International Journal of Nursing Studies 47 (2010) 67–77

Contents lists available at ScienceDirect

International Journal of Nursing Studies

journal homepage: www.elsevier.com/ijns

The Hyperemesis Impact of Symptoms Questionnaire: Development and

validation of a clinical tool
Zoe Power a, Malcolm Campbell a, Pamela Kilcoyne b, Henry Kitchener a,b,
Heather Waterman a,*
a
University of Manchester, Oxford Road, Manchester M13 9PL, UK
b
St. Mary’s Hospital, Central Manchester University Hospitals NHS Foundation Trust, Whitworth Park, Manchester, UK

A R T I C L E I N F O A B S T R A C T

Article history: Background: The Hyperemesis Impact of Symptoms Questionnaire is a clinical tool designed
Received 18 January 2009
to assess holistically the impact of the physical and psychosocial symptoms of hyperemesis
Received in revised form 24 June 2009
gravidarum (HG) on individuals. Its purpose is to aid planning and implementation of
Accepted 26 June 2009
tailored care for women with HG. To our knowledge no similar tool exists.
Objective: To assess the validity and reliability of the HIS questionnaire.
Keywords:
Hyperemesis gravidarum Design: As no similar tool exists, we compared the HIS with three tools that reflect its key
Patient care planning areas: physical impact (Pregnancy Unique Quantification of Emesis – PUQE score and
Validation studies markers of severity of HG), psychological impact (Hospital Anxiety and Depression Score –
HADS) and social impact (SF12 quality of life score).
Setting: A large regional referral, women and children’s hospital in the North West of
England.
Participants: The HIS was evaluated on 50 women admitted to hospital with HG and 50
women recruited from ante-natal clinic without severe nausea and vomiting of pregnancy
and with an uncomplicated pregnancy.
Results: Good criterion validity was demonstrated by strong significant correlations with
all three scores (PUQE, r = 0.75, p < 0.001, HADS, depression r = 0.76, p < 0.001, and SF12,
mental component r = 0.65, p < 0.001). The HIS showed good internal consistency,
Cronbach alpha 0.87, split half 0.80.
Conclusions: There is evidence for the validity and reliability of the HIS to assess the impact
of the physical and psychosocial symptoms of HG. Further research is currently underway
to establish the clinical utility of the HIS questionnaire in the care of women hospitalised
with HG.
ß 2009 Elsevier Ltd. All rights reserved.

What is already known about the topic?  The impact of HG affects women psychologically and
socially as well as physically.
 Hyperemesis gravidarum is a self-limiting but poten-  Current standard care relies largely on rehydration and
tially debilitating condition, often requiring repeated anti-emetic therapy, but is ineffective for many women. A
admissions to hospital. more individualised model, addressing social and psycho-
logical, as well as physical needs, may be more effective.

What this paper adds

* Corresponding author at: Room 6.314a, University Place, School of
Nursing, Midwifery and Social Work, University of Manchester, Man-
 This study shows that the HIS is a valid and reliable
chester M13 9PL, UK. Tel.: +44 0161 306 7864; fax: +44 0161 306 7894. clinical tool with which to assess the impact of the
E-mail address: heather.waterman@manchester.ac.uk (H. Waterman). physical and psychosocial symptoms of HG.

0020-7489/$ – see front matter ß 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijnurstu.2009.06.012
68 Z. Power et al. / International Journal of Nursing Studies 47 (2010) 67–77

