Nutritional Screening Tools for

Malnutrition in Pediatrics

Gal Rub, Luba Marderfeld, and Raanan Shamir

Abstract
Malnutrition in pediatrics remains a cause for concern due to its considerably high
prevalence and deleterious effects on growth, development, and overall health.
Early identification of malnutrition risk may prevent nutritional deterioration
during hospitalization. There are currently a number of suggested screening
tools for use in pediatrics; however, there is no consensus on a single tool that
is favorable over others. Thus selection of a screening tool for implementation is
perplexing. Presented here is an overview of the screening tools available for use
in pediatrics and further actions needed in order to implement the use of screening
tools in different settings in pediatrics. Seven screening tools that are intended for
use upon admission to the hospital were identified. Two screening tools were
designed for specific medical conditions. One screening tool was designed for
sole use in ambulatory settings. Of the seven tools identified for use upon hospital

G. Rub
Institute of Gastroenterology, Nutrition and Liver Diseases, Schneider Children’s Medical Center of
Israel, Petah Tikva, Israel
e-mail: galarub@gmail.com
L. Marderfeld
Institute of Gastroenterology, Nutrition and Liver Diseases, Schneider Children’s Medical Center of
Israel, Petah Tikva, Israel
Clinical Nutrition and Dietetics Department, Institute of Gastroenterology, Nutrition and Liver
Diseases, Schneider Children’s Medical Center of Israel, Petah Tikva, Israel
e-mail: Lu.marderfeld@yahoo.com
R. Shamir (*)
Institute of Gastroenterology, Nutrition and Liver Diseases, Schneider Children’s Medical Center of
Israel, Petah Tikva, Israel
Sackler Faculty of Medicine, Tel-Aviv University, Tel - Aviv, Israel
e-mail: president@espghan.org; shamirraanan@gmail.com

# Springer International Publishing AG 2017 1

V.R. Preedy, V.B. Patel (eds.), Handbook of Famine, Starvation, and Nutrient
Deprivation, DOI 10.1007/978-3-319-40007-5_66-1
2 G. Rub et al.

admission, some tools were also validated for use in specific medical conditions
and one tool was also validated for use in ambulatory settings. Comparison
between screening tools failed to offer one tool favorable to others. A model
for implementation of nutritional screening in pediatrics in terms of policy change
is suggested. In conclusion, there is currently no single nutritional screening tool
that is superior to others. When selecting a screening tool, one should consider the
setting in which screening will take place, in terms of purposes and applications.
Governments and healthcare providers should promote implementation of nutri-
tional screening in pediatrics in all healthcare facilities.

Keywords
Malnutrition • Malnutrition risk • Nutritional status • Nutritional assessment •
Nutritional screening • Screening tools • Pediatrics • Children • Hospitalization •
Undernutrition

List of Abbreviations
ASPEN American Society of Parenteral and Enteral Nutrition
BIA PhA Bioelectrical Impedance Phase Angel
BMI Body Mass Index
CDC Center for Disease Control
ESPEN European Society of Enteral Nutrition
ESPGHAN European Society of Pediatric Gastroenterology Hepatology
and Nutrition
GI Gastro Intestinal
ICD-10 International Classification of Diseases
LOS Length of (hospital) Stay
MUAMC Mid Upper Arm Muscle Circumference
NRI Nutritional Risk Index
NRS Nutrition Risk Score
nutriSTEP Nutritional Screening Tool for Every Preschooler
PeDiSMART The Pediatric Digital Scaled Malnutrition Risk Screening Tool
PICU Pediatric Intensive Care Unit
PNRS Pediatric Nutrition Risk Score
PYMS Paediatric Yorkhill Malnutrition Score
SCAN Screening tool for childhood CANcer
SGNA Subjective Global Nutritional Assessment
STAMP Screening Tool for the Assessment of Malnutrition in Pediatrics
STRONGkids Screening Tool for Risk on Nutritional status and Growth
UK90 United Kingdom (growth charts)
WHO World Health Organization

Contents
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Nutritional Screening Tools Currently Available for Use in Pediatrics upon Hospital Admission 3
Comparison Between Different Screening Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Nutritional Screening Tools for Specific Medical Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Nutritional Screening Tools for Malnutrition in Pediatrics 3

Policies and Protocals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

Implementation of Nutritional Screening in Pediatrics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Dictionary of Terms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Summary Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

Introduction

Studies published over the past decade show that malnutrition prevalence upon
admission to pediatric hospitals remains considerably high, ranging from 6% to
14% in developed countries (Joosten and Hulst 2011) with higher prevalence rates in
infants and toddlers (Hecht et al. 2015). In specific medical conditions the described
prevalence is much higher (Joosten and Hulst 2008). Hospital stay has deleterious
impact on the nutritional status leading to weight loss during hospitalization even in
children affected merely by mild clinical conditions (Campanozzi et al. 2009;
Pacheco-Acosta et al. 2014). Early detection of children at malnutrition risk has
been discussed by international organizations such as ASPEN (Corkins et al. 2013),
ESPGHAN (Agostoni et al. 2005), and ESPEN (Kondrup et al. 2003).
Nutritional screening is a process aimed to identify an individual who is mal-
nourished or who is at risk for malnutrition, to determine if a detailed nutritional
assessment is indicated (Teitelbaum et al. 2005). An ideal screening tool should
demonstrate good ranking at concurrent validity (the extent to which screening
tools agree with each other), predicative validity (the extent to which screening
tools predict certain outcomes), reproducibility (agreement between users of a given
tool), and applicability in terms of ease and speed of administration (Elia and
Stratton 2011).
However widely used in adults, nutritional screening is not routinely conducted in
children because of the lack of a simple, properly validated screening tool (Hartman
et al. 2012). Also, since there is no one universally accepted definition of malnutri-
tion, there is also lack of consensus on a single definition for nutritional risk
(Huysentruyt et al. 2015). As a result, different screening tools were developed for
different purposes using different methodological methods (Elia and Stratton 2011),
making it nearly impossible to compare between tools in order to favor one tool over
the others. Thus, adding to the difficulties to choose an appropriate screening tool
there is lack of agreement between tools (Chourdakis et al. 2016) and evidence for an
impact of the screening on long-term outcome is lacking. In this chapter we will
review the literature on pediatric screening tools and discuss the complexity of
choosing a single tool. We will also discuss the policy needed to ensure implemen-
tation of nutritional screening in pediatrics.
4 G. Rub et al.

