Summary A prospective cohort study evaluating the clinical effectiveness of the 23-valent pneumococcal polysaccharide vaccine was conducted among 1298 Spanish older adults with chronic respiratory diseases (bronchitis, emphysema or asthma) who were followed between 2002 and 2005. Main outcomes were all-cause community-acquired pneumonia (CAP) and 30 days mortality from CAP. The association between vaccination and the risk of each outcome was evaluated by multivariable Cox proportional-hazard models adjusted for age and comorbidity pneumococcal vaccination did not alter significantly the risk of overall CAP (hazard ratio [HR]: 0.77; 95% confidence interval [CI]: 0.56–1.07) and 30 days mortality from CAP (HR: 0.87; 95% CI: 0.33–2.28). However, a borderline significant reduction of 30% in the risk of all-cause hospitalisation for CAP was observed among vaccinated subjects (HR: 0.70; 95% CI: 0.48–1.00; p = 0.052). The effectiveness of the vaccine on the combined endpoint of pneumococcal and unknown organism infections reached 34% (HR: 0.66; 95% CI: 0.43–1.01; p = 0.059). Although our findings suggest moderate benefits from the vaccination, the evidence of clinical effectiveness appears limited.