Abstract
OBJECTIVE: We determined the effectiveness of a 23-valent-polysaccharide pneumococcal vaccine (PPV-23) and pneumococcal conjugate vaccine (PCV-7) in reducing adult pneumococcal mortality by comparing historically predicted declines in pneumococcal disease mortality with observed patterns since the introduction of PPV-23 and PCV-7, including analyses of age, gender, and racial/ethnic subgroups. METHODS: We analyzed all deaths registered on U.S. death certificates reporting any site of pneumococcal infection (e.g., meningitis, sepsis, pneumonia, bacteremia, and peritonitis) from 1968 to 2006. We used time-series dynamic linear regression on annual pneumococcal mortality rates to determine the percentage reduction in post-1983 mortality rates for a given increase in PPV-23 vaccination rates and post-2000 mortality rates for a given increase in PCV-7 vaccination rates. RESULTS: Pneumococcal mortality decreased well before the introduction of PPV-23 in 1983 and again before the introduction of PCV-7 in 2000. The level of PPV-23 vaccination was associated with a direct and significant reduction in adult mortality, especially white female adults > or = 65 years of age. In contrast, the level of PCV-7 vaccination in the population was not associated with an indirect and significant reduction in pneumococcal mortality beyond the historical pace of decline. CONCLUSIONS: PPV-23 introduction was associated with a reduction in pneumococcal mortality among older adults > or = 65 years of age beyond levels predicted by secular trends, whereas PCV-7 introduction was not. Mortality reduction was not uniformly experienced across the population, revealing the need for additional strategies to reduce pneumococcal mortality in older adults.