day 15 to Day 45 of
the study (Table 2).
Table 2. Number of Days With and Without Symptoms of Illness
Group
Data From Massage Therapy Play Control
Day 15 to Day 45 (n = 14) (n = 16)
of Intervention Mean SD(±) Mean SD(±) Effect Size P Value
Days without symptoms (n) 7.2 6.3 3.9 3.3 1.00 .08
Days data were missing (n) 8.4 5.8 8.3 5.4 .01 .98
For ease of understanding, individual symptoms of illness have been condensed

into categories similar to those used by physicians. Infants in the massage therapy
group had a lower mean incidence of infection symptoms and general fussiness
(Table 3). When individual symptoms were analyzed, the massage therapy
group infants experienced diarrhea less frequently and exhibited a more positive
temperament than the play control group infants.
Table 3. Symptoms of Illness Experienced by Massage Therapy and Play Control

Groups From Day 15 to Day 45
Group
Symptoms Experienced Massage Therapy Play Control
From Day 15 to Day 45 (n = 14) (n = 16)
of Intervention Mean SD(±) Mean SD(±) Effect Size P Value
Upper Respiratory Tract

16.0 12.1 18.1 7.9 .27 .58
Symptoms
Clear-runny or snotty nose 7.4 5.3 9.9 5.6 .45 .22
Yellow/green-runny
1.8 2.1 1.1 1.5 .48 .28
or snotty nose
Congestion 0.9 1.1 1.5 1.2 .50 .19
Coughing 0.6 1.2 1.2 1.4 .40 .25
Dried mucous on nose 3.1 2.8 3.6 3.6 .16 .65
Crusty eyes 2.2 4.3 0.9 1.4 .98 .28
Symptoms of Infection 5.3 4.6 8.4 6.2 .50 .14
Diarrhea 4.6 3.9 7.5 5.5 .53 .11
Vomiting 0.7 1.6 0.9 1.4 .12 .78
Whiny or Fussier Than Normal 0.8 1.2 1.8 2.3 .41 .16
Although these findings were not statistically significant at the P<.05 level, they
did approach statistical significance. Given that this was a pilot study on a small
sample, my colleagues and I opted to use “effect sizes” to determine the practical
significance of the results. Based on findings reported by Cohen, an effect size of
.40 or higher is considered indicative of “practical significance.”17 As can be seen
in Table 3, most effect sizes (8 of 11) exceeded this benchmark.
Im pli c a t i o n s
Results of this pilot project suggest that infant massage may reduce both the
number of days that children in orphanages experience symptoms of illness and the
types of symptoms they experience. The finding that infants in the massage therapy
group averaged more days with no symptoms of illness, despite being in the same
environment as infants in the play control group, warrants further research on the
effects of massage on decreasing infants’ susceptibility to illness. Data gathered on
the levels of urinary cortisol remain to be analyzed. Because there is an established
link between elevated levels of cortisol and compromised immune response,
infants who experienced fewer days of the symptoms of illness are likely to have
experienced decreased levels of cortisol over time.
With respect to the massage therapy group infants experiencing more yellow/green
runny noses and more incidents of crusty eyes (Table 3), a pediatrician colleague
(Odell, personal communication) indicated that this, in fact, could have been
an indication of the babies getting the infection or illness out of their bodies.
However, it would also be imperative to know whether the infants in the massage
therapy group experienced a shorter duration of illness each time. It was not
possible to conduct this analysis during this study because of the large number of
days (mean, 8.4 days) for which we had no data during the course of the illnesses
(Table 2). Future similar studies would need to have more reliable symptom
reporting to allow for this type of analysis.
Another interesting finding was that infants in the massage therapy group
experienced diarrhea less frequently than did infants in the play control group.
Annually, approximately 2 million children worldwide under the age of 5 die as
a result of diarrhea and associated dehydration.18 However, because this study was
a pilot project that enrolled relatively few infants, no conclusions can be drawn.
Nevertheless, the effect sizes obtained indicate that further research is needed on
the effectiveness of massage in preventing or ameliorating diarrhea.
My colleagues and I are planning to replicate this work in other orphanages, with
larger numbers of children, to better understand the effects of massage on children’s
immediate and long-term health outcomes. Several study improvements are
necessary, including stricter adherence to the intervention protocol and to daily
symptoms recording. Only by designing and following strict research protocols

