Opportunistic growth measurements have been recommended in the UK to complement universal measurement of a child’s growth at school entry. This study analysed the effectiveness of opportunistic growth measurements in a random sample of 200 children attending hospital for clinical reasons other than growth concerns. Height and weight were measured in only 12.5% and 51.5% of the children, respectively, and head circumference in 11.1% of children below the age of 2 years. The levels of height measurement were equally poor in all clinical areas, while significantly fewer children had weight measurements in outpatients and non-paediatric areas. The results suggest that the opportunity to check a child’s growth is seldom used in hospital settings. There is a need for an educational programme for health professionals to emphasise the value of growth as an index of health in children, and for the current strategy for growth monitoring in the UK to be re-evaluated.
Statistics from Altmetric.com
Guidelines for monitoring children’s growth in the United Kingdom have been recently updated.1 2 3 Apart from measuring weight during infancy and a single measurement of height and weight at school entry (4–5 years of age), no other systematic growth screening is recommended. However, opportunistic growth measurement is encouraged whenever children come to medical attention for reasons other than concerns regarding their growth. Such a policy has merit if not only are measurements performed but also the data recorded are easily retrieved to enable a referral for specialist assessment to be made if necessary. However, recent publications on evidence-based guidelines for the referral of children with short stature4 and the value of growth screening to detect disease5 have highlighted concerns about the current policy of growth monitoring in the UK.6
What this study adds
Opportunistic growth measurement in children attending hospital is undertaken infrequently.
An educational programme for health professionals is required to stress the value of growth as an index of health in children.
The current UK strategy for growth monitoring merits re-evaluating to ensure the potential for growth as a marker of a healthy childhood is realised.
Since opportunistic growth measurement is an important component of this policy, we undertook an analysis of its effectiveness in a random sample of children attending a large UK hospital for clinical reasons unrelated to growth concerns.
Addenbrooke’s Hospital in Cambridge is a 1000-bed facility which provides secondary and tertiary level services to patients of all ages. Children access hospital services in various clinical areas provided by the Children’s Unit (“paediatric areas”) and by other departments (“non-paediatric areas”), either to attend outpatient appointments (“outpatient areas”) or to be admitted onto the wards or day surgical units (“inpatient areas”).
During the 12-month study period from November 2006 to October 2007, 77 146 clinical episodes involving children aged 0–16 years were recorded, excluding attendance in the accident and emergency department. Of the total, 42.5% and 57.5% were in paediatric and non-paediatric areas, respectively. Outpatient attendances comprised 87.1% of the episodes and inpatient attendances the remainder. Each of these clinical episodes provided an opportunity for the child to be measured.
We randomly selected 200 clinical episodes as follows: 20 in paediatric outpatient areas (dedicated children’s clinics), 50 in paediatric inpatient areas (dedicated children’s wards), 100 in non-paediatric outpatient areas (clinics in orthopaedics, immunology and allergy, ophthalmology, dermatology, dental and maxillofacial surgery, and ear, nose and throat) and 30 in non-paediatric inpatient areas (day surgical units and wards). The number of selected episodes from each clinical area reflected the relative proportion of all episodes during the study period. Random selection of the episodes was carried out by the clinical information department; the authors were blinded from this randomisation process. Random selection of the under 2-year-old children was from outpatient areas only, since opportunistic growth measurements in such young children may not be practical in those requiring hospital admission.
Each selected episode represented a different child. The medical record was reviewed to determine if a height (or length), weight and head circumference (if below 2 years old and attending outpatient areas) had been measured during the selected episode and recorded. The number of episodes with measurements was added, tabulated and analysed. Comparisons were made between measurements recorded in children below and above 2 years of age and between the chosen clinical areas. Differences were considered statistically significant if p<0.05 (using χ2 and Fisher’s exact tests). Approval by an ethics committee was not necessary for this study.
A total of 200 children were randomly selected, each of whom had one clinical episode. Mean (SD) age was 7.86 (4.54) years for the total group and 0.86 (0.59) for the 18 children below 2 years of age. The selected clinical episodes were distributed over all 12 months of the study period; the median number of children seen by month was 17 (range 11–23). Table 1 summarises the growth measurements and provides age group comparisons as well as an analysis between the different clinical areas.
Height was measured in only 25 (12.5%) of the total study group. When analysed by age, just one child under 2 years of age was measured for height (or length) (5.6%), while 24 older children (13.2%) had their heights measured. There was no statistical difference between the two age groups (p = 0.71). No significant differences in height measurement rates were found when the following were compared: 18 of 120 children (15%) in outpatient areas, seven of 80 children (8.8%) in inpatient areas, 10 of 70 children (14.3%) in paediatric areas, and 15 of 130 children (11.5%) in non-paediatric areas.
