Weight-loss and maintenance programs usually focus on the quantity of weight reduction achieved and maintained over time. Caloric restriction and other features of a weight-loss program, however, may be associated with various important health effects. Many overweight persons have related medical conditions, such as hypertension, hypertriglyceridemia, and non-insulin-dependent diabetes mellitus, that may respond favorably to long-term weight loss. Weight loss must be directed in a prudent manner to avoid adverse medical problems such as the development of cardiac arrhythmias, dehydration, or electrolyte imbalance. Therefore, successful weight-loss programs must be presented to the public in a manner showing that weight loss may be achieved safely as a result of the responsible use of the programs.

To address the efficacy and safety of weight-loss programs, the National Institutes of Health and the Food and Drug Administration requested data from industry on the success of programs targeted for use by persons with various degrees of obesity. Six weight-loss centers and three pharmaceutical companies producing nonpharmaceutical weight-loss products were contacted by letter and asked to provide information about their programs and products specifically for the National Institute of Health's Technology Assessment Conference, which was held from 30 March to 1 April 1992. Data were requested about the nature of the weight-loss programs, numbers and characteristics of participants, and outcomes. The companies were assured that the data submitted would be compiled and presented without naming specific programs or products. We have therefore identified companies with letters only, beginning with "A" and progressing sequentially in the order of discussion, and have omitted names of authors of unpublished data provided.

The data reviewed in this paper reflect only industry-submitted medical literature. Much of the data are preliminary findings and are still being revised by company researchers. This paper is not a comprehensive review of weight-loss programs. Rather, its objective is to evaluate industry's supporting evidence for the success of weight-loss programs. We received material from five companies, including 60 reprints of different articles published in peer-reviewed medical literature. Information was provided by the makers of three nonphysician-directed weight-loss programs and two physician-directed programs. Table 1 summarizes published and unpublished data that were reviewed for this article. For the first very-low-calorie diet (VLCD) reviewed, the company provided 55 publications; due to space limitations, we chose 11 of these for inclusion in the table. We chose these publications because they were the most helpful in evaluating the program.

We reviewed the data with three primary considerations in mind: 1) What evidence, if any, suggests that the program is safe and effective? 2) Does the program use foods readily available to most U.S. consumers, food products for special dietary use, or VLCDs supervised by a physician? 3) Does the program also rely on behavior modification, exercise, or drugs? We examined selection methods for study participants and conclusions drawn from data analysis.

Diets without Direct Physician Supervision

Three of the weight-loss programs that we examined did not require direct physician supervision and consisted of some combination of reduced-calorie meals or meal replacements, education, behavior modification, individual or group counseling, and aerobic exercise. One was a self-directed meal replacement plan, and two were structured weight-loss programs.

Meal Replacement

The first program consisted of over-the-counter preparations that were intended as substitutes for one or two meals daily. The preparations were special dietary products whose labels included general guidelines for use. The instructions warned against unsupervised use by pregnant and lactating women, by children younger than 18 years, and by persons compromised by any health problem. The label also stated that this program is not to be used for weight losses of more than 14 kg or more than 15% of starting body weight. The diet regimen was designed to provide a total daily intake of 1250 calories. Each serving of the meal replacement provides 190 calories, with 29% from protein, 67% from carbohydrate (including 2 g of dietary fiber), and 4% from fat.

We were provided with one published study and three associated abstracts from company A. In a published clinical trial, decreases in blood pressure and cholesterol levels in overweight persons with cardiovascular disease and improvement in walking times in overweight arthritic patients were reported as examples of short-term success of weight loss [2]. This trial, although of short duration and small sample size, supports the claim that weight loss can positively affect existing medical conditions. It would be useful to examine these findings in a larger group, with a longer follow-up. The abstracts described studies that evaluated the efficacy of the product for initial weight loss and maintenance [3, 4]. They did not, however, provide sufficient detail regarding study methods to allow a full interpretation of the data. The evaluable information indicated that the studies had basic design problems including poor response rates to surveys, inadequate follow-up to assess maintenance, high dropout rates for clinical trials, and insufficient group sizes to detect statistically significant differences.

Structured Weight-Loss Programs

The second weight-loss program consisted of individualized nutrition instruction, behavior modification, exercise, individualized meal plans, and supervised weight maintenance. The meal plans varied from 1100 to 2000 calories per day, depending on individual needs, and provided approximately 60% of calories from carbohydrate (23 g of fiber), 15% from fat, and 25% from protein. Four studies conducted by company B to evaluate its program were submitted to us for review but were subsequently withdrawn.

A third weight-loss program submitted information that was inadequate for proper assessment of the program's efficacy. On the basis of the information received, the basic program consisted of prescribed menu plans for adults, ranging in caloric content from 1000 to 1800 calories per day. Approximately 60% of calories were provided from carbohydrate, 20% from protein, and 20% from fat. According to the program information provided by company C, one formal study following participants after treatment has been completed but was not available for review. Some data from this study were received for review along with a description that was inadequate to support the results or conclusions about the program's success. Dropouts from this program were not tracked because all participants remain at liberty to redefine their weight goals as they progress.

Physician-Supervised Diet Programs

Two weight-loss programs required physician supervision of patients consuming VLCDs. Both programs included very-low-calorie formulas that were similar in composition. Company D sent us 55 publications, including 5 that are currently in press. Company E provided three published articles and several abstracts that addressed program safety and efficacy.

