Paediatric nutrition risk scores in clinical practice: children
with inammatory bowel disease A. E. Wiskin,* D. R. Owens, V. R. Cornelius,* S. A. Wootton* & R. M. Beattie *NIHR Biomedical Research Unit (Nutrition, Diet & Lifestyle), Southampton, UK University of Southampton, Faculty of Medicine, Southampton, UK Paediatric Medical Unit, University Hospital Southampton NHS Foundation Trust, Southampton, UK Introduction Despite different denitions of malnutrition, nutritional assessments based on height and weight demonstrate that between one-fth and one-quarter (Pawellek et al., 2008; Joosten et al., 2010) of paediatric inpatients are malnour- ished. Recent guidance from the British Association of Parenteral and Enteral Nutrition suggests that, in addition to plotting growth measurements on an appropriate growth chart, tools for detecting nutritional risk should also be employed (Brotherton et al., 2010). This is sup- ported by European Society for Paediatric Gastroenterol- ogy, Hepatology and Nutrition guidance indicating that one of the main functions of a nutrition team is to screen for nutrition risk (Agostini et al., 2005). Four non-disease specic nutrition screening tools designed for paediatrics have been developed for use: Screening Tool for the Assessment of Malnutrition in Paediatrics (STAMP) (McCarthy et al., 2008); Screening Tool for Risk On Nutritional status and Growth (STRONGkids) (Hulst et al., 2010); Paediatric Yorkhill Malnutrition Score (PYMS) (Gerasimidis et al., 2010); Keywords inammatory bowel disease, nutrition risk, nutrition risk screening, paediatrics. Correspondence R. M. Beattie, Paediatric Medical Unit, Southampton General Hospital, Tremona Road, Southampton, Hampshire S016 6YD, UK. Tel.: +44 (0)2380 798688 Fax: +44 (0)2380 796888 E-mail mark.beattie@suht.swest.nhs.uk How to cite this article Wiskin A.E., Owens D.R., Cornelius V.R., Wootton S.A. & Beattie R.M. (2012) Paediatric nutrition risk scores in clinical practice: children with inammatory bowel disease. J Hum Nutr Diet. doi:10.1111/j.1365-277X.2012.01254.x Abstract Background: There has been increasing interest in the use of nutrition risk assessment tools in paediatrics to identify those who need nutrition support. Four non-disease specic screening tools have been developed, although there is a paucity of data on their application in clinical practice and the degree of inter-tool agreement. Methods: The concurrent validity of four nutrition screening tools [Screening Tool for the Assessment of Malnutrition in Paediatrics (STAMP), Screening Tool for Risk On Nutritional status and Growth (STRONGkids), Paediatric Yorkhill Malnutrition Score (PYMS) and Simple Paediatric Nutrition Risk Score (PNRS)] was examined in 46 children with inammatory bowel disease. Degree of malnutrition was determined by anthropometry alone using World Health Organization International Classication of Diseases (ICD-10) criteria. Results: There was good agreement between STAMP, STRONGkids and PNRS (kappa > 0.6) but there was only modest agreement between PYMS and the other scores (kappa = 0.3). No children scored low risk with STAMP, STRONGkids or PNRS; however, 23 children scored low risk with PYMS. There was no agreement between the risk tools and the degree of malnutrition based on anthropometric data (kappa < 0.1). Three children had anthropome- try consistent with malnutrition and these were all scored high risk. Four chil- dren had body mass index SD scores < )2, one of which was scored at low nutrition risk. Conclusions: The relevance of nutrition screening tools for children with chronic disease is unclear. In addition, there is the potential to under recognise nutritional impairment (and therefore nutritional risk) in children with inam- matory bowel disease. Journal of Human Nutrition and Dietetics 2012 The Authors Journal of Human Nutrition and Dietetics 2012 The British Dietetic Association Ltd. 1 and Simple Paediatric Nutrition Risk Score (PNRS) (Sermet-Gaudelus et al., 2000). These tools all attempt to classify children into three nutrition risk categories; low, medium or high. The tools contain different components and therefore may not be freely interchangeable. There is a paucity of research on the application of these tools to specic conditions and it is not clear which tool