Effectiveness of Pneumococcal Vaccination in Older Adults in a Real-World Setting: A Systematic Review and Meta-Analysis of Observational Studies Using Test-Negative Designs

OBJECTIVES: Pneumococcal disease continues to be a major worldwide cause of vaccine-preventable morbidity and mortality, in particular among older adults. The current study aims to understand the vaccine effectiveness (VE) of pneumococcal vaccination in older adults in real-world settings, which is important to inform vaccination policies and the ongoing development of vaccines.

METHODS: We searched PubMed for English observational studies from 2010 using test-negative designs(TND) to estimate VE of pneumococcal vaccination against pneumococcal disease (e.g., hospitalized community-acquired pneumonia) in adults aged ≥65 to produce pooled odds ratios (ORs), stratified by vaccine type (PCV13; PPSV23). Where data availability allowed, results were further stratified into four subgroups: (1) unadjusted ORs for all positive tests (i.e., irrespective of whether the infection matched the vaccine serotype), (2) adjusted ORs for all positive tests (3) unadjusted ORs for only vaccine-serotype infections and (4) adjusted ORs for vaccine-serotype infections. Fixed-effects models were applied for all analyses given low heterogeneity.

RESULTS: We identified seven eligible TND studies (five of which estimated VE of PCV13 and four estimated VE of PPSV23). For PPSV23 (total N= 4193), the adjusted OR for the vaccine-serotype infections (group 4) was 0.80 (95% Confidence Interval [CI]: 0.66-0.96) with similar ORs observed across outcome groups 1 to 3. For PCV13 (total N= 3590), the group 4 OR was 0.40 (95% CI: 0.17-0.95). Though ORs were lower in group 4 than in groups 1 and 3 (indicating the potential for higher effectiveness in preventing vaccine-type infections), CIs were wide and no meaningful statistical comparisons could be made (only one study available for PCV13 group 2).

CONCLUSIONS: The evidence supports both PCV13 and PPSV23 in the prevention of pneumococcal disease in real-world settings. Sample size did not allow for additional comparisons or meta-regression, and further research is warranted on the relative effectiveness of vaccine types, overall and for serotype-specific infections.