Pneumococcal vaccination, but not influenza vaccination, is negatively associated with incident dementia among Japanese older adults: The JAGES 2013-2022 prospective cohort study.

Background: It is unclear whether inactivated influenza vaccination (IIV) or pneumococcal vaccination are associated with the risk of dementia; however, both types of vaccination are recommended for older adults. Studies have shown that the IIV is negatively associated with incident dementia; however, the uptake of pneumococcal vaccinations has not been considered. We investigated the independent associations of IIV and 23-valent pneumococcal polysaccharide vaccine (PPSV23) with incident dementia in older adults.

Methods: Health-related information on older Japanese adults was obtained through a baseline survey conducted in 2013 (baseline survey). The uptake of IIV and PPSV23 was determined in a second survey conducted in 2016 (second wave). Both surveys were conducted among independent Japanese older adults aged ≥ 65 years at the two surveys and who had not been certified as needing long-term care (LTC). In the second wave, 9,865 participants were followed up for 3.5 years (short-term follow-up), and 6,995 participants were followed up for six years and five months (long-term follow-up) until they required LTC due to dementia onset (incident dementia). A competing risk model with stabilized inverse probability weighting (SIPW) was constructed to calculate the hazard ratios (HRs) and 95 % confidence intervals (CIs) of incident dementia.

Results: PPSV23 uptake was negatively associated with incident dementia among participants in both the short- and long-term follow-up periods after SIPW (short-term follow-up: HR: 0.77, 95 % CI: 0.63 - 0.95; long-term follow-up: HR: 0.83, 95 % CI: 0.70 - 0.97). Conversely, IIV uptake was not associated with incident dementia among participants in either follow-up group (short-term follow-up: HR: 0.86, 95 % CI: 0.63-1.16; long-term follow-up: HR: 0.99, 95 % CI: 0.76-1.29). The PPSV23 uptake was negatively associated with incident dementia in participants without the IIV uptake (short-term follow-up: HR: 0.44, 95 % CI: 0.24 - 0.81; long-term follow-up: HR: 0.47, 95 % CI: 0.29 - 0.76). Conversely, the IIV uptake was not associated with incident dementia regardless of the PPSV23 status (short-term follow-up: HR: 0.87, 95 % CI: 0.62 - 1.23; long-term follow-up: HR: 1.00, 95 % CI: 0.74 - 1.35).

Conclusion: Our results suggest that the PPSV23 uptake was independently associated with the incidence of dementia. However, the IIV uptake was not associated with the incidence of dementia.