is best suited for what purpose. It is widely considered that children with inammatory bowel disease (IBD) are at high nutritional risk by deni- tion. However, experience in our clinic suggests that, although many children are underweight, most are of normal weight, and some are overweight for their height (Wiskin et al., 2011). A screening tool should also be able to detect those already malnourished, as well as detect those at nutri- tional risk. The present study aimed to evaluate the con- current validity of these screening tools in children with IBD in comparison with an objective nutritional assess- ment made by anthropometry, using WHO International Classication of Diseases (ICD)-10 criteria. Materials and methods Study design and setting This was a prospective observational study of children recruited from the regional paediatric gastroenterology service between December 2009 and June 2010. Children attending outpatient clinics and those requiring inpatient stay were recruited. Ethics approval was granted from the local research ethics committee. Subjects All children had IBD, which was conrmed histologically in accordance with international criteria (Silverberg et al., 2005) and treated in accordance with published guidelines (Sandhu et al., 2010). Data collection The four nutritional screening tools (STAMP, STRONGk- ids, PYMS and PNRS) were consolidated into one generic assessment from which the scores for each tool were derived. In most cases, the original questions were used but, in a few areas, respondents were asked to provide numerical rather categorical answers. Nutritional risk was determined from each tool. All observations were com- pleted by one observer. Height and weight were recorded and converted to SD scores (SDS) using lms growth software (Harlow Healthcare, South Shields, UK; http:// www.healthforallchildren.co.uk) and the UK 1990 data- sets. Malnutrition was dened using the anthropometric component of ICD-10 into none or mild, moderate or severe; weight SDS < )2, )2 to < )3 and )3, respec- tively. A SDS of 2 is approximately equal to the second centile and an SDS of 3 is less than the 0.4th centile. Statistical analysis Kappa values were calculated to assess the level of agree- ment between each risk score compared to that expected by chance. Statistical analysis was performed using spss, version 16.0 (SPSS Inc., Chicago, IL, USA). Results Forty-six children (25 boys) were studied. Median age was 14.6 years (range 317 years). Of these children, 27 had Crohns disease, 16 had ulcerative colitis and three had indeterminate colitis. Median (25th, 75th percentile) for height SDS, weight SDS and body mass index (BMI) SDS were )0.19 ()1.08, 0.52), )0.3 ()0.87, 0.15) and )0.43 ()1.09, 0.29). Three children had weight SDS < )2 (i.e were malnourished according to ICD-10). Four chil- dren were underweight for their height (BMI SDS < )2) and only one of these had a weight SDS < )2. In addi- tion, two children were short for age (height SDS < )2). No children scored low risk with STAMP, STRONGk- ids or PNRS (Table 1) and there was good agreement between these three tools. Similar numbers of children (1820) were scored high risk by all four tools. Half of the children scored low risk with PYMS and there was only a modest level of agreement between PYMS and each of the other scores. Of the children scored at low risk with PYMS, one scored high risk with STAMP and four scored high risk with PNRS. Table 2 demonstrates the kappa val- ues for the overall agreement between the scores. Three children had anthropometry consistent with moderate or severe malnutrition according to ICD-10, and these were all scored high risk. Interestingly, of the four children with BMI SDS < )2, one was scored at low risk by PYMS, medium risk by STRONGkids and high risk by the other score. Two children had height for age < )2 SDS (stunted) and were attributed high risk by all of the tools. Children who were not malnourished according to ICD-10 were scored mainly at medium or high risk by the screening tools, leading to a lack of agreement between any of the scores and the degree of malnutrition. Discussion There was good agreement between STAMP (McCarthy et al., 2008), STRONGkids (Hulst et al., 2010) and PNRS (Sermet-Gaudelus et al., 2000) but not between PYMS