Drs. Mazza and Yale have identified the two kinds of studies that characterize proof of value for influenza vaccines. Efficacy studies constitute proof that influenza virus infection and infection-specific illness and its consequences are reduced among persons given vaccine, preferably in randomized, placebo-controlled studies. Such studies have been numerous since vaccines were first described (1-3). The second study type is one in which use of influenza vaccine in a population is followed by reductions among vaccinees of influenza-like illnesses or the consequences of these illnesses during periods in which influenza virus infection is prevalent. This type of study, which is called an effectiveness study, generally involves larger population groups than efficacy trials and circumstances in which accurate detection of every infection may not be feasible. Nevertheless, such studies are beneficial for assessing value, particularly in cost–benefit analysis, since use of vaccine as recommended must be beneficial for preventing illnesses during the period of prevalence of the infection. For example, if efficacy was demonstrated for a vaccine but influenza infections were replaced among vaccinees by a similar infection and illness caused by another respiratory virus, and acute respiratory illnesses or their consequences did not change, effectiveness of influenza vaccine would not have been demonstrated, and no cost benefit would have been realized.

It is preferable to have precise data on the efficacy and effectiveness of the influenza vaccine in the same clinical trial, but using the period of prevalence as a surrogate for infection provides useful estimates of effectiveness. It is the combination of both types of study performed many times over many years that provided the consensus for the value of influenza vaccines that supports vaccine recommendations. A thoughtful analysis by Hirota and coworkers (4) and a report on prevention of influenza by the Cochrane Collaboration (5) address both types of study. Because of the variability of influenza and influenza vaccines among different populations and circumstances, it is highly desirable to conduct studies on a continuing basis. Both efficacy and effectiveness studies are needed and, when possible, should be done simultaneously.