will we, as researchers, be able to disentangle the many factors that might impact
the effects of massage on health outcomes of children living in orphanages.
Ackn owledgments
This project was made possible through a research grant awarded by the American
Massage Therapy Association Foundation to Vonda K. Jump, PhD, Senior Research
Associate at the Early Intervention Research Institute at Utah State University,
Logan, Utah. Special thanks go to the Ecuadorian orphanage staffs, the Orphanage
Support Services Organization and other volunteers, as well as to the babies for
allowing me to work with them. Thanks to Bianca for teaching me about massage.
Ref ere n c e s
1. Reite M. Effects of touch on the immune system. In: Gunzenhauser N, ed.; Advances in
Touch: New Implications in Human Development. USA: Johnson & Johnson Consumer
Products, Inc.; 1990.
2. Laudenslager ML, Reite M, Harbeck RJ. Suppressed immune response in infant monkeys
associated with maternal separation. Behavioral and Neural Biology. 1982;36:40-48.
3. Reite M, Harbeck R, Hoffman A. Altered cellular immune response following peer
separation. Life Sciences. 1981;29:1333-1336.
4. Worlein JM, Laudenslager ML. Effects of early rearing experiences and social interactions
on immune function in nonhuman primates. In: Ader R, Felten DL, Cohen N, eds.
Psychoneuroimmunology. Vol 1. 3rd ed. Boston, Mass.: Academic Press; 2001:73-85.
5. Laudenslager ML, Capitanio JP, Reite ML. Possible effects of early separation experiences on
subsequent immune function in adult macaque monkeys. The American Journal of Psychiatry.
1985;142:862-864.
6. Worlein JM, Sackett, GP. Social development in nursery-reared pigtailed macaques (Macaca
nemestrina). American Journal of Primatology. 1997;41:23-35.
7. Hostetter MK, Iverson S, Thomas W, McKenzie D, Dole K, Johnson DE. Medical

evaluation of internationally adopted children. The New England Journal of Medicine.
1991;325:479-485.
8. Johnson DE, Miller LC, Iverson S, et al. The health of children adopted from Romania.
JAMA. 1992;268:3446-3451.
9. Albers LH, Johnson DE, Hostetter MK, Iverson S, Miller LC. Health of children adopted
from the former Soviet Union and Eastern Europe: comparison with preadoptive medical
records. JAMA. 1997;278:922-924.
10. Diego MA, Hernandez-Reif M, Field T, Friedman L, Shaw K. HIV adolescents show
improved immune function following massage therapy. The International Journal of
11. Ironson G, Field T, Scafidi F, et al. Massage therapy is associated with enhancement of
the immune system’s cytotoxic capacity. The International Journal of Neuroscience.
1996;84:205-218.
12. Field T, Schanberg SM, Scafidi F, et al. Tactile/kinesthetic stimulation effects on preterm
13. Jinon S. The effect of infant massage on growth of the preterm infant. In: Yarbes-Almirante
C, De Luma M, eds. Increasing Safe and Successful Pregnancy. Amsterdam, Netherlands:
Elsevier Science; 1996:265-269.
14. Scafidi FA, Field T, Schanberg SM, et al. Massage stimulates growth in preterm infants:
a replication. Infant Behavior and Development. 1990;13:167-188.
15. Field T. Infant massage therapy. In: Field T, ed. Touch in Early Development. Hillsdale, NJ:
Lawrence Erlbaum Associates; 1995:105-114.
pulmonary function after massage therapy. The Journal of Pediatrics. 1998;132:854-858.
17. Cohen J. Things I have learned (so far). The American Psychologist. 1990;45:1304-1312.
18. World Health Organization Web site. Available at http://www.who.int/inf-fs/en/fact178.html.

Accessed July 13, 2002.
CHAPTER 15:
MASSAGE THERAPY
F O R P E D I AT R I C P RO B L E M S
Tiffany Field, PhD

Miguel Diego, MA
Abstract
Numerous studies have reported the benefits associated with massage therapy
in children who have various illnesses and conditions. In this chapter studies
are reviewed demonstrating improvements in children with medical and
psychiatric disorders. The benefits of massage therapy have also extended to
those who performed the massages. Conditions discussed in this chapter include
attention-deficit/hyperactivity disorder, autism, posttraumatic stress disorder,
depression, anorexia, bulimia, cerebral palsy, Down syndrome, various
autoimmune disorders, including asthma and diabetes, and immune disorders,
including human immunodeficiency virus and leukemia. Potential underlying
mechanisms that may contribute to the effects of massage are also discussed.
Atten t i o n a n d Be h a v i o r D i s o rd e r s
ATTENTION-DEFICIT/HYPERACTIVITY DISORDER
Attention-deficit/hyperactivity disorder (ADHD) affects as many as 3% to 6%