Paediatric nutrition risk scores in clinical practice A. E. Wiskin et al. 2012 The Authors 2 Journal of Human Nutrition and Dietetics 2012 The British Dietetic Association Ltd. (Gerasimidis et al., 2010) and the other scores. There was no agreement between the risk tools and the degree of malnutrition based on anthropometric data. The three scores with good agreement between each other automati- cally scored children with IBD as at least medium nutri- tion risk, simply on the basis of having IBD. The lack of this element within PYMS accounts for the poor agree- ment with the other scores. From a simple nutritional assessment based on height and weight, several children had obvious anthropometric abnormalities. It is of inter- est that both PYMS and STRONGkids did not score all these children at high nutritional risk, which is relevant to any clinical application of the tools. Most children had acceptable weight for height and were therefore not mal- nourished, whereas most scores placed the majority of children at moderate nutritional risk. Children with IBD are a heterogenous group, some of whom are malnourished and some are overweight and at risk of obesity (Wiskin et al., 2011). There is therefore an opportunity within this group to attribute different nutri- tion risk scores and different management outcomes. In the present study, children exhibited a range of height SDS and weight SDS and had a range of nutrition risk scores. Despite the small number of patients studied, the poor agreement between risk tools and poor agreement with anthropometry raises the question of what does it mean to be a child at nutritional risk? The tools employed in the present study have been designed to look at different outcomes. The team who developed the PNRS (Sermet-Gaudelus et al., 2000) state that their aim was to develop a score to identify children at risk of acute mal- nutrition during hospitalisation; however, there is no evi- dence provided indicating that their outcome of 2% weight loss is related to the development of acute malnu- trition. Indeed, 45% of their study group lost >2% weight. STRONGkids was tested in a national survey of 424 children. In their study (Hulst et al., 2010) using this tool the prevalence of a signicant anthropometric abnor- mality (weight for height < )2 SDS or height for age < )2 SDS) in low risk children was 12%. Were these children really low risk, or should they have been identi- ed by a screening tool to enable delivery of nutritional support? The four-stage evaluation of the PYMS tool pri- marily determines whether nurses using the tool attrib- uted the same nutritional risk as a dietician assessment. As yet, STAMP has not been published, except as an abstract, and so details of its validation are limited. Experience in our hospital (Moon et al., 2009) and elsewhere (Sullivan, 2010) suggests that routine height and weight measurement is poorly performed; therefore, it is uncertain how uniformly a nutrition screening tool Table 1 Cross-tabulation of nutrition risk using four nutrition screening tools and the degree of malnutrition described by International Classica- tion of Diseases (ICD)-10 STAMP STRONGkids SPNRS PYMS Low Medium High Low Medium High Low Medium High Low Medium High ICD-10 None/low 0 28 15 0 27 16 0 26 17 23 5 15 Moderate 0 0 2 0 0 2 0 0 2 0 0 2 Severe 0 0 1 0 0 1 0 0 1 0 0 1 STAMP Low 0 0 0 0 0 0 0 0 0 Medium 0 25 3 0 24 4 22 5 1 High 0 2 16 0 2 16 1 0 17 STRONGkids Low 0 0 0 0 0 0 Medium 0 22 5 23 3 1 High 0 4 15 0 2 17 SPNRS Low 0 0 0 Medium 19 4 3 High 4 1 15 ICD, International Classication of Diseases; PNRS, Paediatric Nutrition Risk Score; PYMS, Paediatric Yorkhill Malnutrition Score; STAMP, Screening Tool for the Assessment of Malnutrition in Paediatrics; STRONGkids, Screening Tool for Risk On Nutritional status and Growth. Table 2 Kappa values showing the level of agreement of nutrition risk scores STRONGkids SPNRS PYMS ICD10 STAMP 0.774 0.732 0.332 )0.014 STRONGkids 0.600 0.270 )0.013 SPNRS 0.236 )0.013 PYMS 0.079 A kappa value >0.6 represents a good level of agreement, <0.2 is poor. ICD, International Classication of Diseases; PNRS, Paediatric Nutrition Risk Score; PYMS, Paediatric Yorkhill Malnutrition Score; STAMP, Screening Tool for the Assessment of Malnutrition in Paediatrics; STRONGkids, Screening