Does weight cycling present a health risk?

I review research on the hypothesis that weight fluctuations caused by unsuccessful dieting are hazardous to one's health. Recent epidemiologic findings show that weight variability over time is associated with increased total and cardiovascular mortality (relative risk: 1.5-2.0), independent of a variety of possible confounding variables. Although these findings are consistent across studies, methodologic limitations of a lack of a uniform or standard definition of weight cycling, and the linking of weight variability to unsuccessful dieting raise serious questions about whether these findings should be interpreted as supporting the weight-cycling hypothesis. The absence of data identifying a plausible biological mediator for weight fluctuation per se as a health hazard is also a problem. It is concluded that, although epidemiologic data on weight variability and health are intriguing, they are at present insufficient to alter public health recommendations regarding weight control.