The first physician-supervised plan that we examined was described as a 12-week supplemented fast, during which patients consume primarily the program's product and noncaloric beverages. This program is recommended only for patients who are at least 30% and 18 kg overweight and is available only through physicians who have received specialized training to administer the plan. Two formulas have been used, both of which provide vitamin and mineral supplements to meet the U.S. recommended daily allowance or the estimated safe and adequate daily dietary intakes of essential minerals. The 800-calorie formula provided 70 g of protein, 100 g of carbohydrates, and 13 g of fat per day. Another formula, which was evaluated in earlier studies, provided a total of 420 kcal per day with 70 g of protein, 30 g of carbohydrate, and 2 g of fat. Because of poor compliance and insufficient proof of improved weight loss, the 420-calorie formula was recently eliminated from the program. A recent recommendation also included a phase-in period of 7 days, during which 1200 calories per day is consumed, 800 from the formula and an additional 400 from prepared food. The fasting regimen has a duration of 12 weeks, after which food is gradually reintroduced over 6 weeks. Maintenance is directed by the physician.

The studies that we reviewed can be grouped in three areas: the first evaluated initial weight loss and weight-loss maintenance; the second focused on product safety in terms of cardiac function, liver enzyme levels, blood glucose level, and lipid levels; and the third examined the effect of weight loss on underlying medical problems such as hypertension, hypertriglyceridemia, and non-insulin-dependent diabetes.

In 25 studies, six of which are included in Table 1, VLCDs were evaluated for their effectiveness in weight loss and maintenance. A 1989 study by Kanders and colleagues [7] found the 800-kcal formulation to be as effective in weight loss as the 420-kcal formulation. Other studies have also shown that those who are successful in reaching their goal weight are also more successful at maintaining the loss [8, 9]. Several well-controlled studies provided information about adjunct treatment necessary for weight loss and maintenance [10, 11]. It was concluded that behavior modification therapy and supervised exercise are necessary for successful maintenance of weight loss.

Product safety was the focus of 14 studies that examined the occurrence of metabolic abnormalities reflected by changes in liver enzymes, blood glucose levels, and lipid levels. One review found that the administration of VLCDs by trained physicians to patients meeting recommended guidelines resulted in only minor complications [12]. The authors cited examples of major complications resulting from over-restriction of caloric intake, binge eating after severe caloric restriction, or use by persons with only mild obesity. Complications reported include dehydration, electrolyte imbalance, orthostatic hypotension, hyperuricemia, and ventricular arrhythmias. Two other studies that addressed safety are summarized in Table 1. One is an older study supporting the claim that the quality of the VLCD rather than the number of calories is the factor affecting cardiac arrhythmias [13]. The other study examined liver function in patients receiving VLCDs and found no significant changes during the 8-week period [14].

Fourteen studies examined the effect of weight loss on underlying medical problems. In most studies, lower blood pressure, improvement of non-insulin-dependent diabetes mellitus, and lower triglyceride levels have resulted after weight loss. Three studies summarized in Table 1 showed improvement in persons with medical conditions that existed before the start of the VLCD [15–17].

In summary, the data sent by company D suggested that many obese persons benefitted from a physician-controlled VLCD in combination with exercise and behavior modification. When the diet was administered properly, several chronic health problems showed improvement, and no serious adverse reactions developed. The studies cited were, for the most part, well controlled and had adequate sample sizes, low dropout rates, and well-supported conclusions.

Company E, a second manufacturer of physician-supervised VLCD products, provided information on three supplements ranging from 525 to 800 calories per day. Protein components ranged from 38% to 51% of calories, carbohydrates from 44% to 60%, and fat from 2% to 11% (depending on the product). Vitamins and minerals provide 100% to 150% of the U.S. recommended daily allowance in all supplements.

We reviewed two published studies and several unpublished abstracts that presented data on weight loss with the use of this VLCD. In one study, a combination of a VLCD and exercise did not show an effect superior to diet alone on metabolic rate or weight loss in obese women [18]. Weight loss and metabolic responses of the first 100 persons consecutively enrolled in a comprehensive VLCD program were documented in another study [19]. Abstracts presented at meetings within the last year also considered health outcomes and the effects of exercise and behavioral changes with VLCD treatment. Further data on these topics is needed to allow an adequate assessment of the results and conclusions presented.

Conclusions

Manufacturers of several popular weight-control programs in which participants are not supervised directly by a physician provided us with limited data. Some of the programs consisted of meal replacement preparations that are readily available to the public and provided written guidance on the label of the products as to their proper use. It is difficult to track users of such products, and clinical trials regarding weight loss and maintenance with these products have been conducted with inconclusive results. Other weight-loss programs use reduced-calorie products that are provided by weight-loss centers. At the centers, clients may also receive personalized instruction for dieting and maintenance. Although these studies are more controlled in terms of following clients who participate, the data are still not complete enough to draw valid conclusions regarding the success of these programs. In recent years, the number of studies has increased, and the current studies are of better quality, indicating that industry may be responding to the need for more valid data.

Probably because of the potential for adverse health effects through misuse, physician-supervised programs that use VLCDs and that are limited to obese persons are more widely studied and published. We concluded from the available data that VLCDs are helpful in obtaining and maintaining weight loss for 2 to 3 years in some persons when used in conjunction with behavioral modification and an exercise program. The current products promote a safe method of weight loss when used under a trained physician's care and when instructions are followed regarding the proper intake of the product and time limits on its use.

Because controlling excess body weight is a prevalent practice that affects the public health, further investigation into variables that can help persons to achieve and maintain an optimal weight is essential. Weight-loss programs can no longer rely on informal surveys or poorly controlled studies because it has been shown that valid, scientific data can be obtained. Given the importance to public health of reducing overweight and given society's interest and economic investment in food and lifestyle changes aimed at achieving a healthy body weight, rigorous studies of current weight-control practices should be pursued aggressively.