Context: Since 2004, increasing numbers of military personnel have been immunized with the intranasal live attenuated influenza vaccine (LAIV) while most others received the trivalent inactivated vaccine (TIV). However, data about live virus vaccine effectiveness among healthy adults are limited.
Objective: To monitor the effectiveness of vaccines to better inform military vaccination policy.
Design, setting, and participants: Surveillance of population-based, propensity-matched, and/or vaccine-naive cohorts of more than a million active-duty, nonrecruit military service members aged 17 to 49 years stationed in the United States during the 2004-2005, 2005-2006, or 2006-2007 influenza season.
Main outcome measures: Incidence of health care encounters resulting in a primary diagnostic code consistent with pneumonia or influenza. Incident hospitalizations was a secondary outcome.
Results: In all 3 seasons, immunization with TIV was associated with lower incidence rates of health care encounters for pneumonia and influenza when compared with no immunization: 8.6 vs 19.4 for 2004-2005, 7.8 vs 10.9 for 2005-2006, and 8.0 vs. 11.7 per 1000 person-years for 2006-2007 (all P < .001). Similar estimates were obtained from propensity-matched and/or vaccine-naive cohorts. Consistently lower vaccine effect following LAIV immunization was only seen during the 2006-2007 influenza season in the total (10.7; 95% confidence interval [CI], 2.72 to 18.1; P = .03) and propensity-matched cohorts (11.8; 95% CI, 0.85 to 21.5; P = .04), and was less than effect from TIV (TIV vs LAIV, 19.8; 95% CI, 13.6 to 25.5; P < .001). Among vaccine-naive service members, however, estimates for LAIV effect were more robust for both the 2005-2006 and 2006-2007 seasons (P = .01) and were comparable with TIV (eg, LAIV, 30.2; 95% CI, 11.2 to 45.2; vs TIV, 35.3; 95% CI, 25.9 to 43.6; in 2005-2006).
Conclusions: Vaccination with TIV was associated with fewer medical encounters related to pneumonia and influenza compared with LAIV or no immunization. In this annually immunized population, this effect was less apparent in those vaccinated with LAIV.