Session: Poster Session: The World of Community-Acquired Pneumonia
Saturday, October 25, 2008: 12:00 AM
Room: Hall C
Background: Invasive pneumococcal disease (IPD) is an important cause of vaccine-preventable mortality. We evaluated prognostic factors and changes in mortality of patients with IPD over a 30-year period. Methods: Nation-wide population-based cohort study based on the linkage of IPD-laboratory surveillance data between 1977 and 2007 to several Danish national data sources. Patients were followed for mortality within 30-days after hospitalization. Mortality in three decades was compared using χ2-test and examined in a multivariate logistic regression analysis, controlling for confounders. Results: 18.858 patients were included in the study; 90% were ≥15 years. Of the 16.096 bacteremia patients, 71% had a pneumonia diagnosis code. The overall mortality rate was 20%, 19% and 17% in the three decades (χ2-test p<0.02). Mortality among bacteremia patients was 19%, 18%, and 17% (p=0.09), and 21%, 23% and 21% for meningitis patients (p=0.4). For pneumonia patients without meningitis, mortality was stable at 14% in the three decades. By multivariate analysis, increasing age, male sex, pneumococcal serotype, comorbidity, and earlier decade of diagnosis were independent risk factors for death. There was a higher 30-day mortality among bacteremia patients in the first (OR 1.40 [95% CI: 1.20-1.64]) and second decades (OR 1.21 [95% CI 1.10-1.33]). We found a similar association with higher mortality during the first decade among meningitis patients (OR 1.39 [95% CI 1.07-1.81]), and a marginally significant mortality difference between the second and the third decade (OR 1.21 [95% CI 0.97-1.50]). Conclusions: Mortality rates associated with IPD remain high in Denmark. Only after adjusting for individual patient factors, mortality after bacteremia and meningitis was shown to have declined over the last 30 years.