Risk factors for inpatient mortality among children with severe acute malnutrition in Zimbabwe and Zambia

Abstract

Background/objectives: Malnutrition underlies 45% of deaths in children under-5 years annually. Children hospitalised with complicated severe acute malnutrition (SAM) have unacceptably high mortality. We aimed to identify variables from early hospital admission (baseline factors) independently associated with inpatient mortality in this cohort to identify those most at risk.

Subjects/methods: Observational study of 745 children aged 0-59 months admitted with complicated SAM at three hospitals in Zimbabwe/Zambia. Children underwent anthropometry and clinical assessment by a study physician within 72 h of enrolment, and caregivers provided sociodemographic data. Children were followed-up daily until discharge/death. A multivariable survival analysis identified the baseline factors independently associated with mortality.

Results: 70/745 (9.4%) children died in hospital. Age between 6-23 months [aHR 6.53, 95%CI 2.24-19.02], higher mid-upper arm circumference [aHR 0.73, 95%CI 0.59-0.89], presence of oedema [aHR 2.22, 95%CI 1.23-4.05], shock [aHR 8.18, 95%CI 3.79-17.65], sepsis [aHR 3.13, 95%CI 1.44-6.80], persistent diarrhoea [aHR 2.27, 95%CI 1.18-4.37], lack of a toilet at home [aHR 4.35, 95%CI 1.65-11.47], and recruitment at one Harare site [aHR 0.38, 95%CI 0.18-0.83] were all independently associated with inpatient mortality. Oedematous children had a significantly higher birthweight [2987 g vs 2757 g, p < 0.001] than those without oedema; higher birthweight was weakly associated with mortality [aHR 1.50 95%CI 0.97-2.31].

Conclusions: Children with oedema, low MUAC, baseline infections, shock and lack of home sanitation had a significantly increased risk of inpatient mortality following hospitalisation for complicated SAM. Children with high-risk features may require additional care. A better understanding of the pathophysiology of SAM is needed to identify adjunctive interventions.