Comparative effectiveness of high-dose versus standard-dose influenza vaccines in nursing home residents aged 65 and older in France: a nationwide cohort study on the French health data system from the 2022-2023 season.

BACKGROUND

Older adults in nursing homes face high influenza-related morbidity and mortality due to immunosenescence and comorbidities, which may reduce vaccine effectiveness. High-dose (HD) influenza vaccine, containing four times more antigen than standard-dose (SD) formulations, aim to compensate for this impaired immune response.

OBJECTIVES

To compare the Relative Vaccine Effectiveness (rVE) of HD versus SD influenza vaccines against influenza-related hospitalisations (primary), and hospitalisations due to pneumonia or combined influenza/pneumonia (secondary) in nursing home residents aged ≥65 years.

STUDY

DESIGN

Nationwide retrospective cohort (French SNDS). Residents vaccinated between 1 September 2022, and 31 March 2023, were followed until 30 June 2023, discharge or death. Facilities with in-house pharmacies were excluded.

EXPOSURES/OUTCOMES: HD vs. SD influenza vaccine (pharmacy dispensing codes). Hospitalisations identified by ICD-10 discharge codes.

METHODS

HD and SD recipients were matched 1:1 using propensity scores with exact matching on sex, age group, region and vaccination week. Incidence rate ratios (IRR) were estimated using Poisson/negative binomial models. rVE = (1-IRR) × 100.

RESULTS

Among 315 239 residents (123 720 HD; 191 519 SD), 76 858 pairs were matched (mean age 88 years; 75% women). Influenza hospitalisation rates were 302 vs. 438 per 100 000 person-years, corresponding to an rVE of 30.8% (95% CI: 11.8-45.7; P = .0029). No significant rVE was observed for pneumonia or combined outcomes.

CONCLUSIONS AND RELEVANCE: HD vaccine was associated with a significant reduction in influenza-related hospitalisations in French nursing home residents, supporting preferential HD use in this high-risk population. Residual confounding cannot be excluded, and vaccine safety was not assessed in the present analysis.