 The HIS appears to offer a holistic assessment tool that is HG are impacting most on a woman’s life. Treatment,
acceptable to women with HG. It provides a basis for advice, reassurance and referral can then be tailored to the
management that tailors the care of women with HG to individual in the most effective way. Focusing on the
their individual needs. impact of symptoms also allows health care professionals
to provide a plan of care that does not end with discharge
1. Introduction from hospital and the removal of intravenous therapy.
Tailored care planning using the HIS can assist women
Hyperemesis gravidarum (HG) is a state of excessive while supported in hospital to develop strategies to
nausea and vomiting in early pregnancy which usually manage their symptoms when at home. The closest
resolves spontaneously by 16–20 weeks gestation. Symp- available questionnaire to the HIS currently available is
toms of HG can be severe and can affect all aspects of a the Health Related Quality of Life for Nausea and Vomiting
woman’s life, impacting as it does not only on physical and during Pregnancy (NVPQOL) (Lacasse and Berard, 2008;
psychological well-being but also on social functioning Magee et al., 2002). However the NVPQOL aims to measure
(Attard et al., 2002; Poursharif et al., 2008). It is estimated quality of life in nausea and vomiting of pregnancy and
that up to 2% of pregnant women will be admitted with HG is primarily designed for research purposes. The HIS is
(Bailit, 2005; Gazmararian et al., 2002). Prolonged HG with different as it offers a specific clinical function that is
poor weight gain during pregnancy is also reported to be assisting in the assessment of the impact of symptoms
associated with lower birth weight infants (Bailit, 2005; with the purpose of tailoring person specific care plans to
Dodds et al., 2006; Paauw et al., 2005). meet individual women’s needs. The HIS may however
The aetiology of HG is poorly understood. Standard prove to also be useful for research in the future.
treatment is aimed at rehydration and anti-emetic therapy Severe symptoms of nausea and vomiting of preg-
(Jewell and Young, 2006). Best practice guidelines have been nancy may be similar amongst many women. However,
published by several groups (ACOG, 2004; Arsenault et al., how each woman is affected by and able to manage their
2002; Koren et al., 2002) but to our knowledge there is no symptoms will differ depending on various factors such
clinical, validated assessment tool for HG that is specifically as social circumstances and physical and psychological
designed to assist in planning tailored care for women. As well-being. Women often describe HG in terms of a
women with HG often experience problems other than ‘‘vicious circle’’ or ‘‘downward spiral’’ requiring ‘‘con-
nausea and vomiting such as depression and severe impact trol’’ or a ‘‘break in the cycle’’(Power et al., 2007). Nausea
on social functioning (Attard et al., 2002; Gazmararian et al., and vomiting can cause dehydration, exhaustion and low
2002; Mazzotta et al., 2000; Munch and Schmitz, 2006; mood. In turn all of these effects can further exacerbate
O’Brien et al., 2002) our hypothesis is that a purely ‘medical original symptoms in a downward cyclical spiral until
model’ of treating HG which focuses on rehydration and anti- they eventually become unmanageable and a woman
emesis, will often be inadequate for the needs of affected seeks medical assistance to control them. There is
women. Because the mainstay of management-anti-emetic growing research evidence to support the view that
therapy, is frequently unsuccessful, many women require where symptoms are not unmanageable, given the
several readmissions, compounding social difficulties and appropriate explanation, information, reassurance and
feelings of depression (Gazmararian et al., 2002). A move to a support, women are able to rationalise their symptoms
more holistic model of care may address these problems and deal with them in a more positive way (Munch,
more successfully. Reports in the literature (Munch and 2000; Soltani and Taylor, 2003). The aim of the HIS is to
Schmitz, 2006; Soltani and Taylor, 2003) and our own early identify the areas of a woman’s life where symptoms of
research (Power et al., 2007) suggest that there is sometimes HG impact most, therefore allowing clinicians to direct
a feeling of therapeutic nihilism amongst doctors and nurses assistance where it is most required in a tailored and
caring for women with HG regarding treatment and a lack of individualised way. For example as well as questions on
attention to identifying the symptoms and concerns that ability to tolerate diet and fluids, the HIS questionnaire
women are experiencing. In addition to feeling that nothing asks how tired the woman feels, whether she is able to
can be done to relieve their symptoms, women in our care for herself, her home and her family, whether she
exploratory work also reported feeling disbelieved by health feels especially anxious about her unborn child, whether
care professionals as to the severity of their symptoms she feels defeated by her symptoms and whether she
(Power et al., 2007). This led some to feel angry, more feels people understand how ill she is feeling? Each
distressed or unimportant and as if they were wasting staff question can score from 0 to 3. A score of 2 or above
time. We believe that using a valid and reliable clinical should trigger action from the health care professional to
questionnaire based on the holistic assessment of women address the issue raised by the question. This may be in
with HG will lead to a better understanding by clinicians of the form of advice and support or by practical action. To
the impact of HG on individuals. Using the results of the maximise clinical utility the HIS can be used with the
questionnaire can then assist in better care planning that is associated ‘‘HIS handbook’’ developed alongside the
tailored to the woman’s individual needs (Appendix 1). HIS. (See http://personalpages.manchester.ac.uk/staff/
Heather.Waterman/HIShandbookfinessed100908.doc.)
2. The Hyperemesis Impact of Symptoms Questionnaire The ‘‘HIS handbook’’ is a clinical document that suggests
advice and support to offer in the event of a high scoring
The HIS is a ten point questionnaire designed to assist response to any question. The HIS handbook was not
health care professionals in assessing where symptoms of tested as part of the validation of the tool.
 Z. Power et al. / International Journal of Nursing Studies 47 (2010) 67–77 69

3. Methods tone of the interview remains conversational and informal.

When necessary the interviewer probes for a more in-
The study was divided into two phases. The research depth response or guides the conversation to make sure all
objectives of the first phase were to determine key issues topics are covered. In our interviews, if a participant went
for women with HG, to generate the HIS questionnaire, and off the topic guide, but into an area of conversation that
to carry out a pilot exercise. The objectives of phase two seemed relevant, the interviewer let the conversation flow.
were to investigate its validity and reliability. If the The topic guide would then be returned to later in the
questionnaire was successfully validated, further research interview. It was hoped that this would minimise the risk
into its application as a clinical tool was planned. of missing topics important to the participant but not
considered in the topic guide. Later topic guides were
3.1. Phase one: development of the HIS questionnaire influenced by earlier data collection.
The interview schedule covered topic areas such as
The first phase of the project was exploratory and ‘‘How would a typical day with HG go for you?’’, ‘‘Is there
involved collection of data to inform the development of anything that makes the nausea/vomiting worse or
the HIS questionnaire to improve the care of women better?’’, ‘‘What was it that made you decide to come into
hospitalised with HG. Hospitalised women were chosen hospital?’’ and ‘‘Do you feel better or worse in hospital?-
because the development of the HIS arose from a hospital What makes you feel better or worse in hospital?’’
based project instigated by clinicians investigating the All interviews were conducted by ZP. All interviews were
local issue ‘‘Why do women with HG often have repeated taped and transcribed. Data were analysed qualitatively.
admissions to hospital, are they indicative of HG sufferers Examples of some of the more common codes were;
whose needs have not been met in the community? The tiredness, depression, relentlessness of symptoms, weight
HIS questionnaire was, therefore, developed as a response loss, needing support, disbelief of health care professionals,
to the findings of the early project and as a clinical tool for problems with employment, nausea triggers, effect on
use in hospital. mental well-being, guilt at inactivity, the ineffectiveness of
The research questions were: what is the experience of oral anti-emetic therapy and when will it end? Codes
HG from the perspective of women with the condition and were then grouped into categories and presented back to
what factors lead to hospital admission for HG? Initially it the women during later interviews for verification of the
was planned to answer this question by focus groups with importance of the categories. Following verification the
women who had been recently admitted to hospital with categories were collapsed to form ten key factors which
HG. However, perhaps due to the inconvenience of this impact on women’s experience of HG (Fig. 1). These ten
method to the women (some of whom still felt unwell) we factors were triangulated with evidence from a literature
were unable to recruit even a single woman to the study. review on HG, four focus groups with staff caring for women
We changed the design to interviews of women while with HG, a retrospective review of medical records over 2
hospitalised. Women appeared to find this method much years of women with HG and a survey of anxiety and
more acceptable and recruitment progressed. The aim was depression of women with HG (Table 1). As well as
to interview each woman up to three times. Once in discussing their experiences of caring for women with
trimester one, once in trimester two or three and once HG, later staff focus groups were also presented some of the
post-natally. Where women did not complete three women’s interview data and the proposed HIS questionnaire
interviews they were replaced by a woman of similar for comment and discussion. All of the evidence triangulated
gestation. supported the importance of psychosocial as well as
Thirty qualitative one to one interviews took place with physiological support for women with HG and also the
18 women who had been hospitalised with HG (eight relevance of the ten key factors underpinning the develop-
women had two or more interviews) about their experi- ment of the HIS questionnaire.
ences of the condition. For four of the women English was The evidence from phase one of the project showed that
not their first language, however only one required an on the whole, the impact of symptoms on women’s lives
interpreter. was the factor that made the difference in their experience
Interviews as a method in qualitative research have rather than degree of symptoms. Thus while most women
been well described (Rubin and Rubin, 2004; Silverman with HG experienced dehydration from the nausea and
2005; Lambert and Loiselle 2008). Guion (2006) contends vomiting it would impact on them in different ways
that ‘‘The goal of the interview is to deeply explore the depending on their circumstances, lifestyle and ability to
respondent’s point of view, feelings and perspectives. . .’’. cope. Thus an individualised plan of care was needed based
Probing for deeper meaning and understanding of the not only on women’s physical symptoms but also on their
responses is possible in a semi-structured interview and individual psychological and social needs. A lengthy in-
necessary in order to get a fuller understanding of the depth interview of the type used in our research would
experience under investigation. Repeated interview ses- help health care professionals better understand the
sions offer an opportunity to gain an insight into the problems and issues facing women with HG. However,
phenomenon over time. in a time pressed environment this is rarely going to be
Interviews were semi-structured. In this approach, the possible. A systematic method of assessing the most
interviewer has an outline of topics or issues to be covered, important issues for individual woman was needed that
but is free to vary the wording and order of the questions to could be used in a clinical setting. The HIS was constructed
some extent. In this way key topics are covered while the for this purpose. The ten key factors established in phase
70 Z. Power et al. / International Journal of Nursing Studies 47 (2010) 67–77

