Incidence and etiology of community-acquired pneumonia in the elderly in a prospective population-based study

Background: We conducted a prospective population-based epidemiological study to prepare a setting for documentation of the efficacy of novel vaccines against pneumococcal (Pnc) community-acquired pneumonia (CAP) in the elderly. Specific objectives were to demonstrate setting feasibility, to construct a case definition for Pnc CAP, and to estimate its incidence. Methods: We prospectively enrolled patients with clinical and radiological findings compatible with CAP at municipal on-call clinics serving an elderly population (age >= 65 y) of approximately 29,500. Sputum, urine, nasopharyngeal swab (NPS), and blood samples were analyzed using diverse methods for the identification of Pnc (culture, PCR, antigen tests, serology) and of other pathogens. The following case definition for Pnc CAP was derived: encapsulated Pnc in blood culture or in high-quality sputum culture or at least 2 of the following: positive urine Pnc antigen; >= 2-fold increase in serum anti-PsaA or anti-CbpA antibodies; encapsulated Pnc culture or LytA PCR in either sputum or NPS. Results: We enrolled 490 clinical CAP patients during the 2-y follow-up, 53% of all clinical CAP patients in the source population; 323 were radiologically confirmed. The incidence of radiologically confirmed CAP was 5.5/1000 person-y (95% confidence interval (CI) 4.9 -6.1) and 10.5/1000 person-y when adjusted for non-captured patients. The proportion of radiologically confirmed CAP caused by Pnc was estimated at 17%; i.e. 0.95/1000 person-y (95% CI 0.7-1.2) and 1.8 when adjusted for non-captured patients. Conclusions: We developed and documented a feasible methodology for capturing endpoints in a vaccine trial for the prevention of pneumonia. CAP incidence in the elderly population remains considerable and Streptococcus pneumoniae was one of the most commonly detected causative agents.