THE CASE FOR RANDOMIZED CLINICAL-TRIALS ON THE TREATMENT OF OBESITY

Some say that randomized clinical trials on weight loss are unnecessary (''the benefits are 'obvious''') and others say that such trials are not feasible because too few participants will succeed in maintaining weight loss. Although the intermediate term benefits of weight loss are beyond dispute (lowering of blood pressure, lipids, blood sugar, etc), there is no proof that these benefits will translate into long term benefits, i.e., lower rates of cardiovascular disease and/or lower overall mortality. While this extrapolation may seem obvious, the clinical trials' literature is full of unexpected, adverse side effects of theoretically appealing therapies (e.g., higher mortality with clofibrate and higher cardiovascular disease rates with estrogen treatment in men). Although there is clearly a voluntary component to food ingestion, there are also powerful physiological forces at work which impact on energy balance. For example, individuals of similar height and weight may nevertheless have widely different daily energy expenditures and hence energy requirements. It has been shown in Pima Indians that those with low energy expenditure (i.e., those who are ''fuel efficient'') are more prone to future weight gain than those with high energy expenditure. Also, reduced obese individuals have lower 24-hour energy expenditure than individuals who are spontaneously at the same lean weight. It appears that this deficit in energy expenditure may last for several years, if not longer, implying that reduced obese individuals must exercise far greater vigilance over their caloric intake than their spontaneously lean peers. If they anew themselves to ingest the same number of calories as the latter, they are likely to regain weight, thereby exposing themselves to charges of overeating, even though their caloric intake does not exceed that of the spontaneously lean! Epidemiologic data do not support a benefit of weight loss, Populations such as Mexican Americans, among whom obesity is more common than in the general population, do not have excess mortality past age 45. Life expectancy in the U.S. has improved steadily since the early 1970s, despite a rising prevalence of obesity. Lastly, prospective studies have suggested that people who lose weight die at a higher rate than those who maintain a stable weight. This effect persists even after controlling for latent, subclinical disease and cigarette smoking. Although none of the above considerations prove that voluntary weight loss is bad, they indicate that this treatment should lose its hitherto privileged status and be subjected to the rigors of clinical trials as have been treatments for hyper cholesterolemia and hypertension.