Abstract
Abstract
In spite of the obvious differences in the diseases under study, designs of trials of prevention of HIV and cardiovascular disease share some common features. A trial of prevention should identify a population at risk for the disease; it should have a clearly defined, clinically important, outcome; it should be large enough to have sufficient power to detect an effect of public health importance; and participants should be followed long enough for the effect of the intervention to become manifest (but the trial should not take so long as to render the intervention no longer of interest). Many cardiovascular prevention studies have been large, simple, randomized trials leading to easily interpretable results. This paper urges that designers of trials of HIV prevention should consider mimicking the strategies used in cardiovascular disease prevention trials: focus on a clear inferential path to the question being asked while limiting unnecessary data collection, auditing, and complexity. Finally, showing benefit in a trial is only the first step in reducing the burden of disease: aggressive, effective efforts at educating the population at risk so they will implement the intervention is essential to improvement in public health.