Among the paediatric areas, none of the 50 children (0%) in the inpatient areas had height measured, a result significantly worse than in outpatient areas, where 10 of the 20 children (50%) had height measured (p<0.001). For non-paediatric areas, a satellite day surgical unit (DSU) near Cambridge had significantly higher height measurement rates compared to the DSU in Addenbrooke’s Hospital and other outpatient areas. Significantly fewer children in the non-paediatric outpatient areas (eight of 100) had height measured compared with paediatric outpatient areas (10 of 20; p<0.001; see table 1).
Weight was measured in 103 of the total group (51.5%). Six of 18 (33.3%) and 97 of 182 children (53.3%) below and above 2 years of age, respectively, had weight measured. There was no statistically significant difference between the two age groups (p = 0.14).
Rates of measurement were significantly higher in paediatric areas compared with non-paediatric areas, and in inpatient areas compared with outpatient areas. When children were admitted to hospital, weight was measured in at least 96% of cases whether the child was in a paediatric or a non-paediatric area (p = 1.00).
Head circumference was measured in only two of the 18 children below the age of 2 years (11.1%). These two children were among six children who had attended paediatric outpatient areas, while none of the 12 children who had attended non-paediatric outpatient areas had their head circumference measured (p = 0.10).
Our analysis indicates that there were ample opportunities for growth measurements to be undertaken. There were more than 77 000 clinical episodes per year involving children, with 60% occurring in non-paediatric areas. However, the opportunity to check a child’s growth during these clinical episodes was seldom used.
Overall, height was less frequently measured than weight. Weight was recorded in almost all children who were admitted to inpatient areas (both paediatric and non-paediatric). This reflects the importance of having weight measurements for calculating drug doses in children. In contrast, there is no such pressing clinical need in outpatient areas, as suggested by similar levels in new and follow-up cases.
Clinical episodes for this study were randomly selected over a 12-month period, thus minimising any possible seasonal effect. Attendance episodes at the accident and emergency department were excluded as this clinical setting is often not conducive for growth monitoring. Similarly, inpatients under 2 years of age were also excluded.
Other than notable exceptions where a standalone children’s hospital offers a comprehensive service across a range of areas, the provision of hospital-based children’s services in the UK is generally within the framework of a large hospital predominantly catering for adults. A previous practice audit in the children’s unit of another hospital with such a set-up also showed poor rates of growth measurement.7 Thus, the experience in Cambridge may not be atypical of many other centres in the UK.
All clinical areas in this study were equipped with basic measuring equipment, illustrating that a major educational programme is needed for health care staff to rectify the problem. It is unrealistic to believe that such opportunistic measurements would constitute a growth screening programme. Rather, we need to inculcate a process whereby a growth chart is a routine constituent of the formulated set of medical notes (birth to 16 years) as is the temperature chart or a prescription chart. After all, taking the temperature is not a screen for pneumonia or urinary tract infection, it is just sound clinical practice.
For growth measurements to be meaningful, they must be performed accurately and reliably. Outside a research environment, the hospital setting should be a place that offers the opportunity to conduct reliable growth measurements by trained healthcare professionals. This study illustrates that such opportunities are currently missed in this particular hospital setting.
Most health care for children in the UK is undertaken in primary care. This study did not address the effectiveness of opportunistic growth measurements in a community setting. However, we did notice that among the outpatient episodes, few, if any, of the general practitioners’ referral letters contained the height and weight of the child. There is evidence that less than half of general practitioners surveyed in Melbourne, Australia routinely measure children’s height and weight in their practices.8 It is unclear whether a similar survey undertaken in the UK would produce more encouraging results. What we do know is that even the recommendation for a single height measurement at school entry was found to be poorly adhered to in the community: only two thirds of children in a South Wales study were measured and the results then plotted on a growth chart in only 25% of cases.9
Obesity is now high on the agenda as a disorder which captures the attention of health care professionals who are ardent proponents of screening. There is evidence that parents may not recognise their children as being overweight or obese.10 Since body mass index (BMI) is currently the biomarker for obesity (calculated as weight in kg divided by height in metres squared), those who set health policies in the UK should re-evaluate the most effective means of utilising growth as a useful marker of healthy childhood. Opportunistic growth measurement does not appear to be the answer.
We thank the Hospital Medical Records and Clinical Information Departments for providing data on clinical episodes and for random selection of the episodes used in this study. The Hospital Audit Department was responsible for retrieving the patient records.
Competing interests IAH has received lecture fees from Pfizer and Novo Nordisk pharmaceutical companies.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.