Observational epidemiology is the preferred means of evaluating effects of behavioral and lifestyle modification

In evaluating the health effects of behavioral change, randomized trials have the clear advantage of being able to provide unbiased and unconfounded estimates. However, in practice, many difficulties loom that limit the utility of trials. These include the inability to assess adverse behavior (except indirectly), the often unrealistic requirement for very prolonged adherence--perhaps for decades--to the randomized assignment, the requirement for huge numbers for precise estimates of effect (as opposed to simply a "significant result"), and the ethical problem of giving a placebo when an efficacious agent is available. We must be alert to feasible opportunities to test behavioral changes in clinical trials, but the many limitations mean that we will continue to rely on well-conducted, large-scale observational studies for most of our evidence on the effects of behavioral change.