Fig. 1. Important problems affecting women with hyperemesis gravidarum.

Table 1
Example evidence supporting the formulation of questions for the HIS questionnaire.

HIS question Examples of interview evidence from women Other evidence to support question formulation
to support question formulation

1. Are you able to keep small drinks ‘‘You think maybe it is just going to the morning, Fluid balance charts from patient record review.
down without vomiting? a one-off, but then suddenly you know it is twice Literature review (Attard et al., 2002; Bailit,
in an hour and then it is retching and then you 2005; Gadsby et al., 1993; O’Brien et al., 2002;
just can’t stop’’ Soltani and Taylor, 2003)
‘‘I felt that I was dying. I was completely dry, Focus groups
I couldn’t even sip water, I couldn’t even ‘‘they are actually admitted dehydrated, they’ve
swallow, I had no saliva.’’ been vomiting, not eating, which is how they
will present most times’’

2. Are you able to keep small amounts ‘‘Anything I’ve eaten since Sunday has just been As above
of food down without vomiting? coming out. So yesterday, all day, I just didn’t
eat anything. You’re starving as well, which is
funny, because you want to eat as well, but
you can’t keep it in.’’

3. How tired do you feel? ‘‘I was woken up feeling sick, I couldn’t get to sleep, Literature review (Attard et al., 2002; O’Brien
because I was getting up every ten/fifteen minutes et al., 2002)
feeling sick. It was all through the day so I wasn’t Focus group
sleeping properly. . . I couldn’t talk I didn’t have
‘‘They’ll limp down the corridor. They can’t walk
no energy’’
properly and not exaggerating.’’
‘‘just to have a bath or a shower would take so
much energy up and I would just think no
I can’t, I can’t – just to get out of bed’’

4. Think back to your usual mood/ ‘‘but it is the most miserable thing I’ve ever had Literature review (Attard et al., 2002; Mazzotta
emotional state before you felt because I’ve never really been ill before’’ et al., 2000; O’Brien et al., 2002; Poursharif
sick, how do you feel now in et al., 2008; Soltani and Taylor, 2003)
comparison? ‘‘Very depressed because the main thing depressing Survey of anxiety and depression (see Table 2).
was not the baby, it’s the work. Mainly because Focus groups
if I take off sick, people won’t be happy at all
at work.’’

5. Do you worry about the health ‘‘the main thing on your mind when you are not Literature review
of your unborn baby? eating, is the baby not eating.’’ Focus groups
‘‘I was quite worried during that time about my baby, ‘‘when a patients vomiting, the first thing
even though it wasn’t my fault, I was trying to eat that they ask is, because their vomiting
something, but I couldn’t eat anything. so vigorously, ‘‘will it disrupt the pregnancy’’
 Z. Power et al. / International Journal of Nursing Studies 47 (2010) 67–77 71

Table 1 (Continued )
HIS question Examples of interview evidence from women Other evidence to support question formulation
to support question formulation

6. Do you feel defeated by your nausea ‘‘nothing, no anti-sickness, no anti-emetics, nothing Literature review (O’Brien et al., 2002)
and vomiting and that nothing helped at all. . .We tried everything, but nothing helped.’’ Focus groups
will work to make you feel better? ‘‘they think nothing will work for me it’s going
to be that bad and just to make them
psychologically well, it’s quite a bit of effort.’’

7. Do your symptoms effect your ‘‘this is taking over my life and I am not able to look Literature review (Attard et al., 2002; O’Brien
ability to look after your home after my son at the moment without help of other et al., 2002; Poursharif et al., 2008; Soltani
and/or family? members of family and its unfair on him as he is unable and Taylor, 2003)
to understand and he gets upset ‘cos he knows I’m not
well so he is extra clingy, you know if I’m not well I want
to be just left alone but obviously at the same time
I want to be able to care for him and play with him
and everything, which at the moment I am not able to do.’’
Researcher-Your little girl she’s on holiday at the
moment, how do you manage with that?
Pt.-I couldn’t, my Mum’s got her. I couldn’t cope.
I couldn’t even get her dressed, I couldn’t even get
me up and about, so I couldn’t look after *****. I couldn’t.