Table 1 Main components of screening tools. Different screening tools take into consideration
different nutritional related data when assessing nutritional risk. The table describes the different
screening tools with regards to their components
Current
Effect of nutritional Anthropometry Weight Other
Tools disease intake measurements loss components
NRS ٧ ٧ ٧ ٧ GI symptoms,
ability to eat
PNRS ٧ ٧ Pain
STRONGkids ٧ ٧ ٧
PYMS ٧ ٧ ٧ ٧
STAMP ٧ ٧ ٧
PNST ٧ ٧
PeDiSMART ٧ ٧ ٧ GI symptoms
Key: NRS nutrition risk score, PNRS pediatric nutrition risk score, STRONGkids screening tool for
risk on nutritional status and growth, PYMS paediatric yorkhill malnutrition score, STAMP screen-
ing tool for the assessment of malnutrition in pediatrics, SGNA subjective global nutritional
assessment, PeDiSMART the pediatric digital scaled malnutrition risk screening tool, GI gastro
intestinal

Nutritional Screening Tools Currently Available for Use in

Pediatrics upon Hospital Admission

The main components of the following screening tools are described in Table 1.

1. Nutrition Risk Score (NRS) – Developed by Reilly et al. (1995), the NRS is a
screening tool validated for use by the nursing staff to assess the risk of nutritional
depletion in hospitalized patients, both adults and pediatrics. The NRS collects
data on weight loss (amount and duration), BMI for adults or percentile charts for
children, food intake (appetite and the ability to eat and retain food), and stress
factors (effect of medical condition on nutritional requirements). Each parameter
is given a score affected by the severity of the condition described. The scores are
summed and patients are allocated into nutritional risk groups. A course of action
is then advised accordingly. The tool was validated on a sample of 20 patients
ranging from 6 weeks of age to 79 years. The NRS scores were assessed in
comparison to Nutritional Risk Index (NRI) (r = 0.68, p < 0.001) and a dietitian’s
clinical impression of the patient (r = 0.83, p < 0.001). Inter-rater reliability was
also assessed by comparing NRS scores conducted by two dietitians (r = 0.91,
p < 0.001). It should be noted that this tool was validated by comparison to the
NRI (Wolinsky 1990) which was originally designed for use in geriatrics. More-
over, this tool is not specifically intended for use in pediatrics and the study’s
sample size was exceptionally small with a large age range. No data on sample
size calculations was mentioned in the text.
2. Pediatric Nutritional Risk Score (PNRS) – Originally developed by Sermet-
Gaudelus et al. (2000), this tool is designed to identify children at risk of losing
Nutritional Screening Tools for Malnutrition in Pediatrics 5

2% or more of their admission weight, during hospitalization. Nutritional risk was

assessed prospectively in 296 children within the first 48 h of admission. Various
nutritional-related factors were assessed regarding their ability to predict weight
loss during hospitalization. Multivariate analysis indicated that food intake below
50% of dietary allowance, pain, and the severity of the disease were found to be
the most significant predictors of weight loss greater than 2% during hospital stay
( p = 0.0001 for each). It should be noted that although authors described this tool
as simple and easy to use, it has some downfalls. First, the tool is not applicable
upon admission since it requires monitoring nutritional intake over the first 48 h
of hospitalization. Thus, by using this tool, children who could benefit from early
intervention may be overlooked. Second, authors does not specify the resources
needed to implement the screening procedure; however, it appears trained per-
sonnel is required to implement the tool, if not only to assess dietary intake
compared with recommended allowance (Hartman et al. 2012). Third, there is no
data on reproducibility, inter-rater reliability, and the sensitivity of the tool
(Hartman et al. 2012; Joosten and Hulst 2014).
3. Screening Tool for Risk on Nutritional status and Growth (STRONGkids) –
Developed and tested by Hulst et al. (2010) in a prospective observational
multicenter study including 424 children admitted to 7 academic and 37 general
hospitals across the Netherlands. The STRONGkids consists of four items each
given a numeric score: subjective clinical assessment, high-risk disease, nutri-
tional intake, and weight loss. Weight and height were also measured and z-scores
were calculated. Risk scores were then compared with z-scores. Hulst et al.
showed that z-scores decreased with the rise in risk score (rs = 0.25,
p < 0.001), suggesting the higher the score according to STRONGkids, the
greater the risk according to anthropometric measurements. In addition, length
of hospital stay (LOS) was significantly shorter in children in the low-risk score
group compared to children with a moderate or high risk score ( p < 0.001). This
was also demonstrated by Cao et al. (2014) who used STRONGkids to analyze
nutritional risk in hospitalized children and its relationship with clinical outcome.
Cao et al. showed that children in the high-risk group demonstrated higher
complication rate, longer hospital stays, greater weight loss, and greater hospital
expanses, compared with children with moderate or low risk ( p < 0.001).
Hulst et al. (2010) described this tool as practical and easy to use; however, it
should be noted that both subjective clinical assessment and determining the
severity of the disease rely on expert knowledge and experience and thus might
not be suitable for use by all healthcare workers in different settings.
4. Paediatric Yorkhill Malnutrition Score (PYMS) – Developed by Germasimidis et
al. (Gerasimidis et al. 2010), the PYMS consists of four parameters: BMI, weight
loss history, changes in nutritional intake, and the severity of the underlying
disease and its impact on nutritional status. The tool was validated by comparing
its allocation to nutritional risk groups, with that determined by full nutritional
assessment. Inter-rater reliability was determined by comparing the PYMSs’
scores as rated by the nursing staff with those rated by dietitians. PYMSs’
allocation into nutritional risk groups was also compared with other commonly
6 G. Rub et al.

