Incidence and predictors of attrition rate after children started inpatient treatments for complicated severe acute malnutrition in North West Ethiopia

BACKGROUND: Retaining children for inpatient treatment of complicated severe acute malnutrition (SAM) is a growing challenge until achieved the reference weight of a child. In Ethiopia, there is limited information regarding the time to be lost from the stabilizing centers after initiation of treatment. Thus, this study aimed to identify incidence and predictors of attrition for children suffering from SAM after started inpatient treatment in North West Ethiopia. METHODS: A retrospective cohort study was conducted among under-five children admitted and started inpatient treatment for complicated SAM from 2015/2016 to 2020/2021. Data were entered using Epi-data version 4.2 and then exported to STATA (SE) version R-14 software for further analysis. The analysis was computed using Cox proportional hazard regression model after checking all proportional hazard assumptions. Covariates having < 0.2 of P values in the bi-variable analysis were candidates transferred to the multivariable Cox proportional hazard regression model. Finally, a statistical significance was declared at a P value of < 0.05. RESULT: Overall, 760 files of under-five children were analyzed with a mean (± SD) age of participants 27.8 (± 16.5) months. About 6944 child-days of treatment observation were recorded with the crude incidence of attrition rate of 9.7% (95% CI 7.9–12.6). The overall median time of attrition and half-life time S(t(1/2)) of survival rates was determined as 14 (IQR = ± 7) days and 91.6% (95% CI 88.2–93.1), respectively. The attrition rate was significantly associated with cases living in rural residents (AHR = 6.03; 95% CI 2.2; 25.2), being re-admitted SAM cases (AHR = 2.99; 95% CI 1.62; 5.5), and caregivers did not have formal education (AHR = :5.6, 95% CI 2.7; 11.7) were all independent predictors for attrition from inpatient treatment. CONCLUSIONS: Nearly one in every ten severely acute malnourished under-five children defaulted at the end of treatment observation with a median time of 14 (IQR = ± 7) days. Living in a rural residence, being re-admitted cases, caregivers who did not have a formal education were significantly associated with the attrition rate. Hence, it is crucial to detect and control the identified causes of defaulting from treatment observation promptly. Furthermore, serious counseling during admission and nutritional provision strategies are essential for virtuous treatment outcomes.