8. Do you feel people understand ‘‘it was like he wouldn’t believe what I’m saying, Literature review (Munch and Schmitz, 2006;
how ill you are feeling? that I’ve been sick and everything, literally. He O’Brien et al., 2002; Poursharif et al., 2008;
was just like, nothing’s wrong with you and I was Soltani and Taylor, 2003)
like. . ., that day I was so bad, I couldn’t even talk Focus groups
properly, so I felt a bit terrible’’ ‘‘Some of these ladies actually believe the hospital
‘‘I felt dismissed many, many times.’’ is a 5 star hotel, where you can just come and
chill out and get taken care of and will have the
nurses doing everything for them, get me this get
me that. Too lazy, won’t stand up and get something
from the cupboard. Some of them have this notion
that that is what the ward is supposed to be and
these are the type that will keep coming back with
vague symptoms and no matter what you do.’’

9. Do your symptoms effect your ‘‘I haven’t been able to go to work. I tried a couple Literature review (Attard et al., 2002; O’Brien
worklife? of times in the last two weeks and had to come et al., 2002; Poursharif et al., 2008)
back early.’’
‘‘Even when the vomiting stopped, I could not return
to work as I felt dizzy.’’

10. Do your symptoms effect your ‘‘you can’t do anything yourself; totally depending on Literature review (O’Brien et al., 2002; Soltani
ability to look after yourself? others, especially my husband. I couldn’t even stand and Taylor, 2003)
up myself last time; even to go to the toilet I had my
husbands help. I had no strength at all even to get up.’’
‘‘sometimes just going up the stairs to have a bath
drains you so very much, so completely. By the time
I get back downstairs, my legs are shaking.’’

one formed the foundation of the ten questions of the HIS 3.3. Phase two: assessment of validity and reliability of the
questionnaire. Each question response was scored on the HIS
basis of severity of impact it represented.
The validity and reliability of the HIS was assessed using
3.2. Pilot testing of HIS a consecutive sample of women: 50 women admitted to
hospital with HG and 50 women recruited from ante-natal
After its development, the HIS was initially piloted on 11 clinic with an uncomplicated pregnancy. As the intention
women with HG aged 16 years or older who were less than of the HIS is to assess the impact of HG on women and not
16 weeks pregnant and admitted to the wards at a women’s to be a diagnostic tool, the study was not powered to detect
hospital in the North West of England. The diagnosis of HG specific differences between a group of controls and a
was given by the medical staff admitting the women group of women with HG. In addition the statistical
according to hospital protocol. Hospital admission criteria properties of the HIS score were not yet known, therefore it
included a history of unremitting nausea and vomiting, was not possible to base a sample size calculation on the
signs of clinical dehydration and presence of ketones on score itself. However, using formula for a confidence
urinalysis. The questionnaire was administered as soon as interval (CI) of Cronbach’s alpha given by Fan and
possible after admission. Feedback was positive and analysis Thompson (2001), a sample of size 100 would lead to a
of responses and comments led to the alteration of two 95% CI of 0.60–0.78 for an alpha of 0.70 and one of 0.74–
questions because they lacked clarity. 0.85 for an alpha of 0.80. A sample size of 100 (50 per
72 Z. Power et al. / International Journal of Nursing Studies 47 (2010) 67–77

group) was therefore judged to be adequate to give As there were no other symptom impact scores for HG
reasonable numbers in the various categories to demon- available, criterion validity was assessed by comparison
strate sufficient validity to warrant further study. The with other validated measures of concepts included in the
inclusion criteria for the control arm of the study were HIS. The concepts chosen for assessment were quality of
pregnant women aged 16 years or older unaffected by HG life, anxiety and depression and nausea and vomiting
or severe NVOP and attending for their first visit to the because it was felt these were the measurable concepts
ante-natal clinic. Data were collected during the ante-natal that most closely reflected the key factors underpinning
clinic visit. the construction of the HIS. The Short Form 12 version two
All women aged 16 years or older, admitted to the Quality of Life Assessment (SF12v2) (Ware et al., 1996), the
sample hospital with a diagnosis of HG who were willing Hospital Anxiety and Depression Score (HADS) (Zigmond
and able to give informed consent were eligible for and Snaith, 1983) and the Pregnancy Unique Quantifica-
recruitment to the HG arm of the study. Women unable tion of Emesis (PUQE) (Koren et al., 2002, 2005) were
to comprehend and speak English were excluded unless an selected respectively to measure these concepts as they are
interpreter was found. Every effort was made to make all reliable validated measures as well as relatively short.
interpreters available. As before, the diagnosis of HG was Brevity was an important factor in choice of questionnaires
given by the medical staff according to hospital protocol; for the women with HG, who felt very unwell. Recently a
all women presented with a history of unremitting nausea nausea and vomiting of pregnancy (NVOP) specific quality
and vomiting, signs of clinical dehydration and presence of of life questionnaire has been validated (Lacasse and
ketones on urinalysis. Data were collected during hospi- Berard, 2008). However when our study was undertaken
talization, as soon as possible after admission. insufficient validation data were available to warrant its
Content validity was achieved by deriving the symp- use. Therefore, the SF12v2 was chosen. Both groups of
toms of HG from phase one of the research as described women were asked to complete all questionnaires. Data
earlier in this paper. This was achieved by interviewing did not meet the assumptions for Pearson’s product–
women, conducting focus groups with staff, undertaking a moment correlation and so Spearman’s rank order
retrospective review of medical records and by surveying correlation was used to compare SF12v2, PUQE and HADS
women for anxiety and depression to better understand scores with HIS scores. Strength of correlation was
the constructs underlying the diagnosis of HG in order to interpreted using the guidelines suggested by Cohen
develop the 10 key categories that led to the development (1988) and Pallant (2007).
of the HIS questions (see Table 1). In the group of women with HG, the tool was also
Face validity was assessed by interviews with a random measured for construct validity. This was against accepted
sample of 20% of the main sample of women with HG (10 markers of HG as surrogate, as there was no measure of
women). Interviews were semi-structured and ranged up impact of symptoms of HG (Fig. 2). Spearman’s correlation
to 45 min in length. Topic areas focused around the was again used.
experience of hospitalization for HG and the appropriate- For discriminant validity, the t-test was used to test for
ness and utility of the HIS. Interviewing stopped when data differences in the mean HIS scores between the control
saturation was reached. All interviews were conducted by group and HG group. Where a sample size is over 40 and
ZP. All of the interviews were digitally recorded and the mean is a good measure of the centre, a t-test is
transcribed verbatim. No interviews in this group required considered appropriate (Moore and McCabe, 2003).
an interpreter, although one woman’s first language was Data were cross-tabulated and the chi-square test for
not English. The data from the interviews were analysed trend (Bland, 2000) was used to assess the difference in
separately using content analysis to see if there were any responses to the questions of the HIS between the control
differences in viewpoint. The transcripts of all interviews group and HG. Pearson’s chi-square test for independence
were searched for any information on the appropriateness, and where necessary, the Freeman–Halton extension to
feasibility and usefulness of the HIS. The meaning of each Fisher’s exact test (Freeman and Halton, 1951; Mehta and
was coded and each code grouped into a category. These Patel, 1996) were used to compare nominal characteristics
categories formed the coding frame for analysis and were of the women between groups, while the t-test was used to
applied to each interview. test for differences in non-skewed interval characteristics