used tools (STAMP and SGNA). The study included 247 children, of whom
nurse-rated PYMS identified 59% of those rated at high risk by full nutritional
assessment. Of those rated at high risk by nurse-rated PYMS, 47% were con-
firmed at high risk by the full nutritional assessment. These results could be
interpreted that approximately 40% of children considered at high risk were not
identified by nurse-rated PYMS. Moreover, 53% of children were falsely identi-
fied at high risk by nurse-rated PYMS and inadequately referred to dietitians
(Hartman et al. 2012). PYMS demonstrated a moderate agreement with full
dietitians’ assessment (k = 0.46) and inter-rater reliability (k = 0.53) when
nurse-rated PYMS was compared with dietitian-rated PYMS.
5. Screening Tool for the Assessment of Malnutrition in Pediatrics (STAMP) –
Developed by McCarthy et al. (2012), STAMP consists of three steps and gathers
information aimed to detect low percentile weight for age, reported weight loss,
discrepancy between weight and height percentile, recently changed appetite, and
the expected nutrition risk of clinical diagnoses. All were identified as predictors
of nutrition risk, in the development phase of the tool. Nutritional risk is trans-
lated into the need for a referral to a full nutritional assessment. No outcomes
were evaluated in the validation study (Hartman et al. 2012). The tool was
validated in a cohort of 238 children upon admission to hospitalization. STAMPs’
allocation to nutritional risk groups was compared with a classification deter-
mined by a registered dietitian. The tool demonstrated fair to moderate reliability
(k = 0.54), and sensitivity and specificity were estimated at 70% and 91%,
respectively. It should be noted that this tool assesses growth parameters using
specific charts that were developed for this purpose (based on either UK90 or
CDC charts). Therefore one might claim implementation of such tool requires
resources, such as time and money, in order to train healthcare staff accordingly.
However, McCarthy et al. (2012) claim STAMP utilizes information that should
be routinely collected by nursing staff upon admission, thus introduction of the
tool required minimal training. Moreover, in the validation study, STAMP was
completed by different members of the nursing staff, including student nurses and
nursing supporting staff, suggesting implementation of the tool requires no
specific training.
6. Pediatric Nutrition Screening Tool (PNST) – Developed by White et al. (2016),
PNST consists of four dichotomous (yes/no) nutritional screening questions and
requires no data collection on medical condition or anthropometric measure-
ments. The PNST was validated on 295 hospitalized children from birth to
16 years in Australia. The pediatric SGNA and anthropometry were chosen as
the gold standards in defining nutritional risk. The sensitivity and specificity for
the PNST compared with the pediatric SGNA were moderate and high, scoring
77.8% and 82.1%, respectively. However, it should be noted that screening
procedure was performed by the same investigator using both PNST and
SGNA. No clinical outcome was investigated.
7. The Pediatric Digital Scaled Malnutrition Risk Screening Tool (PeDiSMART) -
Developed and validated by Karagiozoglou-Lampoudi et al. (2015), the
PeDiSMART is a software that consists of four elements: weight for age z-scores,
Nutritional Screening Tools for Malnutrition in Pediatrics 7

nutrition level intake, overall disease impact, and symptoms affecting intake.
Other than anthropometric data, all other parameters are described as categorical
variables rated 0–4. In order to validate the tool, PeDiSMART was compared to
bioelectrical impedance phase angle (BIA PhA) on 161 hospitalized children
aged 1 month to 17 years in Greece. Data showed inverse correlation between the
tool and PhA values (R = 0.582, p < 0.001). PeDiSMART was then compared
to STAMP, PYMS, and STRONGkids screening tools in 500 hospitalized chil-
dren. Patient allocation to nutritional risk groups on admission was associated
with clinical outcomes such as weight loss or nutritional support and LOS. ROC
curves showed sensitivity of 87% and specificity of 75%, ranking better than
STRONGkids and STAMP. In regards to outcome measurements, PeDiSMART
accuracy in predicting weight loss/nutritional intervention was higher than PYMS
and comparable to STAMP and STRONGkids. PeDiSMART accuracy in pre-
dicting LOS > 7 days was higher than STAMP and STRONGkids and compa-
rable to PYMS. Inter-rater reliability was evaluated in 57 patients by two
dietitians and showed moderate agreement of k = 0.474. Authors claim
PeDiSMART is rapid, easy to use, and suitable for use by all clinical staff
members. Since the software incorporates information documented on electronic
medical files automatically, the authors suggest the use of this technology may
facilitate and possibly improve the prediction of nutritional risk.