of children and is characterized by inattention, impulsiveness and hyperactivity.1
Overactivity is typically the most prominent feature. Although drug therapy has
improved the symptoms of ADHD in more than 75% of cases, medications often
have undesirable side effects.1
In a study on 28 adolescents with ADHD, each participant was provided with

either massage therapy or relaxation therapy for 10 consecutive school days.1
At study completion, those in the massage therapy group, but not those in the
relaxation therapy group, rated themselves as “happier,” and observers rated them
as “fidgeting less” following the sessions. In addition, those in the massage
therapy group were reported by their teachers as spending more time on assigned
tasks, and they assigned these children lower hyperactivity scores based on current
classroom behavior.
AUTISM
Autism affects 2 to 5 of every 10,000 children. Autism takes many forms with
several different degrees of severity and its causes are still unknown.2 Symptoms
of autism include limited language, social communication and ability to develop
relationships. Also children with autism show abnormal responses to sensory
stimuli (usually sound), atypical movements including immobility and
hyperactivity, a limited attention span and excessive off-task behaviors and an
aversion to touch.
A variety of therapies have been tried in children with autism, including behavior
modification and sensory-integrative approaches. In a recent study we reported
that children with autism who were provided with massage therapy showed
less inattentiveness (off-task behavior), reduced touch aversion and decreased
withdrawal (Figures 1 and 2).2
Chapter 15: Massage Therapy for Pediatric Problems 241
Figure 1. Benefits of massage

therapy in children with attention
and behavior disorders.1
Figure 2. Benefits of massage therapy in off-task behavior in children with autism.2
In another study we conducted,3 20 children with autism were given daily massages
by their parents. Ranging in age from 3 years to 6 years, these children were
randomly assigned to massage therapy and reading attention control groups. A
massage therapist trained the parents in the massage therapy group. Every night for
1 month, these parents massaged their children for 15 minutes prior to bedtime,
while the parents in the attention control group read stories by Dr. Seuss to their
children on the same time schedule. Sleep, which is often disturbed in children
with autism, was monitored. Teacher- and parent-rated scales, classroom and
playground observations and sleep diaries were used to assess the effects of massage
therapy on various behaviors, including hyperactivity, stereotypical and off-task
behaviors and sleep problems. The children in the massage therapy group showed
fewer stereotypical behaviors and more on-task and social-relatedness behaviors
during classroom observations. In addition, the massaged children experienced
fewer sleep problems at home.
POSTTRAUMATIC STRESS DISORDER IN

CHILDREN FOLLOWING HURRICANE ANDREW
Past studies have described posttraumatic responses in children that are similar
to those described for adults with the diagnosis of posttraumatic stress disorder
(PTSD).4 Symptoms include reduced responsiveness, increased arousal and conduct
problems that persist for several months following the traumatic event.
In a study we conducted following hurricane Andrew in Miami, Florida,

60 grade-school children who showed classroom behavior problems were given
massage therapy for 8 consecutive days.4 Scores on the PTSD Reaction Index
suggested that these 60 children were experiencing severe posttraumatic stress.
When compared with a video attention control group, the children who received
massage therapy reported being happier and less anxious and had lower saliva
cortisol (stress hormone) levels posttherapy. In addition, the children in the
massage therapy group showed more sustained changes, as manifested by
lower scores on the Children’s Manifest Anxiety Scale and The Center for
Epidemiological Studies Depression Scale, as well as in self-drawings. They were
also observed to be more relaxed. These positive results are quite promising, given
the persistence of the symptoms of PTSD for children who had not received
interventions following disasters such as hurricanes.
CHILD AND ADOLESCENT PSYCHIATRIC PATIENTS
Depression and adjustment disorder are often accompanied by anxiety, increased

stress hormones and sleep disturbances. In a past study we determined the effects
of massage therapy on child and adolescent psychiatric patients.5 A 30-minute
back massage was given daily for a 5-day period to 52 hospitalized children and
adolescents with depression and/or adjustment disorder.5 Compared with a control
group who viewed relaxing videotapes, the massaged subjects were less depressed
and less anxious and had lower saliva cortisol levels after the massage. In addition,
nurses rated the massaged children as being more cooperative on the last day of
the study, and nighttime sleep increased over this 5-day period. Lastly, cortisol and
norepinephrine levels were decreased.
ANOREXIA
Anorexia, a body weight disorder affecting women almost exclusively, has increased
in incidence over the past 40 years.6 Symptoms of anorexia are extreme thinness
and poor body image. In a recent study we observed adolescent women with
anorexia who received massage therapy twice per week for 5 weeks and those
who received standard treatment alone.6 The women who received massage
reported lower stress levels and anxiety levels and had lower cortisol levels. Over
the 5-week treatment period, these women also reported improved body image
on the Eating Disorder Inventory and they experienced increased dopamine and
norepinephrine levels.
BULIMIA
Bulimia is characterized by recurrent episodes of binge eating; self-induced

