Influenza remains a significant cause of death in the elderly, particularly in frail residents of long-term hospitals and residential care facilities. Direct immunization of frail, elderly patients reduces the risk of respiratory infection and death, but immunologic protection is frequently incomplete because of the impaired immune response in many of these patients. Studies indicate that about one quarter of health care workers in long-term care facilities had serologic evidence of influenza during one winter season. Therefore, elderly patients are at considerable risk of direct and repeated exposure to infection. Carman and colleagues studied the effect of influenza immunization of health care workers on mortality rates in elderly patients in long-term care hospitals.

A parallel-group study was conducted in 20 long-term care hospitals. Nurses, physicians, therapists, housekeeping staff and other ancillary staff were included in the study. The hospitals were paired, based on size and other characteristics, with one hospital of each pair randomly assigned to participate in the intervention. The other hospital served as the control. All health care workers at the intervention hospitals were interviewed and offered influenza immunization. To establish the levels of influenza immunization, staff at both sets of hospitals were sent a questionnaire after the study concluded. During the study, patients were monitored for clinical influenza symptoms, and nose and throat swab samples were obtained from some patients in all hospitals to monitor levels of influenza activity.

Overall, 1,217 health care workers were offered immunization against influenza; 620 (50.9 percent) were vaccinated. Questionnaires from the same sites showed an immunization rate of 49.8 percent in intervention hospitals compared with a rate of 4.8 percent in control hospitals. The rate of return of questionnaires was 68 percent from intervention hospitals and 49 percent from control hospitals. In the patient group, 749 patients from the intervention hospitals and 688 patients from the control hospitals were included in the study. Age and sex were similar for both groups. The mean age for both groups was 82 years, and the proportion of men was approximately 30 percent. The uncorrected mortality rate in intervention hospitals was 13.6 percent compared with 22.4 percent in control hospitals. The surveillance program found no significant difference in the rates of influenza infection between the intervention hospitals and the control hospitals (5.4 versus 6.7 percent, respectively).

The authors conclude that immunization of health care workers was associated with a significant decrease in the mortality rates of patients in long-term care hospitals, probably through prevention of nosocomial transmission. The immunization of the health care workers at these facilities could have raised awareness of the need to immunize patients because rates of patient immunization were significantly higher in intervention hospitals (48 percent) than in control hospitals (33 percent). Although the surveillance program showed little difference in total rates of influenza infection, immunization could have protected patients from the virulant Wuhan H3N2 variant. This variant was encountered during the study period, but a good match between the prevailing influenza virus and the vaccine was likely to be an important factor in the protective effect of the vaccine. Surveillance of patients who died during the study revealed that no influenza virus was found (by polymerase chain reaction and culture) in 17 deceased patients from the intervention hospitals, but influenza virus was detected in 10 of 30 patients who died in control hospitals.