No long-term excess mortality after measles infection: a community study from Senegal.

Because measles immunization has been found in all studies to reduce mortality with more than the share of deaths attributed to acute measles, the authors examined mortality after measles infection in a study in a rural area of Senegal that included 6,924 unimmunized children, whom 1,118 developed measles. Age-adjusted post-measles mortality was similar to the mortality of unvaccinated, uninfected children (mortality ratio (MR) = 1.04, 95% confidence interval (CI) 0.80-1.35). When controlling for source of infection, mortality rate was significantly different for children who contracted measles from a person outside the home (index cases vs. unvaccinated, uninfected MR = 0.27, 95% CI 0.09-0.85) and for children infected at home (secondary cases vs. unvaccinated, uninfected MR = 1.10, 95% CI 0.80-1.51). Hence, secondary cases had markedly higher long-term mortality than did index cases (MR = 4.13, 95% CI 1.26-13.58). These estimates were essentially unchanged when the effects of season, period, separation from mother, size of community, and size of compound were investigated using a multivariate Cox regression model. The authors conclude that measles infection was not associated with increased mortality after the acute phase of infection and that index cases had lower mortality than uninfected, unvaccinated children. The reduction in mortality after measles immunization can therefore not be explained by the prevention of post-measles mortality.