Fig. 2. Markers of severity of hyperemesis gravidarum.

 Z. Power et al. / International Journal of Nursing Studies 47 (2010) 67–77 73

and the Mann–Whitney U test was used to test for marital status and parity, but the control group had a
differences in skewed interval characteristics between higher mean gestational age (12.7 vs 9.8 weeks). There
groups. were also differences in the distribution of ethnic group
The reliability or internal consistency of the HIS was between the two groups, with twice as many Caucasian
assessed by estimating the Cronbach alpha coefficient. This women in the control compared with the HG group.
indicates homogeneity of the items of a questionnaire, that Almost all of the women sampled spoke English well
is, the extent to which the items of the questionnaire enough to understand the questionnaire without inter-
broadly measure the same construct (Bland and Altman, pretation. One woman in the control group and one
1997). We expected all the questions to form one woman in the HG group required interpretation.
construct. There were strong significant correlations between HIS
SPSS 13.0 was used for all statistical analysis. Data were scores and SF12v2, HADS and PUQE scores across the
first cleaned and then analysed descriptively before the combined sample (Table 3). Correlations between HIS
methods mentioned above were applied. scores and SF-12v2 scores were negative as higher scores
on the HIS indicate decreased well-being and higher scores
4. Results on SF12 components indicate improved quality of life and
well-being.
The number of women recruited to the study was 101; Amongst the women admitted with HG, there were
51 women with HG (54 approached, 3 declined) and 50 moderate correlations between the total HIS score and
controls (57 approached, 7 declined). Accrual took place markers of severity of HG (Table 4). Strong and moderate
between August 2006 and March 2007. Reasons provided correlations were found between the physically orientated
for refusal included not wanting to be bothered and feeling HISQ1 (Are you able to keep small drinks down without
too ill. The characteristics of the women in the study are vomiting?) and HISQ2 (Are you able to keep small amounts
shown in Table 2. There was no significant difference of food down without vomiting?) and the markers of
between the control group and women with HG for age, severity of HG. PUQE scores (which also measure the

Table 2
Additional studies carried out which informed the development of the HIS.

Aim Sample size

Retrospective survey To gather demographic, hospital (re)admission, 119

treatment, discharge and infant outcome data
Survey of anxiety and depression To assess the level of anxiety and depression in Control group of women without
a group of women with and without HG HG = 100 HG group = 90
Focus groups with staff To determine the experience of nursing and Four focus groups Sample size of
medical staff who care for women with HG each group was: 9, 9, 6, and 24.

Table 3
Characteristics of women recruited to the study.

Characteristic Control group (n = 50) Hyperemesis group (n = 51) Findings

Age (years) – mean (SD) 28.3 (5.72) 27.3 (6.64) Mean diff = 0.97, t = 1.19, df = 99, p = 0.44
2
Marital status x = 0.94, df = 2, p = 0.63
Single 7 (14%) 8 (16%)
Married 28 (56%) 32 (63%)
Lives with partner 15 (30%) 11 (22%)

Ethnicity Fisher’s exact test p = 0.007

White British 29 (58%) 17 (33%)
White European 4 (8%) 0 (0%)
Black British 2 (4%) 4 (8%)
Asian British 5 (10%) 6 (12%)
African 1 (2%) 6 (12%)
Middle Eastern 1 (2%) 6 (12%)
Asian 4 (8%) 10 (20%)
Mixed 2 (4%) 2 (4%)
Other 2 (4%) 0 (0%)

Paritya Mann–Whitney Z = 1.62, p = 0.11

0 20 (42.6%) 27 (57.4%)
1 17 (54.8%) 14 (45.2%)
2 8 (53.3%) 7 (46.7%)
3 3 (75.0%) 1 (25.0%)
4 1 (100%) 0 (0%)
5 or more 2 (66.7%) 1 (33.3%)

Gestation (weeks) – mean (SD) 12.7 (0.35) 9.8 (0.47) Mean diff = 2.9, t = 4.84, df = 99, p < 0.001
a
Number of times have given birth to a baby of gestational age 24 weeks or more.
74 Z. Power et al. / International Journal of Nursing Studies 47 (2010) 67–77

Table 4
Spearman’s rho correlations between HIS score and validated measures across both groups.
SF12 mental component summary score rho = 0.65 p < 0.001
SF12 physical component summary score rho = 0.62 p < 0.001
SF12 individual component score, role physical rho = 0.65 p < 0.001
SF12 individual component score, bodily pain rho = 0.64 p < 0.001
SF12 individual component score, general health rho = 0.32 p < 0.001
SF12 individual component score, vitality rho = 0.54 p < 0.001
SF12 individual component score, social functioning rho = 0.68 p < 0.001
SF12 individual component score, role emotional rho = 0.59 p < 0.001
SF12 individual component score, mental health rho = 0.58 p < 0.001
PUQE score rho = 0.75 p < 0.001
HADS depression score rho = 0.76 p < 0.001
HADS anxiety score rho = 0.54 p < 0.001

HIS, Hyperemesis Impact of Symptoms Score; HADS, the Hospital and Anxiety Depression Score; SF12, Quality of Life Assessment; PUQE, Pregnancy Unique
Quantification of Emesis.

