OBJECTIVES: The 23-valent pneumococcal polysaccharide vaccine (PPV) is recommended for prevention of pneumococcal diseases in adults at risk. Few data exist on time-, age- and sex-dependent vaccine effectiveness on outcomes in pneumonia.

METHODS: We performed a population-based cohort study including adults ≥60 years of age (n = 738 927) based on statutory health insurance data from 2005 to 2011. Primary outcomes were all-cause pneumonia incidence and 30-day all-cause mortality. Pneumonia was identified by International Classification of Diseases, 10th Revision, German Modification (ICD-10-GM) codes, with ambulatory cases validated by antibiotic prescription within 7 days. The effect of PPV within 5 years was analysed after propensity score-based matching (three controls per case with PPV vaccination) including comorbidities, care status, age, sex and influenza vaccination. Evaluations were stratified by age group, sex and time of PPV.

RESULTS: Two-year incidence of all-cause pneumonia in 213 431 vaccinated individuals was 7501 (3.51%) of 213 431 vs. 23 243 (3.63%) of 640 293 in matched controls (difference -0.11, 95% confidence interval (CI) -0.22 to -0.002, p 0.046). After sex-dependent analysis, PPV effectiveness on pneumonia incidence was observed only in women (difference -0.15, 95% CI -0.28 to -0.02, p 0.02). Thirty-day mortality in vaccinated individuals with pneumonia was 1302 (17.36%) of 7501 vs. 4267 (18.96%) of 22 503 in matched controls (difference -1.60%, 95% CI -2.83 to -0.38, p 0.011). After analysis according to age group, significant mortality reduction was present only in adults aged 60 to 79 years (difference -2.31, 95% CI -3.79 to -0.83, p 0.002). Year of PPV vaccination showed no effect on outcomes.

CONCLUSIONS: The findings support consideration of sex and age dependence of PPV effectiveness in future studies.