Current issues of ACP Journal Club are published in Annals of Internal Medicine


The long-term effects of weight cycling are inconclusive

ACP J Club. 1995 Mar-April;122:51. doi:10.7326/ACPJC-1995-122-2-051

Source Citation

National Task Force on the Prevention and Treatment of Obesity. Weight cycling. JAMA. 1994 Oct 19;272:1196-202.



To determine, by a review of the literature, the effects of weight cycling and to evaluate the risks and benefits of attempts at losing weight.

Data sources

Original studies were identified by searching MEDLINE and Psychological Abstracts from 1966 to 1994 for articles on weight cycling, yo-yo dieting, and weight fluctuation. Additional studies were identified by manually reviewing bibliographies.

Study selection

English-language articles that evaluated the effects of weight change or weight cycling on humans or animals were reviewed. Studies of humans were emphasized. Studies were reviewed by experts in nutrition, obesity, and epidemiology to evaluate study design and the validity of the conclusions.

Data extraction

Data on age, sex, weight, main outcome measures, and the definition used for weight cycling were extracted.

Main results

43 studies met the selection criteria: 22 were cross-sectional or prospective studies evaluating the metabolic and psychological effects of weight cycling on humans; 6 were observational population-based studies with a primary goal of determining mortality in humans; and 15 were animal studies. Most studies evaluating the influence of weight cycling on metabolism showed no adverse effects on body composition, resting metabolic rate, body fat distribution, future successful weight loss, or risk factors for cardiovascular disease. 4 large population-based observational studies showed increased risks for all-cause and cardiovascular mortality with weight variation, but 2 smaller studies showed no such effect. The 4 large studies had several limitations; for example, no distinction was made between intentional and unintentional weight loss, body composition and fat distribution were not controlled for, and the studies were not designed to determine the effects of weight cycling in obese persons and in those with normal weight. Most participants in these studies were either nonobese or only mildly obese. Few well-controlled studies have assessed the effect of weight cycling on psychological functioning. Conclusive data on the long-term health effects of weight cycling are lacking.


Most studies did not show an adverse effect of weight cycling on metabolism. Several large studies showed an association between weight variation and increased mortality and morbidity, but the study designs have limitations. Conclusive data on the long-term health effects of weight cycling are lacking.

Source of funding: Not stated.

For article reprint: Dr. S.Z. Yanovski, Division of Digestive Diseases and Nutrition, National Institute of Diabetes and Digestive and Kidney Diseases, Natcher Building, Room 6AN-18, 45 Center Drive, Bethesda, MD 20892-6600, USA. FAX 301-480-8300.


Weight cycling, or "yo-yo" dieting, is common in our weight-conscious society. Its frequent antecedent, obesity, is the leading form of malnutrition in the developed world. Obesity is second only to smoking as a contributor to morbidity and mortality (1) and bears an undeserved social stigma. As with nicotine addiction, obesity is difficult but not impossible to prevent and treat. Although new molecular evidence (2) suggests that some persons have a genetic predisposition, the dramatic increase in the prevalence of obesity from 25% to 35% in the past decade (3) must be caused by environmental factors such as physical inactivity, caloric increases, and smoking cessation.

The Task Force's finding that metabolic effects of weight cycling are not worrisome will be less controversial than its finding that the data on mortality were inconclusive and should not deter the obese patient from losing weight. Faced with the same inconclusive mortality data and the high recidivism rate among patients who diet, one could conclude that primum non nocare requires that we discourage most dieting. More studies that focus on this question are needed to settle this issue.

What should we advise our patients? It seems prudent to discourage the preoccupation with weight on both scientific and humanistic grounds for those who are not obese or who are marginally obese and, perhaps, even for patients who are moderately obese. Weight loss should be encouraged for the severely obese or for the less obese patient with or at high risk for complications because of family history or abdominal obesity.

Weight control must be individualized. Not all patients with the same body profile have the same risk profile or the same antecedents for weight gain. We can best serve our obese at-risk patients by helping them to find personally compelling reasons for losing weight, to choose attainable rather than "ideal" goal weights, and to set themselves up for success.

How? By taking the long-term view, making gradual changes, increasing physical activity, identifying inappropriate eating cues, reading food labels, and becoming more skeptical and knowledgeable consumers who believe in their ability to make positive lifestyle changes. Although only the patient can do it, weight control is one area in which the involved, caring physician can truly make a difference.

Lawrence J. Cheskin, MD
Johns Hopkins UniversitySchool of MedicineBaltimore, Maryland, USA


1. McGinnis JM, Foege WH. Actual causes of death in the United States. JAMA. 1993;270:2207-12.

2. Zhang Y, Proenca R, Maffei M, et al. Positional cloning of the mouse obese gene and its human homologue. Nature. 1994;372:425-32.

3. Update: prevalence of overweight among children, adolescents,and adults—United States, 1988-1994. MMWR Morb Mortal Wkly Rep. 1997;46:198-202.