ASSOCIATION BETWEEN NUTRITIONAL STATUS OF A CHILD USING SCREENING TOOL FOR ASSESSMENT OF MALNUTRITION IN PEDIATRICS AND CLINICAL OUTCOMES IN INPATIENTS AT A TERTIARY CARE HOSPITAL SETTING IN EASTERN NEPAL

1* 2 3 4 Sonia Dahal , Hemsagar Rimal , Hiralal Sahani , Kumud Pyakurel Received : 13 October, 2021 Accepted : 02 January, 2022 Published : 21 February, 2022 Dahal S et al ISSN: 2542-2758 (Print) 2542-2804 (Online) 1631 Birat Journal of Health Sciences Vol.6/No.3/Issue 16/Sep.-Dec., 2021 Original Research Ar cle 1632 ISSN: 2542-2758 (Print) 2542-2804 (Online) Birat Journal of Health Sciences Vol.6/No.3/Issue 16/Sep.-Dec., 202


Introduc on
Malnutri on in children has health and physical consequences resul ng in delayed physical growth and motor development, behavioral and social problems as well as suscep bility to contrac ng diseases. The Screening Tool for Assessment of Malnutri on in Pediatrics (STAMP) is a specially constructed tool that uses anthropometric measurements, body composi on, and pa ent's clinical status to predict the risk of malnutri on for inpa ents at admission. There is no agreement on the ideal screening tool or scales for detec ng and determining the level of malnutri on in pediatric inpa ents. Various studies in the pediatric popula on have analyzed the associa on of malnutri on during hospitaliza on with pa ent outcomes, but few have analyzed whether the STAMP tool in par cular could be used to predict length of hospital stay and disease severity in these pa ents.

Objec ves
We studied the use of STAMP screening tools to assess the risk of malnutri on in inpa ents by including a nutri onal screening in the admission process and analyze the associa on between the risk of malnutri on and pa ent outcomes (length of hospital stay and severity of disease).

Methodology
We conducted a prospec ve single-center study between May to July 2021 at a ter ary care hospital. All pa ents admi ed to the pediatric ward aged 1 month to 15 years with a length of stay of at least 24 hours were included into the study. Data were collected about clinical, anthropometric and demographic variables for each pa ent including STAMP grading, length of hospital stay and severity of disease.

Results
Our study suggests comparable demographic and other pa ent-related variables in the three categories (mild, moderate and severe) of STAMP grading (p-value >0.05). STAMP grading was not significantly associated with both length of hospital stay and severity of disease.

Conclusion
The use of STAMP as a malnutri on screening tool for pediatric inpa ents is not significantly associated with length of stay or severity of disease.

KEYWORDS
Length of hospital stay, Malnutri on, Pediatrics, Severity of disease, STAMP tool

INTRODUCTION
Malnutri on of children has always been a global concern. Health and physical consequences of malnourishment in children are delay in their physical growth and motor development; lower intellectual quo ent (IQ), greater behavioral problems and deficient social skills; suscep bility 1,2 to contrac ng diseases and the list goes on. The 2017 data from the Ministry of Health and Popula on of Nepal suggests 1 million children under 5 years (36 %) suffer from chronic malnutri on (stun ng or low height-for-age) and 10 % suffer from acute malnutri on (was ng or low weight-for- 3 height). Along with malnutri on the burden of infec ous diseases (pneumonia, diarrhea) resul ng in hospital admissions is more. Children may present with a degree of malnutri on at the me of hospital admission. On top of it, pediatric malnutri on, before and during hospital admission could elevate morbidity and influence the clinical outcomes of disease. When admi ed in hospital, they are generally not subjected to anthropometric assessment therefore may 4,5 not receive nutri onal support. Measuring a child's growth is one of the most efficient ways of assessing their general state of health, making effec ve interven ons possible that can reestablish ideal condi ons for health and avoid the 6 damage resul ng from malnutri on. Nutri onal assessment of hospitalized children is very o en neglected, contribu ng to the occurrence of complica ons and 7 prolonged hospital stays. There are different tools to assess the risk of malnutri on in hospitalized children. The World Health Organiza on (WHO) classifica on for malnutri on or the Waterlow classifica on is widely used in our se ng as a screening tool for classifica on of malnutri on in pediatrics. The Screening Tool for the Assessment of Malnutri on in Pediatrics 7 (STAMP) developed by McCarthy et al is a specially designed tool to determine nutri onal risk for inpa ents at admission. STAMP is a nutri onal screening tool that uses anthropometric measurements, body composi on and the clinical condi on of the pa ent. It has a high sensi vity 8 (90%) but a low specificity. There is no consensus on the op mal screening method or the scales that should be used to detect and establish the level of malnutri on in pediatric [9][10][11] inpa ents but previous studies have recommended the use of these tools even in outpa ents at the primary care 12 level. Various studies in the pediatric popula on have analyzed the associa on of malnutri on during hospitaliza on with pa ent outcomes, but only few have analyzed whether the STAMP in par cular could be used to predict outcomes and severity in these pa ents. The main objec ve of our study was to use the STAMP screening tool to assess risk of malnutri on in inpa ents by including a nutri onal screening in the admission process to analyze the associa on between the degree of malnutri on and pa ent outcomes (length of hospital stay and disease severity) during the hospital stay.

