Effect of influenza vaccination of nursing home staff on mortality of residents: a cluster-randomized trial.

OBJECTIVES To evaluate the effect of staff influenza vaccination on all-cause mortality in nursing home residents. DESIGN Pair-matched cluster-randomized trial. SETTING Forty nursing homes matched for size, staff vaccination coverage during the previous season, and resident disability index. PARTICIPANTS All persons aged 60 and older residing in the nursing homes. INTERVENTION Influenza vaccine was administered to volunteer staff after a face-to-face interview. No intervention took place in control nursing homes. MEASUREMENTS The primary endpoint was total mortality rate in residents from 2 weeks before to 2 weeks after the influenza epidemic in the community. Secondary endpoints were rates of hospitalization and influenza-like illness (ILI) in residents and sick leave from work in staff. RESULTS Staff influenza vaccination rates were 69.9% in the vaccination arm versus 31.8% in the control arm. Primary unadjusted analysis did not show significantly lower mortality in residents in the vaccination arm (odds ratio=0.86, P=.08), although multivariate-adjusted analysis showed 20% lower mortality (P=.02), and a strong correlation was observed between staff vaccination coverage and all-cause mortality in residents (correlation coefficient=-0.42, P=.007). In the vaccination arm, significantly lower resident hospitalization rates were not observed, but ILI in residents was 31% lower (P=.007), and sick leave from work in staff was 42% lower (P=.03). CONCLUSION These results support influenza vaccination of staff caring for institutionalized elderly people.