Background: Routine immunizations have non-specific and sex-differential effects on childhood mortality and morbidity in low-income countries; BCG and measles vaccine (MV) may reduce and diphtheria-tetanus-pertussis vaccine (DTP) may increase the mortality of girls relative to boys.
Setting: Urban area in Guinea-Bissau, with a demographic surveillance system and registration of all pediatric hospitalizations. Guinea-Bissau experienced a large outbreak of measles infection in 2003-2004.
Design: We used hospital and community data to examine the impact of other vaccines on the risk of hospitalizations for measles infection. Vaccine efficacy (VE) against hospitalization for children aged 6 to 59 months of age was examined. We assessed whether VE depended on vaccination status for other vaccines and whether the pattern differed for boys and girls.
Main outcome measure: Sex-specific vaccine efficacy against hospitalization for children aged 6 to 59 months of age.
Results: The VE depended on sex and the sequence of vaccinations. The VE of MV against hospitalization for measles was better for girls than for boys. Among children who had received MV as the most recent vaccine VE against hospitalization was as high as 96% for girls, but only 81% for boys (P = 0.002). Among children who had received DTP simultaneously with MV or DTP after MV, VE declined for girls (91%) and increased for boys (90%). Compared with having received MV as most recent vaccination, DTP simultaneously with MV or DTP after MV improved the efficacy significantly for boys and the effect was significantly different for boys and girls (P = 0.023). The female-male risk ratio of hospitalization varied significantly, depending on the most recent vaccination (P = 0.014); it was 0.28 (0.11-0.68) for MV alone, but 1.21 (0.82-1.77) for DTP but no MV, and 1.13 (0.58-2.18) for DTP simultaneously with MV or after MV. Among MV-unvaccinated children, BCG-vaccinated girls had a lower risk of measles hospitalization than DTP-vaccinated girls (RR=0.0 (0.0-0.99), exact test).