vomiting, the use of laxatives or diuretics and strict dieting; and concern with body
shape and weight. In an earlier massage therapy study, 24 adolescents with bulimia
were randomly assigned to massage therapy or standard treatment control groups.7
Results indicated that the women who received massage showed immediate
reductions in anxiety and depression, based on self-reports and observed behaviors.
In addition, by the last day of therapy, the women had lower depression scores,
lower cortisol levels and higher dopamine levels. These findings suggest that
massage therapy is effective for adolescent women with bulimia.
AGGRESSIVE/VIOLENT ADOLESCENTS
In 1998 the Centers for Disease Control and Prevention published statistics
showing that the violence rate (ie, homicide rate) among adolescent and young
adult males in the US exceeded that of any other industrialized country. Excessive
violence has been noted in adolescents and adults in primitive countries where
young children receive minimal affection. Touch taboos and mandates not to touch
children in school may be contributing to the high rates of aggression/violence in
the US.
In a study on violent adolescents we attempted to reduce aggression by providing

them massage therapy.8 Seventeen aggressive adolescents were assigned randomly
to a massage therapy group or a relaxation therapy group to receive 20-minute
therapy sessions, twice a week for 5 weeks. The adolescents who received massage
therapy showed lower levels of anxiety after the first and last therapy sessions. By
the end of the study, the massaged adolescents also reported feeling less aggressive
and were perceived by their parents as being less aggressive than the adolescents
who were assigned to the relaxation therapy group (Figure 3).
Figure 3. Benefits of 5 weeks of massage therapy in aggressive adolescents.8
POTENTIAL UNDERLYING MECHANISMS
Depressed mood, anxiety levels and stress hormones (norepinephrine, epinephrine

and cortisol) were reduced in all of the studies discussed.1-9 A potential mechanism
that might explain these changes may be the increase noted in vagal activity
following massage therapy.9 The nucleus ambiguous branch of the vagus (the
“smart” vagus) stimulates facial expressions and vocalizations. This contributes
to less depressed affect, which, in turn, could feed back to produce less depressed
feelings.10 Increased vagal activity may also account for the enhanced attentiveness
in the children with ADHD1 and the children with autism.2
Mus c ulo ske l e t a l Pro b l e m s
CEREBRAL PALSY
Symptoms of cerebral palsy (CP) in children include spasticity (rigidity of muscles),

impaired motor coordination and deficits in sitting, standing, locomotion and daily
living skills. In addition, children with CP may be impaired cognitively, socially
and emotionally. As a result of the severity of the disorder, multidisciplinary
treatments are often prescribed.
In a study we conducted, 20 preschool-age children with CP were recruited from

early-intervention programs to receive 30 minutes of massage twice weekly for
12 weeks or participate in a control group who heard stories.11 At the end of the
study the children who received massage therapy showed reduced spasticity, and
improved fine and gross motor functioning. In addition, the children in the
massage therapy group received higher cognition, social and dressing scores on the
Developmental Profile, and they showed more positive facial expressions and less
disorganized limb movements during face-to-face play interactions. These findings
suggest that massage therapy attenuates the physical symptoms associated with CP,
it enhances development and it should be considered as an early intervention for
children with this disorder.
DOWN SYNDROME
Down syndrome (DS), a genetic disorder affecting 1 in approximately 700 children

born in the US,12 is characterized by cognitive deficits, speech problems and
motor- and perceptual-development delays. Children with DS also frequently
exhibit decreased muscle tone, or hypotonia. Recent research findings from studies
of non-DS children have shown that massage therapy may improve muscle tone
and enhance motor development.12
Twenty-one moderate- to high-functioning preschool-age children with DS already

