Sex Differential Effects of Routine Immunizations and Childhood Survival in Rural Malawi

Abstract

Recent studies from West Africa suggest that routine vaccinations may have sex differential nontargeted effects. To answer whether these patterns are found elsewhere, we reanalyzed data about vaccinations and mortality in a Malawian cohort. Children were followed from birth with monthly home visits until 18 months of age. Anthropometric measurements and vaccination information were obtained at these visits. We investigated survival in relation to vaccination status using 2 different methods. First, we included all children updating vaccination information retrospectively. Second, we included only those present at monthly home visits for whom prospective vaccination information was likely to be most complete. This study included 751 children surviving the first week of life. Main outcome measures consisted of mortality rate ratios (MR) for BCG, diphtheria-tetanus-pertussis (DTP) and measles-vaccinated children compared with children who had not received these vaccines and female-male MRs among children who had received the same vaccine. In all analyses, BCG and MV were associated with lower mortality. Children with better nutritional status received DTP1 significantly earlier (P = 0.02). However, DTP1-vaccinated children did not have lower mortality, the MR being 0.99 (0.45-2.19) and 3.19 (0.80-12.8), respectively, using the 2 analyses with retrospective updating or with prospective determination of vaccination status. The female-male MR varied significantly with age, being increased at 3 to 8 months of age, the age of DTP vaccination, and reduced between 9 and 17 months of age, the age of measles vaccination (MV). Using prospective vaccination information, the MR for girls declined after BCG vaccination of unvaccinated girls but increased after DTP1 (P = 0.10) and declined after MV (P = 0.01). Children who received DTP and MV simultaneously had a MR of 5.27 (1.11-25.0) compared with children who received MV only. The female-male mortality ratio varied with age. The mode of analysis was important. Using prospective vaccination information like in previous studies from West Africa, BCG, DTP1 and MV had differential effects for girls and boys.