Table 5
Spearman’s rho correlations between total HIS score and HIS q’s 1 and 2 with markers of severity of HG.

HIS, Q.1 HIS Q.2 Total HIS score

Total nights admitted rho = 0.54, p < 0.001 rho = 0.287, p < 0.05 rho = 0.40, p < 0.05
Total admissions rho = 0.55, p < 0.001 rho = 0.35, p < 0.05 rho = 0.34, p < 0.05
Total presentations rho = 0.48, p < 0.001 rho = 0.28, p = 0.054 rho = 0.31, p < 0.05
Total IV therapy rho = 0.51, p = 0.001 rho = 0.331, p < 0.05 rho = 0.44, p < 0.05

HIS, Hyperemesis Impact of Symptoms Score.

physical symptoms of HG) and the physical components of ing, as it confirmed that their symptoms and reactions
the SF12 were found to have lower correlations with the were common for this condition and that they were not
markers of severity than total HIS scores and Q1 and 2 unique in experiencing them. Which question was found to
(Tables 5 and 6). be the most relevant varied from woman to woman,
There was a significant difference between the control suggesting where symptoms impact most on women’s
and HG group mean HIS scores (total HIS score: control lives varies from woman to woman. Inability to eat and
group mean 5.6, SD 4.7, 95% CI for mean 4.2–6.9, HG group drink without vomiting, tiredness, feeling defeated and
mean 16.3, SD 4.3, 95% CI for mean 15.1–17.5; t = 11.87, emotional state were mentioned the most often, but some
df = 98, p < 0.001, 95% CI for mean difference 8.9–12.5). found nearly all question areas to be of almost equal
Cross-tabulation of the women’s responses by group impact.
revealed clear delineation of responses to individual
questions and total scores between groups. The chi-square 5. Discussion
test for trend showed significant differences (p < 0.001) in
responses between control group and HG group to all The HIS questionnaire represents a single clinical tool
questions within the HIS. that appears capable of assessing holistically the impact of
The HIS showed good internal consistency with a HG for a particular woman. Used with the HIS handbook, it
Cronbach alpha 0.87, split half 0.80. Cronbach alpha for the can be used to design a tailored care plan. Face validity was
validation instruments in this group were: SF12 shown in the qualitative evaluation, as women found all
alpha = 0.88, split half 0.88. PUQE alpha = 0.85, split half questions very relevant to their condition. Good criterion
0.74 and HADS anxiety alpha = 0.81, split half 0.75, HADS validity of the HIS was demonstrated by strong correla-
depression alpha = 0.89, split half 0.88. tions between the HIS and the SF12, the HADS and PUQE.
The overwhelming response of the women to the HIS The strong correlations between these measures suggest
was that they found all questions to be relevant to their that the HIS does indeed measure the areas of impact it
condition and did not feel that important aspects had been aims to do in a HG-specific context. Construct validity was
omitted. Furthermore, many women expressed finding also suggested by a degree of correlation between the
their experiences described by the questionnaire reassur- clinical markers of severity and HIS scores.

Table 6
Spearman’s rho correlations between the PUQE score and SF12 (physical component) scores with markers of severity of HG.

SF12 PCS SF12 PFS SF12RPS PUQE

Total nights admitted rho = 0.059, p = 0.701 rho = 0.071, p = 0.636 rho = 0.051, p = 0.734 rho = 0.29, p = 0.05
Total admissions rho = 0.028, p = 0.857 rho = 0.014, p = 0.924 rho = 0.014, p = 0.924 rho = 0.27, p = 0.07
Total presentations rho = 0.083, p = 0.589 rho = 0.052, p = 0.730 rho = 0.029, p = 0.847 rho = 0.29, p = 0.05
Total IV therapy rho = 0.018, p = 0.915 rho = 0.004, p = 0.979 rho = 0.004, p = 0.982 rho = 0.20, p = 0.21

SF12, Short form 12; PCS, physical component summary; PFS, physical functioning score; RPS, role physical score; PUQE, Pregnancy Unique Quantification
of Emesis.
 Z. Power et al. / International Journal of Nursing Studies 47 (2010) 67–77 75

