Effect of the 23-valent pneumococcal polysaccharide vaccine on the incidence of hospitalization with pneumonia in adults aged ≥65 years: retrospective cohort study using a population-based database in Japan

Objectives The effectiveness of the 23-valent pneumococcal polysaccharide vaccine (PPV23) in preventing pneumonia in older adults remains controversial. Some studies have suggested differences in their effectiveness according to age or sex. Methods We conducted an observational study using a database of vaccine subsidization data and health insurance claims for a city in Japan. Participants were residents from 2014 to 2018 turning 65, 70, 75, 80, 85, 90, or 95 years during a given fiscal year, and PPV23 during the first year of observation were identified. We matched vaccinated and non-vaccinated individuals of the same age using propensity scores for vaccination. The incidence of hospitalization with pneumonia was compared using the Fine-Gray regression model. We summarized the results for each age using random-effects meta-analysis and conducted a subgroup analysis by sex. Results A total of 102 136 participants were included, of whom 35% received PPV23. Propensity score matching selected 32 510 pairs of vaccinated and non-vaccinated individuals. Overall, PPV23 administration was associated with a decreased incidence of hospitalization with pneumonia (17.2 vs. 20.4 per 1000 person-years, sub-distribution hazard ratio: 0.84, 95% CI: 0.77 to 0.91). Vaccine effectiveness was the highest among those aged 70 years and decreased with increasing age. No statistically significant effect was observed in those aged 90 or 95 years. Vaccine effectiveness was observed in both males and females. Conclusions PPV23 was associated with an overall decrease in hospitalization with pneumonia in older adults. However, vaccine effectiveness was significant in those aged 65 to 85 years but not in the older population.