receiving early intervention (physical therapy, occupational therapy and speech
therapy) were randomly assigned to receive 30-minute massage therapy or reading
sessions (control group) twice per week for 2 months.12 Based on last-day versus
first-day assessment measures, the children in the massage therapy group versus
the control group showed developmental gains in fine and gross motor functioning
and less severe hypotonicity in their upper limbs.
Au toi mmun e Pro b l e m s
ASTHMA
Complementary forms of therapy are being explored for children with asthma: For
example, a study of facial-relaxation training documented increases in immediate
peak expiratory flow rates in children.13 However, these increases fell short of the
standard criterion for clinical significance used in evaluating asthma medications—
a 15% increase in air-flow rates (Figure 4).
Figure 4. Standard criterion for clinical significance

used in evaluating asthma medications is a 15% increase
in air-flow rate.13
We evaluated massage therapy with children who have asthma because it has
lowered anxiety and cortisol levels in children who have other problems.14 Also,
massage therapy requires less compliance from children. Another reason for the
effectiveness of massage therapy is that the children’s parents provide the therapy.
Parents of children who have asthma have been shown to experience higher anxiety
levels compared with parents of children who do not have asthma—and increased
anxiety on the part of the parents can affect the children’s health.14 Giving parents
an active coping role in their child’s therapy may reduce their own anxiety. In this
study, 32 children with asthma were randomly assigned to receive either massage
therapy or relaxation therapy (control group).14 The children’s parents were taught
to provide the therapy, which was given for 20 minutes before bedtime each
night for 30 days. The children who received massage therapy showed immediate
decreases in anxiety and cortisol levels. Also, their attitudes toward asthma and
their peak air flow and other pulmonary functions improved over the course of
the study (Figure 5).
Figure 5. Effect on peak air-flow rates in children with asthma following 30 days
of massage therapy given by their parents.14
320
309
Peak Air Flow
300
279
280
260
First Day Last Day
CYSTIC FIBROSIS
Cystic fibrosis (CF) is a chronic, life-threatening disease diagnosed typically in

early childhood. The prevalence of CF is approximately 1 in every 3300 live births,
making it the most common genetic disease in the US.15,16 In a recent study we
assessed the effects of parents giving massage therapy to their children who had
CF.17 This was done to reduce anxiety in the parents and their children as well as
to improve the children’s moods and peak air flow readings. Twenty children with
CF and their parents were assigned randomly to a massage therapy or a reading
control group. Parents in the treatment group were asked to give their children a
20-minute massage every night at bedtime for 1 month. Parents in the control
group were requested to read to their children for the same amount of time.
On day 1 and day 30, parents and children answered questions related to their
anxiety levels, and children answered questions related to their mood. In addition,
peak air flow readings were measured in the children. Following the first and last
massage sessions, the children and parents reported lower anxiety levels. Mood and
peak air flow readings also improved for children in the massage therapy group.
These findings suggest that the parents of children with CF may reduce their
anxiety levels by massaging their children, and that the children may benefit from
receiving the massage by having less anxiety and improved mood, which, in turn,
may facilitate breathing.
DIABETES
Parents’ involvement in the treatment of their children who have diabetes can be
a negative experience; for example, when monitoring dietary compliance, taking
blood samples and giving insulin injections. We tried to give parents of children
with diabetes a more positive role in their children’s treatment by being called on to
massage their children daily before bedtime for 1 month (Figure 6).18 Immediately
after each massage therapy session, anxiety and depression levels were lower in
both the parents and their children. At the end of the 1-month study period, the
children’s insulin and food-regulation scores improved, and the children’s blood
glucose levels decreased to the normal range (158 mg/dL to 118 mg/dL; Figure 7).
Figure 6. Massage therapy offers parents a more

positive role in the treatment of their child’s
diabetes.18
Figure 7. Effects on blood glucose levels in children with diabetes following 1 month of
nightly massage therapy provided by their parents.18
ATOPIC DERMATITIS
Atopic dermatitis (eczema) has been related to increased depression,19 stress and
anxiety,20 which, in turn, lead to increased cortisol levels that then dampen
immune function.21 In a study we conducted, children with atopic dermatitis were
treated with standard topical care and massaged by their parents for 20 minutes
daily for 1 month.22 A control group received standard topical care only. The
children’s affect and activity levels improved significantly, and their parents’ anxiety
levels decreased immediately after finishing each massage therapy session. Over the
span of the 1-month study period, parents of the massaged children reported lower
anxiety levels in their children, and the children improved significantly on all
clinical measures regarding their skin, including redness, scaling, lichenification
(violet patches caused by scratching), excoriation (removal of skin caused by
scratching) and pruritus (itching).
JUVENILE RHEUMATOID ARTHRITIS
Juvenile rheumatoid arthritis (JRA), one of the most common chronic diseases
of childhood, is the most common rheumatic disease of this period of life.23 The
diagnosis of JRA is based largely on the observation of persistent arthritis (6 weeks
or more in duration) in one or more of the child’s joints. This disease manifests
itself typically before 16 years of age, with peak onset in the following age groups:
1- to 3-year-old children and 8- to 12-year-old adolescents.24 Common symptoms
of JRA include night pain and joint stiffness in the morning and following long
periods of inactivity.
In a study we conducted, children with mild-to-moderate JRA were massaged by