It should be noted that although some may refer to the Subjective Global
Nutritional Assessment (SGNA) tool published by Secker and Jeejeebhoy (2007)
as a nutritional screening tool, we chose not to describe this tool as it resembles a
nutritional assessment tool rather than a nutritional screening tool. In fact, it was
sometimes used as a gold standard for validation of a new tool, as was performed in
the validation of the PNST (White et al. 2016).
While a number of screening tools had emerged in the field of nutritional
screening upon hospitalization in the last decade, very few focused on nutritional
screening in community settings. Randall Simpson et al. (2008) developed the
nutriSTEP, a parent-administered questionnaire intended for nutritional risk screen-
ing in preschoolers. However, we choose not to focus on this tool since it is intended
to be solely used in community settings and was validated for that purpose only.
Despite Rub et al. (2016) validated STAMP for ambulatory use in pediatrics, further
research in the field is necessary before a definitive recommendation can be made.

Comparison Between Different Screening Tools

Since nutritional screening is widely recommended by numerous organizations such

as ASPEN (Corkins et al. 2013), ESPGHAN (Agostoni et al. 2005), and ESPEN
(Kondrup et al. 2003) a number of papers were recently published comparing
different screening tools in the hopes of identifying one tool that can be considered
superior to others. Examining the volume of these papers fails to yield consistent
results. The comparison between the tools brings forward the complexity of
8 G. Rub et al.

Table 2 Aims of the nutritional screening tools. Different screening tools were developed for
different aims, thus providing different uses. Table 3 describes the different screening tools with
regards to their aims
Determine
nutritional Identify need for Predict clinical outcome without
Tools status nutritional intervention nutritional intervention
NRS ٧
PNRS ٧ ٧
STRONGkids ٧ ٧
PYMS ٧ ٧ ٧
STAMP ٧ ٧
PNST ٧ ٧
PeDiSMART ٧ ٧
Key: NRS nutrition risk score, PNRS pediatric nutrition risk score, STRONGkids screening tool for
risk on nutritional status and growth, PYMS paediatric yorkhill malnutrition score, STAMP screen-
ing tool for the assessment of malnutrition in pediatrics, SGNA subjective global nutritional
assessment, PeDiSMART the pediatric digital scaled malnutrition risk screening tool

prioritizing one tool over the other due to the different design and evaluation
methods used. As was previously proposed by (Elia and Stratton 2011), different
screening tools were designed for diverse purposes, for use by people with different
backgrounds and for application by one or more settings, one or more age groups,
and one or more disease groups.
With regards to the aims of the tools, while NRS, STAMP, PYMS, and
STRONGkids are designed to be completed upon admission, PNRS requires the
assessment of nutritional intake over the first 48 h of hospitalization. Ergo, using the
PNRS requires time and resources spent on nutritional intake documentation and
analysis, suggesting this tool has greater resemblance to a nutritional assessment tool
rather than a nutritional screening tool. The aims of the aforementioned screening
tools are displayed in Table 2.
Within the aforementioned tools STAMP, SGNA, and PYMS include anthropo-
metric measurements thus identifying nutritional status upon admission, while the
other tools merely provide the perceived risk of deterioration. PNRS, PYMS, and
STRONGkids were also designed to prognostically predict clinical outcomes (with-
out nutritional intervention) such as weight loss of >2% or LOS. However, it should
be noted that LOS may not function as a direct assessment of nutritional risk because
it may be influenced by many other factors and a causative relation has not been
shown (Huysentruyt et al. 2015).
Another complexity arising from comparison is the validation methods used for
each tool. There is currently no agreed upon “gold standard” for the assessment of
malnutrition and malnutrition risk (since there is no universally accepted definition
of malnutrition). In the absence of a nutrition screening tool that can act as a “gold”
standard, information on the agreement between tools (concurrent validity) is used,
especially when the comparison involves tools developed for the same purpose and
when no judgment is made about the superiority of one tool over another (Elia and
Table 3 Studies comparing different screening tools with regards to clinical outcomes. The table presents studies recently published comparing different
screening tools in regards to their ability to predict clinical outcome, in the hopes of identifying one nutritional screening tool superior to other. Different studies
compared different tools with regards to different outcomes
NRS PNRS STRONGkids PYMS STAMP PNST PeDiSMART Clinical outcome
Chourdakis et al. ٧ ٧ ٧ Anthropometry, LOS, infection rates
2016
Huysentruyt et al. ٧ ٧ ٧ ٧ >2% weight loss, referral to dietitian,
2015 dietitians’ assessment, nutritional intervention
Karagiozoglou- ٧ ٧ ٧ ٧ Weight loss, nutritional support, LOS
Lampoudi et al.
2015
Galera-Martinez et ٧ ٧ Anthropometry, LOS
Nutritional Screening Tools for Malnutrition in Pediatrics

al. 2017
Thomas et al. 2016 ٧ ٧ Anthropometry
Ling et al. 2011 ٧ ٧ Anthropometry, nutritional intervention
Key: NRS nutrition risk score, PNRS pediatric nutrition risk score, STRONGkids screening tool for risk on nutritional status and growth, PYMS paediatric
yorkhill malnutrition score, LOS length of (hospital) stay, STAMP screening tool for the assessment of malnutrition in pediatrics, SGNA subjective global
nutritional assessment, PeDiSMART the pediatric digital scaled malnutrition risk screening tool
9
10 G. Rub et al.

