Effectiveness of integrated nutrition interventions on childhood stunting: a quasi-experimental evaluation design.

Background: Although malnutrition particularly stunting is recognized as multi-causal, there has been limited integrated nutrition interventions to reduce its burden in children under-fives and those existing are not well evaluated. This study tested the effectiveness of provision of health and nutrition education and promotion of home gardening in child stunting.

Methods: The study used a quasi-experimental evaluation design. Two rounds of household surveys were done to assess changes in behaviors (uptake of Antenatal Care services and child feeding practices) and stunting among children under-5 years. The sample size was calculated to detect a 10% percent absolute baseline-to-end-line change in stunting. A two-stage stratified sampling process was used to sample 896 and 1736 households at each round of data collection in the intervention and control districts respectively. Mothers delivered in the past 24 months preceding the survey and all children under-5 years residing in selected households were eligible. The difference in difference (DID) analysis was used to estimate effect of the interventions. All ethical clearances were obtained from relevant authorities prior to data collection.

Results: A total of 3467 and 4145 children under 5 years were recruited at baseline and endline respectively. The proportional of stunted children decreases from 35.9 to 34.2% in intervention and from 29.3 to 26.8% in the control sites. Overall, no statistically significant stunting reduction was observed between intervention and control sites. However, a significant effect was observed in intermediate outcomes; Uptake of iron folic acid (DID: 5.2%, (95% CI: 1.7-8.7), p = 0.003), health facility delivery (DID: 6.5%, (95% CI: 1.8-11.2), p = 0.006), pre-lacteal feeding (DID: - 5.9%, (95%CI: - 9.2, - 2.5), p = 0.001), breast feeding within 1 h after birth (DID: 7.8%, (95%CI: 2.2-13.4), p = 0.006) and exclusive breast feeding in children under 6 months (DID:20.3%, (95% CI: 10.5-30.1), p = 0.001).

Conclusion: The 3 years program did not result in significant evidence of stunting reduction, but the observed effect on health and nutrition behavioural indicators are at the causal pathways to improved child nutritional outcomes in the long run. Implementation of these integrated packages over a longer duration is needed to witness significant reduction in the prevalence of stunting.