BACKGROUND: Streptococcus pneumoniae is the primary causative agent of pneumonia in older adults. Vaccination is the only tool to protect against pneumococcal infection; however, vaccination rates remain far below the Healthy People 2010 objective of 90% coverage. The number one reason for such low rates is attributed to controversy over the protective efficacy of the vaccine in preventing nonbacteremic pneumonia (eg, community-acquired pneumonia [CAP]). OBJECTIVES: The primary objectives of this study were to assess the incidence of pneumonia, pneumonia requiring hospitalization, and pneumonia hospitalization costs. METHODS: In this retrospective cohort study of Medicare beneficiaries aged 65 years in 2003, subjects were selected based on exposure status. Exposure was defined as receipt of the 23-valent pneumococcal polysaccharide vaccine (PPV23). Vaccinated persons were then matched 1:1 on gender and the presence of any comorbidity to unvaccinated persons. Subjects were followed up for 1 year (January 1, 2004 through December 31, 2004). The primary outcomes were pneumonia, pneumonia requiring hospitalization, and hospitalization costs. Mantel-Haenszel chi-square or logit was used to estimate the relative risk (RR) associated with vaccination and each outcome and Proc Ttest was used to test the difference between mean hospital costs of the vaccinated and non-vaccinated. RESULTS: During the follow-up period, 443 patients were diagnosed with pneumonia; 266 had previously been vaccinated and 177 had no documented receipt of prior vaccination. Results of the Chi-square analysis revealed a significant association between vaccination and the risk of pneumonia, as the vaccinated were 50% more likely to develop pneumonia than were the non-vaccinated (Adjusted RR: 1.50; 95% CI: 1.25, 1.81). Approximately 67% of patients diagnosed with pneumonia required hospitalization; of which, 183 were previously vaccinated and 115 had no documented receipt of prior vaccination. There was no association between vaccination and risk of pneumonia requiring hospitalization (P value 0.4001). However, the vaccine was associated with a significant reduction in hospital costs (P value 0.004). CONCLUSIONS: The results of this study suggest that use of the vaccine may be associated with cost savings due to a reduction in hospitalization.