Age-specific changes in the female-male mortality ratio related to the pattern of vaccinations: an observational study from rural Gambia.

BACKGROUND: According to studies from Guinea-Bissau and Senegal, live vaccines may reduce the female-male mortality ratio (MR) whereas inactivated vaccines increase this ratio. We used data from The Gambia to examine whether similar tendencies could be found in a different setting.
SETTING: Forty villages in the Farafenni area in rural Gambia.
SUBJECTS: A population of 17,000 was followed with demographic surveillance between 1998 and 2002; 537 children less than 5 years of age died in this period.
METHODS: We used two vaccination surveys and community mortality data to examine, first, the female-male mortality ratio (MR) in the age groups in which DTP and MV are recommended and have a high coverage. Second, using vaccination cards seen post-mortem, we examined the distribution of live or inactivated vaccines as last vaccination in different age groups. Third, we examined the effect of DTP and MV administered simultaneously. MAIN OUTCOME MEASURES: The female-male MR in different age groups and for different vaccines.
RESULTS: Vaccination coverage was high for BCG, third dose of DTP (DTP3) and MV, reaching a level of 80-90% within a few months of the recommended age of vaccination. First, the female-male MR was 0.93 (0.63-1.38) in the first 2 months of life when children had received no vaccination or the combination of BCG, HBV and OPV. From 2 to 8 months of age, with DTP and HBV being the main vaccinations, the female-male MR was 1.28 (0.86-1.89). Between 9 and 17 months of age, with MV as the main vaccination, this ratio dropped to 0.73 (0.50-1.07), a significant inversion of the female-male MR (p=0.045). Second, using information from vaccination cards of dead children, boys who died at 2-4 months of age were more likely to have received live BCG and girls to have received inactivated DTP and HBV as last vaccination (p<0.001). At 5-8 months of age, essentially all dead children had received DTP as last vaccination and the female-male MR was 1.68 (0.96-2.93), whereas the MR was 0.70 (0.43-1.15) at 12-17 months of age when nearly all dead children had received MV (p=0.022). Third, compared with the general population of children who had received MV, dead children who had received MV were more likely to have received DTP3 simultaneously with MV (relative risk (RR)=5.59 (2.10-14.8)) or after MV (RR=2.61 (1.13-6.05)).
CONCLUSION: Most children dying at a specific age had received the recommended vaccines. BCG and MV as last vaccination was associated with a low female-male MR, whereas DTP as last vaccination was associated with a high female-male MR. These trends are consistent with observations from other African countries.