Preventing Influenza in Healthy Adults

INACTIVATED INFLUENZA VACCINE WAS ORIGINALLY DEveloped during World War II to protect healthy adult members of the United States armed services. With its major interest in preventing influenza to maintain military readiness, the US military conducted a large number of randomized controlled trials from 1943 through 1969, which demonstrated that the vaccine was 70% to 90% efficacious in preventing laboratory-confirmed clinical influenza as long as the virus in the vaccine resembled the virus that was circulating. The current influenza vaccine is similar to the vaccines used in those trials, with higher potency and fewer adverse events. Its efficacy in healthy adults was recently reconfirmed in a randomized trial in which efficacy against laboratory-confirmed influenza was 88%. Recommendations for use of influenza vaccine in the United States for years have emphasized prevention of influenza in persons most likely to experience complications: those aged 65 years or older and younger individuals with cardiac, pulmonary, and other chronic conditions. Ironically, the clinical efficacy of the vaccine in older individuals continued to be questioned until the 1990s, when a series of observational studies demonstrated significant effectiveness of influenza vaccine in preventing hospitalization and death in persons aged 65 years or older. For this population, however, a new issue was introduced. Congress questioned whether the Medicare system should cover the costs of annual influenza vaccine; thus, not only effectiveness but also cost-effectiveness needed to be evaluated. These studies showed that the vaccine reduced hospitalizations and their high cost, making it more likely that vaccination would be cost-effective for the elderly population. Younger individuals experience influenza infection more frequently, but because they are much less likely to experience complications than elderly persons, they have not been targets of public health vaccination programs. However, many large employers traditionally provide vaccination to workers to reduce absenteeism, and to offer a workplace benefit. This approach has been quite successful, so much so that in the past, there was talk of the vaccine going into the “wrong arms,” the correct arms being those with risk conditions. Others believed that there were no wrong arms, and that the answer was to increase production of vaccine so that all who wished to receive it could do so. Although vaccination of healthy adults is known to be effective in preventing clinical influenza, this practice also has been scrutinized in terms of its cost-effectiveness. Results frequently have been conflicting, explained in part by varying intensity of influenza outbreaks and the limited time span of the studies. In addition, much of the benefit for healthy adults comes not from preventing expensive complications but from reducing time lost in the workplace and the home, the costs of which vary substantially by country. Thus, for the US population, the most relevant study needed to be done in the United States. The first such study, conducted by Nichol et al, combined a 1-year randomized controlled trial of inactivated vaccine with a careful analysis of the economic implications of the results and showed that in the year of the study, the influenza vaccine saved $46.85 per person vaccinated. The article by Bridges and colleagues in this issue of THE JOURNAL reaches a different conclusion. This carefully conducted trial of 2375 healthy working adults during the 19971998 and 1998-1999 influenza seasons demonstrates that vaccination provided some clinical benefit but, from the societal perspective, no overall economic benefit. However, this study differs from the study by Nichol et al in several important characteristics. Bridges et al determined illness rates and infection virologically, whereas Nichol et al did so only using reports of respiratory illness obtained by telephone interviews from recipients of vaccine and placebo. Because influenza vaccine should protect only against influenza, the difference between the 2 groups in total illness rates can be taken as the approximate attack rate of influenza. The rate is actually likely to be higher because influenza vaccine is not 100% efficacious and some of the illnesses in vaccine recipients must have been influenza. It was this difference, an imputed attack rate of at least 35 per 100 persons, that resulted in the estimation by Nichol et al that