Tool for Risk On Nutritional status and Growth. A. E. Wiskin et al. Paediatric nutrition risk scores in clinical practice 2012 The Authors Journal of Human Nutrition and Dietetics 2012 The British Dietetic Association Ltd. 3 may be applied. Despite differences in scoring and the subsequent suggested management of children, there is no clear evidence of the impact of different scores on patient outcomes. In addition, would the use of a nutrition screening tool provide extra benet to routine height and weight monitoring combined with a recent diet history? Further discussion on the role of nutrition screening tools in paediatrics is necessary (Sullivan, 2010). Acknowledgments A conference poster was previously presented at UEGW 2010. Conict of interest, source of funding and authorship The authors declare that they have no conict of interests. This study was funded by the National Institute for Health Research Biomedical Research Unit (Nutrition, Diet & Lifestyle) in Southampton. All authors contributed to the design and analysis of the study and all contributed to the nal manuscript. DRO carried out the data collection. All authors critically reviewed the manuscript and approved the nal version submitted for publication. References Agostini, C., Axelson, I., Colomb, V., Goulet, O., Koletzko, B., Michaelsen, K.F., Puntis, J.W.L., Rigo, J., Shamir, R., Szajewska, H. & Turck, D. (2005) The need for Nutrition Support Teams in Pediatric Units: a Commentary by the ESPGHAN Committee on Nutrition. J. Pediatr. Gastroenter- ol. Nutr. 41, 811. Brotherton, A., Simmonds, N. & Stroud, M. (2010) Meeting Quality Standards in Nutritional Care: A Toolkit for Commis- sioners and Providers in England. Worcestershire: BAPEN. Gerasimidis, K., Keane, O., Macleod, I., Flynn, D.M. & Wright, C.M. (2010) A four-stage evaluation of the Paediatric York- hill Malnutrition Score in a tertiary paediatric hospital and a district general hospital. Br. J. Nutr. 104, 751756. Hulst, J.M., Zwart, H., Hop, W.C. & Joosten, K.F. (2010) Dutch national survey to test the STRONGkids nutritional risk screening tool in hospitalized children. Clin Nutr. 29, 106 111. Joosten, K.F., Zwart, H., Hop, W.C. & Hulst, J.M. (2010) National malnutrition screening days in hospitalised chil- dren in The Netherlands. Arch. Dis. Child. 95, 141145. McCarthy, H.M.H., Dixon, M. & Eaton-Evans, M.J. (2008) Nutrition screening in children - the validation of a new tool. J. Hum. Nutr. Diet. 21, 395396. Moon, R.J., Wilson, P., Kirkham, F.J. & Davies, J.H. (2009) Growth monitoring following traumatic brain injury. Arch. Dis. Child. 94, 699701. Pawellek, I., Dokoupil, K. & Koletzko, B. (2008) Prevalence of malnutrition in paediatric hospital patients. Clin Nutr. 27, 7276. Sandhu, B.K., Fell, J.M.E., Beattie, R.M., Mitton, S.G., Wilson, D.C. & Jenkins, H. on behalf of the IBD Working Group of the British Society of Paediatric Gastroenterology, Hepatol- ogy, and Nutrition. (2010) Guidelines for the management of inammatory bowel disease in children in the United Kingdom. J. Pediatr. Gastroenterol. Nutr. 50, S1S13. Sermet-Gaudelus, I., Poisson-Salomon, A.S., Colomb, V., Brus- set, M.C., Mosser, F., Berrier, F. & Ricour, C. (2000) Simple pediatric nutritional risk score to identify children at risk of malnutrition. Am. J. Clin. Nutr. 72, 6470. Silverberg, M.S., Satsangi, J., Ahmad, T., Arnott, I.D., Bern- stein, C.N., Brant, S.R., Caprilli, R., Colombel, J.F., Ga- sche, C., Geboes, K., Jewell, D.P., Karban, A., Loftus, E.V. Jr, Pena, A.S., Riddell, R.H., Sachar, D.B., Schreiber, S., Steinhart, A.H., Targan, S.R., Vermeire, S. & Warren, B.F. (2005) Toward an integrated clinical, molecular and sero- logical classication of inammatory bowel disease: report of a Working Party of the 2005 Montreal World Congress of Gastroenterology. Can J Gastroenterol. 19(Suppl. A), 536. Sullivan, P.B. (2010) Malnutrition in hospitalised children. Arch. Dis. Child. 95, 489490. Wiskin, A.E., Wootton, S.A., Hunt, T.M., Cornelius, V.R., Afzal, N.A., Jackson, A.A. & Beattie, R.M. (2011) Body composition in childhood inammatory bowel disease. Clin Nutr. 30, 112115. Paediatric nutrition risk scores in clinical practice A. E. Wiskin et al. 2012 The Authors 4 Journal of Human Nutrition and Dietetics 2012 The British Dietetic Association Ltd.