Breastfeeding and complementary feeding practices in Madagascar: results of a secondary analysis of the freeBily trial.

Breastfeeding and complementary feeding are key public health practices for child development and attainment of optimal nutritional status, with additional beneficial effects on maternal health. In Madagascar, we conducted a secondary analysis of a cluster randomised clinical trial data (Pan-African Clinical Trial Register PACTR201905784271304) to assess the practice of delayed complementary feeding (DCF) at nine months postpartum (T3 visit) and continued breastfeeding (CB) at 24 months (T4). Factors associated with the prevalence of DCF were investigated using a generalized estimating equation (GEE) for binomial outcome with logarithmic link function to account for clustering, prevalence ratio (PR) with 95% confidence intervals (CIs) were estimated. A Cox proportional hazards model was used to assess factors associated with breastfeeding termination (BT). A total of 1863 women were evaluated at T3, 2332 at T4. At T3, 1848 (98.8%) were breastfeeding and 479 (25.7%) were delaying complementary feeding. At T4, 1269 (54.4%) were breastfeeding. Median time for BT was 21 months (range 2-25). Secondary education (PR = 0.67, CI 0.49-0.93) and anaemia (PR = 0.75, CI 0.65-0.88) were associated with lower prevalence of DCF at T3, being a farmer with a higher prevalence (PR = 3.45, CI 1.90-6.28). Older age was associated with longer CB (hazard ratio 0.68, CI 0.49-0.96 for 30-50 years). The association between being farmer mothers and DCF at nine months requires further investigations. Two main lines of public health actions are identified: health education, with intergenerational mentoring initiatives from older to younger women, and integrating maternal and child health programmes to improve both maternal anaemia and child development in a life-cycle approach.