their parents for 15 minutes each day for 30 days.23 A control group engaged in
relaxation therapy. The children’s anxiety and cortisol levels decreased immediately
following the massage, and, over the 30-day period, their pain levels “decreased”
according to self-reports, parent reports and their physicians’ assessments (both
the incidence and severity) of their pain and pain-limiting activities.
POSSIBLE MECHANISMS
Autoimmune problems may occur because of an imbalance in the type-1/type-2

immune systems. Regulation by the type-2 system is thought to lead to
autoimmune conditions, such as asthma.23 By decreasing the levels of stress
hormones following massage, control by the type-2 system may be “switched”
to control by the type-1 system.
Im mun e Pro b le m s
HUMAN IMMUNODEFICIENCY VIRUS
Human immunodeficiency virus (HIV) is a complex retrovirus that affects the

human immune system by attacking CD4+ T cells primarily. Despite the fatal
prognosis following the contraction of acquired immune deficiency syndrome
(AIDS), several treatments are currently being investigated to help enhance
immune function. The treatments researched most popularly for HIV-infected
individuals include pharmacological and psychological interventions.
Pharmacological agents, consisting typically of a combination of antiretroviral
drugs, have proven expensive and difficult to adhere to and are often accompanied
by severe adverse effects. Psychological interventions have been administered
to reduce the levels of anxiety and depression, factors associated with decreased
quality of life and diminished immune function.
Preliminary, unpublished data from an Office of Alternative Medicine-sponsored

study reveal that up to 60% of those receiving alternative treatments for HIV are
receiving massage therapy. In a study we conducted, HIV-positive adolescents were
recruited from the outpatient clinic of a large urban university hospital and were
randomly assigned to receive massage therapy or progressive muscle relaxation twice
weekly for 12 weeks.25 To evaluate the effects of treatment, the adolescents were
assessed for depression, anxiety and immune changes before and after the 12 weeks
of treatment.
The adolescents who received massage therapy, versus those who experienced
relaxation therapy, reported feeling less anxious. In addition, they were less
depressed and showed enhanced immune function by the end of the study.
Changes in immune function included increased numbers of natural killer cells
(cells that kill viral cells). In addition, the cells that are normally killed by the
HIV virus increased for the massage therapy group only.
LEUKEMIA
The outlook for children who have acute leukemia has improved steadily, so that
now at least 60% to 70% of children with leukemia achieve long-term cure and
survival.26 We postulated that massage therapy would be indicated for children
with leukemia because immune function (eg, natural killer cell number and natural
killer cell activity) has been enhanced following massage therapy in HIV-positive
adolescents, as discussed previously.25
Accordingly, we conducted a study on massage therapy given to children with

leukemia.26 Massage was performed daily by the children’s parents. Children who
received massage therapy were compared with a standard treatment control group.
Following 1 month of massage therapy, levels of depression decreased in the
children’s parents, and the children’s white blood cell and neutrophil counts
increased.
POSSIBLE MECHANISMS
Stress hormones (eg, cortisol) have been noted to kill immune cells, particularly
natural killer cells (cells that kill viral and cancer cells).25 Since massage therapy
increases vagal activity, which, in turn, decreases cortisol, immune cells can survive.
Su mm a r y
We have documented clear improvements in several medical and psychiatric

conditions—including depression and eating disorders, musculoskeletal problems,
pain syndromes and immune and autoimmune conditions—following the
application of massage therapy. Although several potential underlying mechanisms
have been explored to explain the improvements seen with massage therapy,
including decreased stress (and decreased cortisol) and enhanced immune function,
further research is needed in this area.
Acknowledgments
We would like to thank the children, adolescents and parents who participated in
these studies. This research was supported by a National Institute of Mental Health
(NIMH) Senior Research Scientist Award (#MH00331) and an NIMH Merit
Award (#MH46586) given to Tiffany Field, as well as by funding provided by the
Johnson & Johnson Pediatric Institute, L.L.C. to the Touch Research Institutes.
Re f e re n c e s
1. Field TM, Quintino O, Hernandez-Reif M, Koslovsky G. Adolescents with attention deficit
hyperactivity disorder benefit from massage therapy. Adolescence. 1998;33:103-108.
2. Field T, Lasko D, Mundy P, et al. Brief report: autistic children’s attentiveness and
responsivity improve after touch therapy. Journal of Autism and Developmental Disorders.
1997;27:333-338.
3. Escalona A, Field T, Singer-Strunck R, Cullen C, Hartshorn K. Brief report: improvements
in the behavior of children with autism following massage therapy. Journal of Autism and
Developmental Disorders. 2001;31:513-516.
4. Field T, Seligman S, Scafidi F, Schanberg S. Alleviating posttraumatic stress in children

