ABSTRACT
In low-resource settings such as Ethiopia, where predictors of time to recovery from SAM remain understudied, identifying predictors and longitudinal weight patterns associated with time to recovery (TTR) from SAM is crucial for early and appropriate clinical interventions. Therefore, this study aimed to determine the incidence and predictors of recovery from SAM, as well as its association with longitudinal weight changes, among children aged 6–59 months in Ethiopia. An institution-based retrospective follow-up study was conducted among SAM children aged 6–59 months at Amhara Region Comprehensive Specialized Referral Hospitals (CSEHs), Northwest Ethiopia. Cox PH and linear mixed effect models were jointly used to evaluate the effects of time-dependent weight trajectories on the time to recovery from the SAM via the Bayesian approach. An adjusted hazard ratio (AHR) with a 95% credible interval (CrI) was used to determine statistically significant predictors. The data were entered into the Kobo toolbox and analyzed with R software version 4.5.1. A total of 480 children aged 6–59 months were followed for 4356 person–days of observation (PDO). At the end of the follow-up, 291 patients (291/480, 60.6%) had recovered, with an incidence density of 6.69 cases per 100 PDOs (95% CI: 5.93–7.47). The median recovery time from severe acute malnutrition was 10 days (IQR: 7–16). The time-dependent current value of weight was significantly positively associated with the time to recovery (AHR = 1.15 95% CrI: [1.04, 1.26]). Additionally, being HIV seropositive (AHR = 0.46, 95% CrI: [0.22, 0.90]), having edematous malnutrition (AHR = 0.48, 95% CrI: [0.32, 0.70]), vomiting at admission (AHR = 0.62, 95% CrI: [0.44, 0.87]), anemia at admission (AHR = 0.45, 95% CrI: [0.31, 0.64]), folic acid supplementation (AHR = 0.70, 95% CrI: [0.53, 0.94]), F-100 formula milk (AHR = 1.40, 95% CrI: [1.08, 1.80]), and antibiotic use (AHR = 1.58, 95% CrI: [1.04, 2.46]) were significant predictors of the time to recovery from severe acute malnutrition. The proportion of recovery reported in this study was lower than the minimum 75% threshold set by the national and international SPHERE standards. The median recovery time of 10 days was optimal. The unobserved true current value of weight change was significantly associated with the time to recovery from severe acute malnutrition. Additionally, HIV status, anemia, edema, vomiting at admission, folic acid supplementation, F-100 formula milk, and IV antibiotic use were significant predictors of the time to recovery from SAM. F-100 supplements and IV antibiotics should be used to improve and increase the recovery rate.
Maternal &Child Nutrition, Volume 22, Issue 3, July 2026. Read More