Stratton 2012). For example, the NRS was compared to NRI for validation, the
STRONGkids was compared to WHO cutoff reference for malnutrition, STAMP
was compared with a dietitian’s nutritional assessment, and PYMS was compared to
a dietitians’ assessment as well as to both STAMP and SGNA.
In the absence of a universally agreed upon reference, attempts have been made to
rank nutritional screening tools by their ability to predict outcome. Summary of the
studies comparing the different tools with regards to clinical outcome measured is
displayed in Table 3. For instance, the PNRS was designed to predict weight loss of
more than 2% during hospitalization. Also, as part of its validation study,
STRONGkids’ allocation into nutritional risk groups was compared to LOS showing
higher risk group had longer hospitalizations. Chourdakis et al. (2016) recently
evaluated PYMS, STAMP, and STRONGkids compared with and were related to
anthropometric measurements and clinical variables such as LOS and infection rates.
Children categorized in medium and high-risk groups according to all tools were
found to have significantly longer LOS compared with children in low-risk group.
However, authors stated that a considerable portion of children with subnormal
anthropometric measures were not identified by neither tool. Thus the use of these
tools is not recommended by the authors. Using LOS as a means to rank nutritional
screening tools raises some reservations. First, LOS as well as other clinical out-
comes is subjected to many confounders and can be influenced by the medical staffs’
policy or even by work load and availability of medical staff to discharge patients.
This was also supported by Huysentruyt et al. (2015) who chose not to consider LOS
as a direct assessment of nutritional risk, in their systematic review comparing
PNRS, STAMP, PYMS, and STRONGkids. Moreover, when LOS as an outcome
measure was controlled for confounders in the PeDiSMART validation study
(Karagiozoglou-Lampoudi et al. 2015), it was found not to be significantly associ-
ated with nutritional risk. Another reservation is that a tool that is good at predicting
outcomes in the absence of nutritional interventions is not necessarily good at
predicting outcomes induced by nutritional interventions (Elia and Stratton 2012).
Karagiozoglou-Lampoudi et al. (2015) used weight loss or nutrition support during
hospitalization as an outcome measure and showed it was independently associated
with the malnutrition risk groups’ allocation on admission, regardless of the tool
used for allocation.
In terms of practicality, a screening tool should be fast and easy to use and should
be suitable for use by untrained personnel. In terms of time needed for administra-
tion, original validation studies did not report speed of administration. Ling et al.
(2011) reported time of administration by two trained investigators, and found that
while STAMP took 10–15 min due to anthropometric measurements, STRONGkids
took merely 5 min. These findings were supported by Huysentruyt et al. (2013) that
reported median time of 3 min for administration of STRONGkids in a validation
study in Belgium. In terms of personnel needed to perform nutritional screening,
while STAMP and PYMS were developed for use by nurses, the PNRS requires
qualified personnel to assess nutritional intake, and STRONGkids was originally
developed to be completed by junior physicians or pediatricians. Nevertheless, in a
different validation study (Huysentruyt et al. 2013) STRONGkids was administered
Nutritional Screening Tools for Malnutrition in Pediatrics 11

Table 4 The care plans advised by the different nutritional screening tools according to screening
result. Each screening tool offers a specific care plan according to the allocated risk group. The table
presents the course of action advised by each tool according to the different risk groups allocated
Tools Low risk Moderate risk High risk
NRS N/A N/A N/A
PNRS None Weight surveillance, report Nutritional assessment, monitor
intake, consider dietetic intake, consider nutritional
consult intervention
STRONGkids Repeat Check weight twice a week, Refer the child to a dietitian
screening consider referring the child
weekly to a dietitian
PYMS Repeat Repeat screening after Refer the child to a dietitian
screening 3 days
weekly
STAMP Repeat Monitor the child’s Refer the child to a dietitian/
screening nutritional intake for nutritional support team/
weekly 3 days, repeat screening consultant
after 3 days
PNST None Not relevant (moderate risk Refer the child for further
category does not appear in nutritional assessment, check if
the PNST) child was previously cared for by
a dietitian, measure weight and
height, commence food and fluid
intake record
PeDiSMART Check Check weight status twice a Refer to a dietitian
weight week, repeat the screening
status after a week
weekly
Key: N/A no care plan available, NRS nutrition risk score, PNRS pediatric nutrition risk score,
STRONGkids screening tool for risk on nutritional status and growth, PYMS paediatric yorkhill
malnutrition score, LOS length of (hospital) stay, STAMP screening tool for the assessment of
malnutrition in pediatrics, SGNA subjective global nutritional assessment, PeDiSMART the pediat-
ric digital scaled malnutrition risk screening tool

by nurses and was found to be easy to use with substantial intra- and inter-rater
reliability rates. Furthermore, in another very recently published study by Galera-
Martinez et al. (2017), STAMP and STRONGkids were assessed for reproducibility
and inter-rater reliability between expert staff specialized in pediatric nutrition
(physicians and dietitians) and clinical staff nonexpert in nutrition. Agreement
between expert and nonexpert staff was good: 94.78% for STRONGkids
(k = 0.72 [p < 0.001]) and 92.55% for STAMP (k = 0.74 [p < 0.001]). These
findings suggest whether STRONGkids was originally developed for administration
by qualified personnel, it can be used for practice by clinical staff, as it is already
widely used in current clinical practice by nurses (Joosten and Hulst 2014). The
PeDiSMART was claimed to be appropriate for use with all clinical staff members.
Nonetheless, it was validated for use by certified dietitians (Karagiozoglou-
Lampoudi et al. 2015). Further research is needed to assess PeDiSMARTs’ inter-
rater reliability between a dietitian’s assessment and other members of the clinical
staff.
12 G. Rub et al.

Not all nutritional screening tools describe the course of action that is advised
according to screening results, and the ones who do, describe different follow-up
care plans. The care plans advised by the different nutritional screening tools are
displayed in Table 4.