Discriminant validity was shown by highly significant affected individual. This offers an adjunct to standard care
differences in HIS scores between women with HG and which usually aims simply to control vomiting. Using the
those without. There was no difference between the two HIS can promote the planning of holistic nursing care.
groups in parity, but the mean gestation in the control
group was higher, while there were more women from 6. Conclusions
minority ethnic groups in the hyperemesis arm. Women of
white European origin have also been found to have a The importance of the psychosocial as well as the
lower incidence of HG compared with other ethnic groups physical burden of HG has been clearly demonstrated in
in previous studies (Jordan et al., 1999; Vilming and this study, and therefore, a plan of care that has identified a
Nesheim, 2000). While these factors are unlikely to have woman’s individual concerns may be crucial in the
affected discriminant validity, they should be accounted successful management of HG. The HIS covers the
for in future research. manifestations of HG that mostly affect women, including
We compared the HIS with the SF12 and HADS, but they inability to tolerate diet and fluids, social dysfunction and
are generic scores so although helpful they do not pinpoint psychological distress, thus pinpointing the specific needs
an individual’s problems with HG in the same way the HIS of women. Our study has shown that the HIS is a valid and
does. While the PUQE is a quick and useful score for nausea reliable tool with which to assess the impact of the
and vomiting in pregnancy, it does not gather information physical and psychosocial symptoms of HG.
on other symptoms like social dysfunction which are
experienced by women with HG. A recently validated Contributors
quality of life tool (Lacasse and Berard, 2008; Magee et al.,
2002) is also different from the HIS as it measures NVOP in ZP undertook the research, devised the HIS, collected
general and not HG, and its purpose is for research rather and analysed data and co-wrote the article. MC provided
than for clinical assessment. It is not unreasonable to statistical support and contributed significantly to revi-
conclude therefore that the HIS in a unique clinical tool sions of the article. PK participated in setting up and
that has a different role to the other measures. conducting the research and in reviewing the article. HK
This study has developed and validated an evidence jointly conceptualised the design of the study, managed
based clinical tool, that is acceptable to women affected by the project and contributed significantly to revisions of the
HG. We found that in our sample the HIS questionnaire was article. HW jointly conceptualised the design of the study,
suitable for all women admitted with HG and could assist managed the project, and co-wrote the article.
care planning to address the problems caused by the
impact of HG on women’s lives. We believe that the HIS
Acknowledgements
does not need amendment in light of the findings. The
clinical effectiveness of the HIS and handbook in an
We would like to thank staff and patients at St. Mary’s
inpatient hospital setting is now being evaluated by a
Hospital, Manchester, UK who very kindly supported and
randomised controlled trial funded by the (UK) NIHR.
participated in the project. Without their input we could
Further testing is required of the HIS for its relevance in
not have carried out this research.
other countries and settings, for example in the commu-
nity, where it may be of particular benefit, and reduce the Conflicts of interest: None.
need for hospital admission.
Funding: The study was funded by a grant from the
A limitation of this study is that it was conducted in a
Burdett Trust for Nursing.
hospital setting and having been validated on women
requiring hospitalization may not be generalisable. Con- Ethical approval: Ethics approval was obtained from the
ducting a study in a hospital setting makes it more difficult local research ethics committee (REC reference number 04/
to disassociate the worries and problems caused by QQ1402/9) and all women and staff who agreed to
hospitalization from those caused by the condition. It is participate gave informed consent.
possible, therefore, that some confounding may have
occurred, also comparing the women with HG with a well
Appendix A. Hyperemesis Impact of Symptoms Score
comparison group who are not in hospital may have
exaggerated the differences between the two groups. The
bigger problem, however, appeared to be that it was Read the questions below. Tick the answer nearest to how
sometimes difficult to assess the impact of symptoms after you have been feeling for the last 24hrs (normally presented
hospitalization. This was because in many cases hospita- on two sides of A4):
lization had eased the burden of symptoms considerably.
The difference between the two groups, therefore, may in 1. Are you able to keep small drinks down without
fact have been larger than was actually detected, had the vomiting?
women been assessed at the point of admission, as would Yes, I can drink without vomiting.
be the case in practice rather than research. The ongoing I sometimes vomit, but I can often drink without
trial involves women from four hospitals. We anticipate vomiting.
that the clinical utility of the HIS will lie in its ability to More often than not I will vomit back even small
assess the impact of HG not only in terms of severity, but drinks.
also the elements of distress that are most relevant to the No, I can’t keep down any drinks.
76 Z. Power et al. / International Journal of Nursing Studies 47 (2010) 67–77

