Determinants affecting utilisation of health services and treatment for children under-5 in rural Nepali health centres: a retrospective cohort study

Abstract

Background: Large inequalities in child health remain in Nepal, with caste, ethnicity and sex being major determinants of deprivation and negative outcomes. The purpose of this study was to explore whether key demographics on under 5s were associated with health seeking behaviours, utilisation of health care, and treatment received.Methods: Data came from Integrated Management of Neonatal & Childhood Illness (IMNCI) records of 23 health centres across five districts. After digitising the paper records, district, ethnicity, sex, age and temperature of the child were used to predict the number of days taken to seek medical care for Acute Respiratory Infection (ARI), diarrhoea and fever. In addition to this, correct diagnosis and subsequent treatment of pneumonia was assessed against IMNCI guidelines, again using the demographic factors of interest to predict these outcomes.Results: From 116 register books spanning 23 health centres, 33,860 child patient records were considered for analysis. The median age of attendance was 16 months (Inter-Quartile Range= 9, 30), while there were more male children that attended (55.8% vs. 44.2% for females). There were statistically significant differences for the time taken to attend a health centre between different districts for ARI, diarrhoea and fever, with children in the remote Humla and Mugu districts taking significantly longer to present at a health facility after the onset of symptoms (all p<0.012). Children from underprivileged ethnic groups, Madhesi and Dalit, were less likely to be given a correct diagnosis of pneumonia (p=0.001), while males were more likely to receive a correct diagnosis than females (73% vs. 67%, p=0.001). This sex difference remained in the adjusted regression models for diagnosis of pneumonia (p=0.011) but not for treatment of pneumonia (p=0.202).Conclusions: Significant demographic differences were found based on ethnicity, sex, and district when examining health seeking behaviours for ARI, diarrhoea, and fever. Significant associations were seen for these same factors when exploring accuracy of diagnoses of pneumonia, but not for treatment. This study has emphasised the importance of a digitalised healthcare system, where inequalities can be identified without the reliance on anecdotal evidence.