Nutritional Screening Tools for Specific Medical Conditions

Different screening tools are designed for different purposes, to be used on one or
more underlying disease (Elia and Stratton 2011). Some tools were originally
designed to be used upon specific conditions such as SCAN, nutritional screening
tool for childhood cancer (Murphy et al. 2016), or the nutrition risk screening tool in
cystic fibrosis (McDonald 2008), while other tools were later on validated for
specific purposes. For example, PeDiSMART was validated in 30 children with
chronic kidney disease (Apostolou et al. 2014). Moderate inverse correlation was
found between PeDiSMART score and PhA ( p = 0.001), MUAMC ( p = 0.008) as
well as protein intake ( p = 0.016). STAMP was also validated in 51 pediatric spinal
cord injury (SCI) patients admitted to a tertiary SCI center (Wong et al. 2013).
STAMP had moderate agreement with dietitians’ assessment (k = 0.507). STAMP,
PYMS, and PMST (modified STAMP) were tested in acute pediatric setting in 300
children (Thomas et al. 2016).
The tools were compared to WHO growth reference cutoffs for malnutrition.
Those who scored medium or high risk by the tools were compared with those who
could be considered malnourished or at risk of malnutrition using the WHO’s
definitions. The results showed poor sensitivity and specificity rates; however, it
should be noted that WHOs’ definitions for malnutrition assess current state rather
than the risk for malnutrition. Thus it may not be the most suitable reference to test
validity. Moreover, the majority of children at PICU are at nutritional risk of some
degree, thus it may be more effective to directly perform nutritional assessment in
the form of growth and intake assessment rather than nutritional screening.
PNRS, PYMS, STAMP, and STRONGkids were also tested in children with IBD
(Wiskin et al. 2012). The tools were tested on 46 children and risk score was
compared to the degree of malnutrition according to WHO’s definition (as expressed
in ICD-10). The tools showed good agreement with one another (k = 0.6); however,
no agreement was found between each tool and anthropometric measures (k < 0.1).
Nevertheless, it should be noted that authors compared nutritional risk as assessed by
the aforementioned tools, with criteria of nutritional assessment such as set by the
WHO’s definition for malnutrition.

Policies and Protocals

Nutritional Screening Tools for Malnutrition in Pediatrics 13

Implementation of Nutritional Screening in Pediatrics

In this chapter we have described recently published research in the field of nutri-
tional screening in pediatrics, and the complexity of comparing the different tools in
the hopes of identifying one screening tool that can be considered superior to others.
Below we describe the detailed policies that should be adopted in order to make
implementation of nutritional screening in pediatrics feasible in various settings.
Policy should acknowledge a number of main areas:

• The government’s role in prioritizing nutritional screening and allocating

resources for that matter
• The standardization of nutritional screening across different healthcare facilities
• The role of healthcare facilities in implementation of nutritional screening
• The need to establish follow-up protocol when advised and the means to do so
• The assessment of long-term nutritional outcomes

As nutritional screening in pediatrics is recommended by international organiza-

tions (Corkins et al. 2013; Agostoni et al. 2005; Kondrup et al. 2003), it should be
made into a governmental policy in order to reduce disease complications (Sermet-
Gaudelus et al. 2000; Cao et al. 2014), LOS (Hulst et al. 2010; Cao et al. 2014;
Karagiozoglou-Lampoudi et al. 2015; Chourdakis et al. 2016), and overall economic
burden on the healthcare system (Ahmed et al. 2012). To this date the majority of
nutritional screening tools in pediatrics are suitable for use in specific healthcare
facilities. Ergo, some tools are suitable for use in primary healthcare centers while
other tools are suitable for ambulatory setting. Government should allocate resources
to validate one nutritional screening tool for pediatrics that can be feasible for use in
different healthcare settings. The use of a single screening tool in different healthcare
settings will allow follow up and monitoring of nutritional status when child is being
transferred from one healthcare facility to another. Government should also instruct
that nutritional screening become mandatory upon arrival or admission to healthcare
facilities, and should enforce implementation upon periodic inspection. Nutritional
screening should be named as a quality measure at the national state, to ensure
quality of care.
Different healthcare facilities should incorporate nutritional screening as part of
their routine patient care. In order to do so, healthcare facilities should introduce
nutritional screening tools and initiate training sessions on performance of such tools
to all healthcare staff. All healthcare staff should be educated on the importance of
nutritional screening in order to promote adherence. Healthcare facilities should
develop a follow-up protocol which will define tendency required and the actions
taken to ensure follow-up takes place. It is advised that each child will be assigned a
case manager (such as the primary physician), who will supervise the process.
Healthcare facilities should create a system that will allow physician minimal time
consuming, easy and effective means to supervise nutritional status of children under
their care, such as a computerized alert system for instance.
14 G. Rub et al.

Only after nutritional screening is widely implemented, research can focus on

long-term effects and assess whether early recognition of malnutrition risk produced
appropriate nutritional intervention. Resources should be allocated, in the national
and institutional level, to the study of nutritional risk screening effect on clinical
outcomes such as rate of childhood infections, readmissions to hospitals, LOS as
well as long-term outcomes such as long-term health parameters, social and aca-
demic achievements, and overall economic burden.