2. Are you able to keep small amounts of food down References

without vomiting?
Yes, I can eat without vomiting. ACOG, 2004. ACOG Practice Bulletin. Clinical management guidelines for
obstetrician-gynecologists number 52: nausea and vomiting of preg-
I sometimes vomit, but I can often eat without nancy. Obstetrics and Gynecology 103 (4), 803–814.
vomiting. Arsenault, M.-Y., Lane, C.A., MacKinnon, C.J., Bartellas, E., Cargill, Y.M.,
More often than not I will vomit back even small Klein, M.C., Martel, M.-J., Sprague, A.E., Wilson, A.K., 2002. The
management of nausea and vomiting of pregnancy. Journal of
amounts of food.
Obstetrics & Gynaecology Canada 24 (10), 817–831 (see comment)
No, I can’t keep down any food. quiz 832–813.
3. How tired do you feel? Attard, C.L., Kohli, M.A., Coleman, S., Bradley, C., Hux, M., Atanackovic,
About the same as usual (sometimes tired, some- G., Torrance, G.W., 2002. The burden of illness of severe nausea
and vomiting of pregnancy in the United States. American Journal
times not as tired). of Obstetrics & Gynecology 186 (5 Suppl. Understanding),
A bit more tired than usual. S220–S227.
Very tired, more than usual. It affects my work/ Bailit, J.L., 2005. Hyperemesis gravidarium: epidemiologic findings from a
large cohort. American Journal of Obstetrics & Gynecology 193 (3 Pt
home life. 1), 811–814.
Exhausted, very much more tired than usual. I want Bland, J.M., Altman, D.G., 1997. Cronbach’s alpha. BMJ 314 (7080), 572.
to sleep all the time. Bland, M., 2000. An Introduction to Medical Statistics. Oxford University
Press, Oxford.
4. Think back to your usual mood/emotional state before Cohen, J.W., 1988. Statistical Power Analysis for the Behavioral Sciences.
you felt sick, how do you feel now in comparison? Lawrence Erlbaum Associates, Hillsdale, NJ.
I feel happier than before. Dodds, L., Fell, D.B., Joseph, K.S., Allen, V.M., Butler, B., 2006. Outcomes of
pregnancies complicated by hyperemesis gravidarum. Obstetrics and
I feel pretty much the same as before.
Gynecology 107 (2 Pt 1), 285–292.
I feel less happy than before. Freeman, G.H., Halton, J.H., 1951. Note on an exact treatment of con-
I feel much more unhappy than before. tingency, goodness of fit and other problems of significance. Biome-
trika 38 (1/2), 141–149.
5. Do you worry about the health of your unborn baby?
Gazmararian, J.A., Petersen, R., Jamieson, D.J., Schild, L., Adams, M.M.,
I’m not especially worried about that. Deshpande, A.D., Franks, A.L., 2002. Hospitalizations during preg-
Yes, I’m a bit worried, but a lot of things worry me. nancy among managed care enrollees. Obstetrics and Gynecology
Yes, I’m worried more than usual, because I’m ill, 100 (1), 94–100.
Jewell, D., Young, G., 2006. Interventions for nausea and vomiting in early
that it might hurt the baby. pregnancy. The Cochrane Library (4(CD000145)).
I’m really frightened about the health of my baby. Jordan, V., Grebe, S.K., Cooke, R.R., Ford, H.C., Larsen, P.D., Stone, P.R.,
6. Do you feel defeated by your nausea and vomiting and Salmond, C.E., 1999. Acidic isoforms of chorionic gonadotrophin in
European and Samoan women are associated with hyperemesis
that nothing will work to make you feel better? gravidarum and may be thyrotrophic. Clinical Endocrinology 50
Yes, I’ve lost hope, nausea and vomiting have taken (5), 619–627.
over my life. Koren, G., Boskovic, R., Hard, M., Maltepe, C., Navioz, Y., Einarson, A., 2002.
Motherisk-PUQE (pregnancy-unique quantification of emesis and
Often, I feel nausea and vomiting have taken over nausea) scoring system for nausea and vomiting of pregnancy. Amer-
my life, but sometimes it’s manageable. ican Journal of Obstetrics & Gynecology 186 (5 Suppl. Understanding),
It depends, mostly I feel the nausea and vomiting S228–S231.
Koren, G., Piwko, C., Ahn, E., Boskovic, R., Maltepe, C., Einarson, A., Navioz,
are manageable.
Y., Ungar, W.J., 2005. Validation studies of the Pregnancy Unique-
Nausea and vomiting are not at all unmanageable Quantification of Emesis (PUQE) scores. Journal of Obstetrics &
for me. Gynaecology 25 (3), 241–244.
Lacasse, A., Berard, A., 2008. Validation of the nausea and vomiting of
7. Do your symptoms effect you’re ability to look after
pregnancy specific health related quality of life questionnaire. Health
your home and/or family? & Quality of Life Outcomes 6, 32.
No, I manage at home. Magee, L.A., Chandra, K., Mazzotta, P., Stewart, D., Koren, G., Guyatt,
A bit, some things don’t get done. G.H., 2002. Development of a health-related quality of life
instrument for nausea and vomiting of pregnancy. American
Yes, but I have help. Journal of Obstetrics & Gynecology 186 (5 Suppl. Understanding),
Yes, I can’t manage and I don’t have help. S232–S238.
8. Do you feel people understand how ill you are feeling? Mazzotta, P., Stewart, D., Atanackovic, G., Koren, G., Magee, L.A., 2000.
Psychosocial morbidity among women with nausea and vomiting
Yes, no problems. of pregnancy: prevalence and association with anti-emetic
Most people understand. therapy. Journal of Psychosomatic Obstetrics & Gynecology 21
Most people don’t understand. (3), 129–136.
Mehta, C.R., Patel, N.R., 1996. SPSS Exact Tests 7.0 for Windows. SPSS Inc.,
No one understands. Chicago.
9. Do your symptoms affect your work-life? Moore, D.S., McCabe, G.P., 2003. Introduction to the Practice of Statistics.
No, I manage to got to work. W.H. Freeman, Bedford.
Munch, S., 2000. A qualitative analysis of physician humanism: women’s
A bit, I go to work, but I don’t feel 100%. experiences with hyperemesis gravidarum. Journal of Perinatology 20
Yes, some days I can’t go to work. (8 Pt 1), 540–547.
Yes, I can’t go to work at all. Munch, S., Schmitz, M.F., 2006. Hyperemesis gravidarum and patient
satisfaction: a path model of patients’ perceptions of the patient–
N/A.
physician relationship. Journal of Psychosomatic Obstetrics & Gyne-
10. Do your symptoms effect your ability to look after cology 27 (1), 49–57.
yourself? (e.g. washing, preparing meals) O’Brien, B., Evans, M., White-McDonald, E., 2002. Isolation from ‘‘being
No I can manage. alive’’: coping with severe nausea and vomiting of pregnancy. Nursing
Research 51 (5), 302–308.
A bit, some things get missed. Paauw, J.D., Bierling, S., Cook, C.R., Davis, A.T., 2005. Hyperemesis grav-
Yes, but I have help. idarum and fetal outcome. Journal of Parenteral & Enteral Nutrition
Yes, I can’t manage it’s a problem. 29 (2), 93–96 (see comment).
 Z. Power et al. / International Journal of Nursing Studies 47 (2010) 67–77 77

Pallant, J., 2007. SPSS Survival Manual. A Step by Step Guide to Data Soltani, H., Taylor, G.M., 2003. Changing attitudes and perceptions to
Analysis Using SPSS for Windows. Open University Press, Maiden- hyperemesis gravidarum. RCM Midwives 6 (12), 520–524.
head. Vilming, B., Nesheim, B.I., 2000. Hyperemesis gravidarum in a contem-
Poursharif, B., Korst, L.M., Fejzo, M.S., MacGibbon, K.W., Romero, R., porary population in Oslo. Acta Obstetricia et Gynecologica Scandi-
Goodwin, T.M., 2008. The psychosocial burden of hyperemesis grav- navica 79 (8), 640–643.
idarum. Journal of Perinatology 28 (3), 176–181. Ware, J.E., Kosinski, M., Keller, S.D., 1996. A 12-item Short Form Health
Power, Z.L., Waterman, H., Kitchener, H., 2007. Exploring management of Survey: construction of scales and preliminary tests of reliability and
hyperemesis with women and their health care professionals. In: validity. Medical Care 34 (3), 220–233.
Royal College of Nursing International Nursing Research Conference, Zigmond, A.S., Snaith, R.P., 1983. The Hospital Anxiety and Depression
Royal College of Nursing, Dundee. Scale. Acta Psychiatria Scandinavia 361–370.