following hurricane Andrew. Journal of Applied Developmental Psychology. 1996;17:37-50.
5. Field T, Morrow C, Valdeon C, Larson S, Kuhn C, Schanberg S. Massage reduces anxiety
in child and adolescent psychiatric patients. Journal of the American Academy of Child and
Adolescent Psychiatry. 1992;31:125-131.
6. Hart S, Field T, Hernandez-Reif M, et al. Anorexia symptoms are reduced by massage
therapy. Eating Disorders. 2001;9:289-299.
7. Field T, Schanberg S, Kuhn C, et al. Bulimic adolescents benefit from massage therapy.
Adolescence. 1998;33:555-563.
8. Diego MA, Field T, Hernandez-Reif M, et al. Aggressive adolescents benefit from massage
therapy. Adolescence. 2002;37:597-607.
10. Porges SW. Emotion: an evolutionary by-product of the neural regulation of the autonomic
nervous system. Annals of the New York Academy of Sciences. 1997;807:62-77.
11. Hernandez-Reif M, Field T, Largie S, et al. Cerebral palsy symptoms in children decreased
following massage therapy. Early Child Care and Development. In review.
12. Hernandez-Reif M, Field T, Largie S, et al. Children with Down syndrome improved in
motor function and muscle tone following massage therapy. Early Child Care and
Development. In review.
13. Kotses H, Hindi-Alexander M, Creer TL. A reinterpretation of psychologically induced
airways changes. The Journal of Asthma: Official Journal of the Association for the Care of
Asthma. 1989;26:53-63.
pulmonary functions after massage therapy. The Journal of Pediatrics. 1998;132:854-858.
15. Cystic Fibrosis Foundation. About cystic fibrosis: What is CF. Available at:
http://www.cff.org/about_cf/what_is_cf.cfm?CFID=1754499&CFTOKEN=98980601.
Accessed February 11, 2004.
16. Boucher RC. Cystic fibrosis. In: Isselbachner KJ, Martin JB, Braunwald E, Fauci AS, Wilson
JD, eds. Harrison’s Principles of Internal Medicine. 13th ed. New York, NY: McGraw-Hill;
1994:1194-1197.
17. Hernandez-Reif M, Field T, Krasnegor J, Martinez E, Schwartzman M, Mavunda K.
Children with cystic fibrosis benefit from massage therapy. Journal of Pediatric Psychology.
1999;24:175-181.
18. Field T, Hernandez-Reif M, LaGreca A, Shaw K, Schanberg S, Kuhn C. Massage therapy

lowers blood glucose levels in children with diabetes. Diabetes Spectrum. 1997;10:237-239.
19. Gupta MA, Gupta AK, Schork NJ, Ellis CN. Depression modulates pruritus perception:
a study of pruritus in psoriasis, atopic dermatitis, and chronic idiopathic urticaria.
Psychosomatic Medicine. 1994;56:36-40.
20. Horne DJ, White AE, Varigos GA. A preliminary study of psychological therapy in the
management of atopic eczema. The British Journal of Medical Psychology. 1989;62:241-248.
21. Engels WD. Dermatologic disorders. Psychosomatics. 1982;23:1209-1211, 1214-1219.
22. Schachner L, Field T, Hernandez-Reif M, Duarte AM, Krasnegor J. Atopic dermatitis
symptoms decreased in children following massage therapy. Pediatric Dermatology.
1998;15:390-395.
23. Field T, Hernandez-Reif M, Seligman S, et al. Juvenile rheumatoid arthritis: benefits from
massage therapy. Journal of Pediatric Psychology. 1997;22:607-617.
24. Varni JW, Jay SM. Biobehavioral factors in juvenile rheumatoid arthritis: implications for
research and practice. Clinical Psychology Review. 1984;4:543-560.
25. Diego MA, Field T, Hernandez-Reif M, Shaw K, Friedman L, Ironson G. HIV adolescents
show improved immune function following massage therapy. The International Journal of
26. Field T, Cullen C, Diego M, et al. Leukemia immune changes following massage therapy.
Journal of Bodywork and Movement Therapies. 2001;5:271-274.
Call 1-877-JNJ-LINK (565-5465)
Fax 1-877-JNJ-FAXX (565-3299)
Outside the USA