Dictionary of Terms

• Malnutrition – A state of over- or undernutrition that is accompanied by micro-

and or macronutrient deficiencies, and causes malfunction at the level of the cell
and/or the organ and/or the body.
• Malnutrition risk – The risk of deterioration into a state of over- or
undernutrition.
• Nutritional assessment – The assessment of current nutritional status of a
subject. Full nutritional assessment is usually a subjective process performed by
a clinical dietitian and includes nutritional intake, review of nutrition-related
blood tests, and anthropometric measurements.
• Screening tool – A questionnaire, form, or method that enables early identifica-
tion of subject with specific conditions within a large group.
• Length of hospital stay (LOS) – Duration of hospitalization to a healthcare
institute. This term is commonly used to represent clinical outcome of assigned
intervention.
• Anthropometry – A term that describes measurements of the human body such
as weight, height, mid-upper arm, or waist circumferences. Anthropometric
indices are usually used for the purpose of nutritional assessment. In children
anthropometric measurements are vital to assess growth and development.
• Inter-rater reliability – The degree of agreement between different raters when
performing the same procedure.
• Intra-rater reliability – The degree of agreement between repeated tests
performed by the same rater.
• Concurrent validity – The degree of agreement between a method and previ-
ously described methods aimed at testing the same thing.
• Predictive validity – The degree in which a certain test corresponds with
measurable clinical outcome.
• Reproducibility – The degree in which a certain test can be reproduced to yield
the same results either by the same rater or by different raters.
• Sensitivity – The proportion of positive observations recognized by the screening
process among all the true positives in the population. For instance, the proportion
of children with high risk for malnutrition as recognized by a certain screening
tool applied, among all the children with high risk for malnutrition (combination
of the children who were recognized by screening and those who were not).
Nutritional Screening Tools for Malnutrition in Pediatrics 15

• Specificity – The proportion of negative observations recognized by the screen-

ing process among all the true negatives in the population. For instance, the
proportion of children with low risk for malnutrition as recognized by a certain
screening tool applied, among all the children with low risk for malnutrition
(combination of the children who were recognized by screening and those who
were not).
• Z-scores – A term that describes distance from the mean as expressed by standard
deviations.
• Clinical outcome – The effect of a certain intervention on the clinical state of a
patient, as expressed by infection rate, comorbidity, duration of hospital stay, and
so forth. The effect can be either short or long term.
• Subjective clinical assessment – An assessment of the clinical state of a subject
which is determined by subjective impression of physical examination. In nutri-
tional assessment this can include signs of physical wasting such as muscle or fat
atrophy.
• Gold standard – The best available test/method/practice in a specific field.
• Bioelectrical impedance phase angle (BIA PhA) – A measurement used to
assess muscle mass as part of the nutritional assessment procedure. The device
applies a low-frequency current to the body and assesses the proportion of
reactance and resistance. The result reflects the ratio of the body’s cell mass to
fat-free mass.

Summary Points

• Nutritional screening is recommended by multiple international organizations.

• While widely used in adults, nutritional screening is not routinely practiced due to
the lack of a universally accepted, validated, easy-to-use screening tool.
• Over the past two decades many different screening tools emerged. Each tool
designed for different purposes, to be used by different members of the clinical
staff, on different healthcare settings.
• Comparison between the tools is complex due to the different methodology
methods used, thus recent research was unable to properly compare between
tools or demonstrate the superiority of one tool over the other. In choosing a
screening tool one should take into account predictive and concurrent validity,
reproducibility, and applicability in terms of ease and speed of application. Tools
should also specify advised course of action.
• Further research should validate the use of a single screening tool in different
healthcare facilities in order to facilitate continuous patient care between different
settings.
• Nutritional risk screening should be administered on admission to healthcare
facilities.
• Outcome measures should integrate the screening process and the nutritional
interventions that follow the screening process.
16 G. Rub et al.

• Follow-up of nutritional risk screening should be performed at the frequency

suitable for the healthcare facility (frequently performed during hospitalization
and less frequent on community settings).
• The implementation of nutritional risk screening in pediatrics should be made into
governmental policy as a primary prevention strategy.
• Following implementation, research should focus on nutrition-related outcomes
to investigate cost-effectiveness.

References
Agostoni C, Axelson I, Colomb V et al (2005) The need for nutrition support teams in pediatric
units; a commentary by the ESPGHAN committee on nutrition. J Pediatr Gastroenterol Nutr
41(1):8–11
Ahmed T, Hossain M, Sanin KI (2012) Global burden of maternal and child undernutrition and
micronutrient deficiencies. Ann Nutr Metab 61(suppl 1):8–17
Apostolou A, Printza N, Karagiozoglou-Lampoudi T, Dotis J, Papachristou F (2014) Nutrition
assessment of children with advanced stages of chronic kidney disease – a single center study.
Hippokratia 18(3):212–216
Campanozzi A, Russo M, Catucci A et al (2009) Hospital-acquired malnutrition in children with
mild clinical conditions. Nutrition 25(5):540–547
Cao J, Peng L, Li R et al (2014) Nutritional risk screening and its clinical significance in
hospitalized children. Clin Nutr 33:432–436
Chourdakis M, Hecht C, Gerasimidis K et al (2016) Malnutrition risk in hospitalized children: use
of 3 screening tools in a large European population. Am J Clin Nutr 103:1301–1310
Corkins MR, Griggs KC, Groh-Wargo S, Han-Markey TL, Helms RA, Muir LV, Szeszycki EE,
Task Force on Standarts for Nutrition Support: Pediatric Hospitalized Patients, The American
Society for Parenteral and Enteral Nutrition Board of Directors (2013) Standarts for nutrition
support: pediatric hospitalized patients. Nutr Clin Pract 28(2):263–276
Elia M, Stratton RS (2011) Considerations for screening tool selection and role of predictive and
concurrent validity. Curr Opin Clin Nutr Metab Care 14:425–433
Elia M, Stratton RJ (2012) An analytic appraisal of nutrition screening tools supported by original
data with particular reference to age. Nutrition 28:477–494
Galera-Martinez R, Morais-Lopez A, Rivero de la Rosa MC et al (2017) Reproducibility and inter-
rater reliability of 2 paediatric nutritional screening tools. J Pediatr Gastroenterol Nutr
64:e65–e70
Gerasimidis K, Keane O, Macleod I, Flynn DM, Wright CM (2010) A four-stage evaluation of the
Paediatric Yorkhill malnutrition score in a tertiary paediatric hospital and a district general
hospital. Br J Nutr 104:751–756
Hartman C, Shamir R, Hecht C, Koletzko B (2012) Malnutrition screening tools for hospitalized
children. Curr Opin Clin Nutr Metab Care 15:303–309
Hecht C, Weber M, Grote V et al (2015) Disease associsted malnutrition correlates with length of
hospital stay in children. Clin Nutr 34:53–59
Hulst JM, Zwart H, Hop WC, Joosten KF (2010) Dutch national survey to test the STRONG(kids)
nutritional risk screening tool in hospitalized children. Clin Nutr 29:106–111
Huysentruyt K, Alliet P, Muyshont L et al (2013) The STRONGkids nutritional screening tool in
hospitalized children: a validation study. Nutrition 29:1356–1361
Huysentruyt K, Devreker T, Dejockheere J et al (2015) Accuracy of nutritional screening tools in
assessing the risk of undernutrition in hospitalized children. J Pediatr Gastroenterol Nutr
61:159–166
Nutritional Screening Tools for Malnutrition in Pediatrics 17