METHODOLOGY
This prospec ve single-center study was conducted at Birat Medical College Teaching Hospital between May to July 2021. All consecu ve pa ents admi ed to the pediatric ward aged 1 month to 15 years with a length of stay of at least 24 hours were included into the study a er consent of their guardian. Pa ents in other pediatric special es (cardiology, nephrology, oncology and hematology), pa ents aged < 1 month and postopera ve pa ents were excluded from the study. Each pa ent's clinical and demographic data, including anthropometric measurements of weight (in kg), height or length (in cm), and body mass index (BMI), were recorded on the day of admission. Using calibrated SECA GmbH375® baby scales, pa ents were weighed naked or lightly clothed in the decubitus posi on for children under 2 years and standing for older children. A stadiometer was used to measure length of the pa ent as the distance between the top of head and heel of pa ents under the age of 2 years. Height of children aged more than 2 years was measured with a pa ent standing with a column scale. We documented the weight for height (WFH), height for age (HFA) and BMI percen le based on the growth tables by WHO. Disease was considered severe if a pa ent required admission to the pediatric intensive care unit (PICU), Wood-Downes clinical scoring system (severity ≥ 8) in pa ents with bronchioli s or pulmonary score in pa ents with asthma exacerba ons (severity ≥7). The risk of malnutri on was assessed using the STAMP screening tool for malnutri on (h p://www.stampscreeningtool.org) Collected data were analyzed with SPSS version 22 (IBM SPSS Sta s cs; Armonk, NY, USA). The Kruskal-Wallis test was used to analyze quan ta ve variables, while the Chi square or Fisher exact test was used to analyze categorical variables. The data is expressed as a median and interquar le range (IQR) or as a frequency.

RESULTS
This study was carried out at Birat Medical College Teaching Hospital a er approval from the ins tu onal review commi ee. A total of 153 pa ents fulfilling the inclusion criteria within the study period were enrolled for the study where 127 pa ents were enlisted for the final analysis. In total, 27 pa ents had severe disease with median length of stay for the study popula on was 6 days with IQR ranging from 4-7 days. The results of our study suggest comparable demographic and other pa ent variable across the three categories (mild, moderate and severe) of STAMP grading (p-value >0.05), thus usage of STAMP grading was not significantly different across the data (Table 1 and 2). Table 3 and 4 suggest STAMP grading and other nutri onal parameters (BMI percen le, HFA and WFH) were not associated with both length of hospital stay and severity of disease. Thus, our study suggests the risk of malnutri on given by STAMP screening tool do not associate significantly to both length of hospital stay and disease severity.

DISCUSSION
Malnutri on is associated with increased length of hospital stay, morbidity and mortality, infec on risk, and increased hospital costs when compared with well-nourished [13][14][15][16][17] children. STAMP grading is a tool intended to be used to assess the risk of malnutri on for hospitalized children. The advantage of this tool is one-me assessment of pa ents generally at the me of hospital admission. This is significantly important as STAMP could also be used for children during the outpa ent visit as well. Thus, this study was relevant to assess whether risk of malnutri on given by STAMP would be able to be related to length of hospital stay and severity of disease thereby highligh ng its importance clinically as well. The results of our study however have demonstrated that risk of malnutri on given by STAMP cannot be related to both length of hospital stay (≥5 days) or severity of disease. The median length of hospital stay of our study popula on was 6 days. The result of our study is in line with a study by Pérez-SolísD where length of hospital stay of 3 days dura on 18 was not associated with the STAMP grading. Our results differ from a previous study with similar methodology to ours where length of hospital stay of 3 or more days has been 19 reported to be associated with STAMP grading. It is to be noted that median length of hospital stay for the above study was 3 days while in our findings was 6 days. Reanalysis of our data revealed a significant associa on between length of hospital stay of 3 days or more with STAMP grading. The results of our study could be explained by two reasons. Firstly, STAMP is a screening tool with generally high sensi vity and low specificity to predict malnutri on as given by various studies. This observa on suggests that the STAMP tool is likely to over diagnose nutri onal risk in hospitalized children, thus could have led to our nonsignificant results. This was highlighted in a recent study by 20 Lesego N et al. The authors in this study observed STAMP grading to be non-significantly related with length of hospital stay. In the same study, when they used a modified STAMP tool that increased the predic on of malnutri on, they found significant correla on with length of hospital stay. Secondly, length of hospital stay has a mul factorial causa on and a causal rela onship between different 21 variables has not been demonstrated.
Our study suggests no clinical significance between the associa on of disease severity and STAMP grading. This is in contrast to a previous study with similar methodology to ours where significant associa on between STAMP grading 19 and severity of disease was noted. A closer inspec on to our data revealed 27 pa ents had severe disease and 5 pa ents had STAMP more than or equal to 4. Our study reveals 19.5% and 21% of pa ents respec vely had mild and moderate STAMP grading whereas the above study had 1.3% and 7%. We believe the discrepancy between the data could be related to hospital admission criteria and difference between the endemic and seasonal diseases, thus influencing the results.

LIMITATION OF THE STUDY
There are some limita ons in the study. Firstly, though our hospital is a ter ary referral center with a diverse set of pa ents seeking medical assistance all year round, limited number of par cipants involved in the study over 3 months of study period limits randomiza on of the overall popula on. Secondly, cross checking with standard diete c evalua on for malnutri on would have enlightened the predic on poten al of STAMP in our study popula on, thus confirming our findings.

CONCLUSION
The use of STAMP screening tool to assess the risk of malnutri on in inpa ents by including a nutri onal screening in the admission process does not associate significantly with length of hospital stay or severity of disease.

RECOMMENDATION
Further studies with overall longer dura on for sample collec on, higher number of par cipants and correla on of finding with standard diete c evalua on for malnutri on is recommended to negate the limita ons of the present study.