Call 1-631-208-9238
Fax 1-631-208-9253
Visit us at www.JJPI.com
TCH03-02

Touch

Uploaded by

Copyright:

Available Formats

Touch

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Touch

Uploaded by

Copyright:

Available Formats

touch

Tiffany Field, PhD

Johnson & Johnson Pediatric Institute, L.L.C.

©Johnson & Johnson Pediatric Institute, L.L.C. 2004

Tiffany Field, PhD

Published by: Johnson & Johnson Pediatric Institute, L.L.C.

Chapter 1: Fetal Responses to Foot and Hand Massage of Pregnant Women . . . . . . . . 3

Chapter 2: Touch Perception in Neonates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

Chapter 3: Maternal Touch Effects on Infant Behavior . . . . . . . . . . . . . . . . . . . . . . . . . 39

Chapter 4: Touching During Mother-Infant Interactions . . . . . . . . . . . . . . . . . . . . . . . . 49

Chapter 5: Differences Across Cultural Groups in Mothers’ Nonverbal

Chapter 6: Pregnancy, Labor and Infant Massage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99

Chapter 7: Skin-to-Skin Care for Breastfeeding Difficulties Postbirth . . . . . . . . . . . . 115

Chapter 8: Tactile Stimulation of Neonatal Intensive Care Unit

Chapter 9: Preterm and Full-term Infant Massage in China . . . . . . . . . . . . . . . . . . . . . 163

Chapter 10: Optimizing Growth and Bone Mass in Premature Infants:

Chapter 12: Massage, Relaxation and Well-Being: A Possible Role for

Chapter 13: Massage Therapy With Preterm Infants and Children

Chapter 14: Massage Therapy Effects on Illness Symptoms in Infants

Chapter 15: Massage Therapy for Pediatric Problems . . . . . . . . . . . . . . . . . . . . . . . . . 239

Gene Cranston Anderson, PhD, FAAN Herminia L. Cifra, MD, MSc,

Tiffany Field, PhD Vonda K. Jump, PhD

Lynda Law Harrison, RN, PhD, FAAN Mary Martini, PhD

Barbara Morrison, PhD, FNP, CNM

Laurie J. Moyer-Mileur, PhD, Kerstin Uvnäs-Moberg, MD

Hiselgis Perez, PhD Zhu Jianxing, MD, MSc

Dale M. Stack, PhD

In the midst of this exciting progress, however, the importance of old-fashioned

• Even a few minutes a day of touching measurably increases a

• During skin-to-skin contact, a mother’s chest naturally adjusts to

• Touch and massage significantly reduce infant crying and distress

• Infant massage under accepted guidelines can increase weight gain

and much more.

Johnson & Johnson Pediatric Institute, L.L.C.

interactions are addressed by Stack, and data on early interactions in different

In other chapters, Field covers therapeutic applications including pregnancy and

Tiffany Field, PhD

FOOT AND HAND MASSAGE

From a prenatal clinic at a large, urban, university hospital, 80 pregnant women in

Figure 1. Ultrasound image of the fetus.

The 3-minute foot-massage procedure was performed by the therapist as follows:

Figure 2. Hand massage procedure.

Results and Discussion

that massaging highly innervated areas produces increased uterine contractions,26,27

While the hand-massage procedure elicited a 10% increase in fetal movement,

Additional research is needed to determine whether massage therapy given

24. Cardini F, Weixin H. Moxibustion for correction of breech presentation: a randomized

Maria Hernandez-Reif, PhD

• These behaviors are exercised in utero, beginning during the second

• Skin or sense receptors are in place and are functional prenatally, so

The Touch “Organs”

The exploration or manipulation of objects with fingers, hands or mouth is defined

Traditionally, the behavioral repertoire of the neonate has been described as

What follows is a detailed review of the research of these behaviors exhibited by

In a follow-up study (1987), Rochat reported that 2- to 3-day-old newborns

Studies of young infants’ cross-modal (touch-to-sight) abilities revealed that

Newborns’ mouthing responses to objects that varied in texture30 or temperature31

• Interestingly, the infants of depressed and nondepressed mothers sucked

• From sucking, neonates discriminated a smooth texture from a nubby