Joosten KF, Hulst JM (2008) Prevalence of malnutrition in pediatric hospital patients. Curr Opin
Pediatr 20(5):590–596
Joosten KFM, Hulst JM (2011) Malnutrition in pediatric hospital patients: current issues. Nutrition
27:133–137
Joosten KFM, Hulst JM (2014) Nutritional screening tools for hospitalized children: methodolog-
ical considerations. Clin Nutr 33:1–5
Karagiozoglou-Lampoudi T, Daskalou E, Lampoudis D, Apostolou A, Agakidis C (2015) Com-
puter-based malnutrition risk calculation may enhance the ability to identify pediatric patients at
malnutrition-related risk for unfavorable outcome. J Pediatr Gastroenterol Nutr 39:418–425
Kondrup J, Allison SP, Elia M, Vellas B, Plauth M (2003) ESPEN guidelines for nutrition screening
2002. Clin Nutr 22(4):415–421
Ling RE, Hedges V, Sullivan PB (2011) Nutritional risk in hospitalized children: an assessment of
two instruments. e-SPEN, Eur e-J Clin Nutr Metab 6(3):e153–e157
McCarthy H, Dixon M, Crabtree I, Eaton-Evans MJ, McNulty H (2012) The development and
evaluation of the screening tool for the assessment of malnutrition in Paediatrics (STAMP) for
use by healthcare staff. J Hum Nutr Diet 25(4):311–318
McDonald CM (2008) Validation of a nutrition risk screening tool for children and adolescents with
cystic fibrosis ages 2–20 years. J Pediatr Gastroenterol Nutr 46:438–446
Murphy AJ, White M, Viani K, Mosby TT (2016) Evaluation of the nutrition screening tool for
childhood cancer (SCAN). Clin Nutr 35:219–224
Pacheco-Acosta JC, Gomez-Correa AC, Florez ID et al (2014) Incidence of nutrition deterioration
in nonseriously ill hospitalized children younger than 5 years. Nutr Clin Pract 29(5):692–697
Randall Simpson JA, Keller HH, Rysdale LA, Beyers JE (2008) Nutrition screening tool for every
preschooler (NutriSTEPtm): validation and test-retest reliability of a parent-administered ques-
tionnaire assessing nutrition risk in preschoolers. Eur J Clin Nutr 62:770–780
Reilly HM, Martineau JK, Moran A, Kennedy H (1995) Nutritional screening – evaluation and
implementation of a simple nutrition risk score. Clin Nutr 14:269–273
Rub G, Marderfeld M, Poraz I et al (2016) Validation of a nutritional screening tool for ambulatory
use in pediatrics. J Pediatr Gastroenterol Nutr 62(5):771–775
Secker DJ, Jeejeebhoy KN (2007) Subjective global nutritional assessment for children. Am J Clin
Nutr 85:1083–1089
Sermet-Gaudelus I, Poisson-Salomon AS, Colomb V, Vrusset MC, Mosser F, Berrier F, Ricour C
(2000) Simple pediatric nutritional risk score to identify children at risk of malnutrition. Am J
Clin Nutr 72:64–70
Teitelbaum D, Guenter P, Howell WH et al (2005) Definitions of terms, style, and conventions used
in a.S.P.E.N. Guidelines and standards. Nutr Clin Pract 20:281–285
Thomas PC, Marino LV, Williams SA, Beattie RM (2016) Outcome of nutritional screening in the
acute paediatric setting. Arch Dis Child 101:1119–1124
White M, Lawson K, Ramsey R et al (2016) Simple nutrition screening tool for pediatric inpatients.
J Pediatr Gastroenterol Nutr 40:392–398
Wiskin AE, Owens DR, Cornelius VR, Wootton SA, Beattie RM (2012) Paediatric nutrition risk
scores in clinical practice: children with inflammatory bowel disease. J Hum Nutr Diet
25:319–322
Wolinsky FD, Coe RM, McIntosh WA et al (1990) Progress in the development of a nutritional risk
index. J Nutr 120(suppl11):1549–1553
Wong S, Graham A, Hirani SP, Grimble G, Forbes A (2013) Validation of the screening tool for the
assessment of malnutrition in Paediatrics (STAMP) in patients with spinal cord injuries (SCIs).
Spinal Cord 51:424–429