Community Health Workers Match Clinic Outcomes for Severe Child Malnutrition
A quasi-experimental study in Somalia finds community-based SAM treatment achieves 89% recovery with 7 fewer days than clinic care.
Summary
A study in Jowhar, Somalia compared two approaches to treating severe acute malnutrition in young children: facility-based outpatient therapeutic programs versus a community health worker model called iCCM+. Over 1,100 children participated between January and May 2024. Recovery rates were nearly identical — 89% in the community group versus 91% in clinics. Remarkably, children treated in the community recovered about 7 days faster and were three times more likely to reach recovery at any given point during treatment. Researchers suggest the community model enables earlier identification of malnourished children before their condition worsens. The findings support expanding community-based nutrition care in low-resource settings where clinic access is limited, potentially reducing the burden on both health systems and caregiving families.
Detailed Summary
Severe acute malnutrition remains a major cause of child mortality in humanitarian settings, yet the evidence base for shifting care from clinics to community health workers has been limited. This study directly addresses that gap with real-world data from Somalia, one of the world's most challenging healthcare environments.
Researchers enrolled 1,183 children with uncomplicated severe acute malnutrition across two facility-based outpatient therapeutic program sites and eight villages using the integrated community case management plus model. From January to May 2024, children in both arms received weekly assessments, ready-to-use therapeutic food, and standard medications until they met discharge criteria based on arm circumference and weight-for-height measurements.
The headline result: recovery proportions were statistically similar — 89% in the community group versus 91% in facility care. More striking was the speed of recovery. Community-treated children had a 3.32-fold higher hazard of recovery at any point during treatment and a median length of stay seven days shorter. Statistical modeling accounted for baseline differences using propensity score weighting, as clinic children presented with more severe deficits at enrollment — a finding that itself suggests the community model captures cases earlier.
The implications reach beyond Somalia. In settings where clinic access is constrained by distance, insecurity, or caregiver burden, a community health worker model that achieves equivalent outcomes while shortening treatment duration represents a meaningful efficiency gain. Earlier case identification may prevent deterioration to complicated SAM, which requires costly inpatient care.
Important caveats apply. This was a quasi-experimental rather than randomized design, making confounding harder to fully eliminate. The study was conducted over a single four-month period in one district, limiting generalizability. The summary is based on the abstract only, as the full text was not accessible for review.
Key Findings
- Community health worker model achieved 89% SAM recovery, matching the 91% rate of facility-based care.
- Children in the community arm recovered 7 days faster with a 3.32x higher hazard of recovery.
- Community model identified children earlier, with less severe malnutrition at enrollment.
- iCCM+ reduced caregiver and health system burden without compromising treatment outcomes.
- Findings support WHO guidelines endorsing community health workers for uncomplicated SAM treatment.
Methodology
Quasi-experimental design enrolling 1,183 children across two OTP sites and eight iCCM+ villages in Jowhar, Somalia from January to May 2024. Outcomes analyzed using generalized linear mixed-effects models and Cox proportional hazards models. Inverse probability weighting based on propensity scores was applied to adjust for baseline differences between groups.
Study Limitations
The quasi-experimental design cannot fully eliminate confounding despite propensity score adjustment, and baseline severity differences between groups suggest possible selection bias. The study was conducted over only four months in a single district of Somalia, limiting broader generalizability. This summary is based on the abstract only, as the full text was not available for review.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.