Recommendations for Medically Supervised Programs

Very-Low-Calorie Diets

In 1988, the American Medical Association's (AMA) Council on Scientific Affairs recognized the medical treatment of obesity and established appropriate treatment guidelines [1]. One of these treatments, the very-low-calorie diet (VLCD), typically provides fewer than 800 calories per day, although the American Dietetic Association considers any diet of less than 1000 calories per day to be a very-low-calorie regimen. These diets are attractive to patients because they induce rapid weight loss and because they are easy to follow due to the imposed limitation on the amount and type of food that can be consumed; patients also report no hunger after several days on the program. They are attractive medically because they can improve weight-related disorders and reduce health risks [2].

Medical supervision of VLCD programs is mandatory due to the metabolic characteristics of semistarvation (fatigue, weakness, and lightheadedness) and related changes in vital signs (blood pressure, heart rate, and respiratory rate). These patients often have obesity-related medical complications such as hypertension or diabetes, which require physician monitoring for appropriate adjustment of medication [3]. Very-low-calorie diets are associated with only minor complications when administered to carefully selected patients by physicians trained in their use [4, 5]. The importance of careful patient selection is shown by attrition rates of 33% to 50% among patients receiving VLCDs [6]. As identified by the Council on Scientific Affairs [1], the Medical Letter on Drugs and Therapeutics [7], and various investigators [8], not all VLCDs have the same metabolic effects, protein-sparing ability, or safety.

At the Technology Assessment Conference, Wadden [9] reviewed the results of clinical trials that used VLCDs and noted that "approximately 90 percent of persons treated by a VLCD and behavior modification will attain a weight loss of this magnitude (9.1 kg or more) and 50 percent a weight loss of 18.2 kg or more". He went on to conclude that "efforts are needed to identify methods of making available. the intensive treatment approach developed in research trials"—an intensive approach that, without exception, involves a supervising physician.

Low-Calorie Diets

The AMA's Council on Scientific Affairs also recognized the use of low-calorie diets (LCDs), which provide more than 800 calories per day [1]. Low-calorie diets represent the most common and, when compliance is monitored, safest treatment of obesity. Compliance with LCDs usually ensures that weight loss does not exceed 0.5% to 1.0% of initial body weight per week, which is recommended as safe and acceptable by the U.S. Dietary Guidelines [10].

When monitored by a physician and supervised by a registered dietitian or clinical nutritionist, the LCD should be sufficient for most patients to promote a 10% to 15% weight loss over 10 to 20 weeks. This level of weight loss results in most of the health benefits associated with weight loss [11, 12]; without documented weight maintenance success, additional weight losses of more than 10% must be justified by the supervising physician [13]. Due to the level of carbohydrate intake [14], these regimens are rarely associated with the consequences of starvation metabolism sometimes encountered with VLCDs.

Physicians must become involved in obesity treatment when the patient has concurrent disease or other risk factors for fatal complications. Overweight persons have an increased risk for morbidity and death that is proportional to their degree of obesity and that also increases with age and the presence of additional disease [15]. A body mass index (BMI) of 25 to 30 kg/m² (independent of concurrent illness) may be described as incurring low risk, 30 to 35 as moderate risk, 35 to 40 as high risk, and anything greater than 40 as very high risk [16]; an excellent nomogram is also available for use [17].

Recommendations for Medical Monitoring of Unsupervised Programs

At any given time, an estimated 65 million Americans diet to lose weight, and many of these persons participate in commercial programs or consume over-the-counter products. Recent reports of gallstone formation (requiring cholecystectomies), cardiac arrhythmias, and even death have generated concern regarding the safety of some unsupervised commercial programs [3]. To a large extent, the lack of medical supervision and the lack of guidelines to indicate when medical supervision is necessary are responsible for diet-related morbidity and death.

In general, however, medically unsupervised dieting is safe provided that the patient does not have any concurrent medical problems and is not severely overweight (BMI >35) and provided that the weight loss does not exceed 1% of initial body weight per week [13]. In addition, the diet macro- and micronutrient composition must be safe and balanced [18]. Deviation from recommended dietary guidelines (for example, too little fat, no meal with at least 10 g fat, excessive protein) can lead to medical problems including the formation of gallstones [19, 20].

Persons planning to lose weight on their own should see their physician for a baseline physical examination to identify any underlying diseases and to discuss weight loss strategies and goals (Table 1). In addition to completing a thorough examination and medical history, physicians should collect information on the patient's nutritional history as well as on the waist and hip circumferences (to monitor reduction in risk through changes in waist-to-hip ratio). Weight loss should not exceed 1% of body weight per week; if weight is lost faster, the physician should suspect unsupervised use of a VLCD (conversely, weight loss at less than 0.5% of body weight per week may indicate low compliance with lifestyle changes). Physician visits should subsequently be scheduled after any 10% change in body weight (loss or gain) and whenever a change in health status is noted. If fatigue, muscle weakness, lightheadedness, or new health problems arise, the physician should be notified promptly [3].

Published Comparison of Medically Supervised and Unsupervised Programs

Although no large, randomized, clinical trials have specifically compared weight loss results between patients receiving medical supervision and those not receiving supervision, data in the literature suggest a benefit for physician involvement. In one study [21], family practice residents were randomly assigned to experimental (received training in weight reduction, which they passed on to their obese hypertensive patients) or to control (no training; obese hypertensive patients received usual care) groups. Control physicians were allowed to offer dietary advice but were given no special education materials or training. During the first 6 months of the study, patients seen by experimental physicians lost weight (1.8 kg) and significantly reduced the number of antihypertensive drugs taken; those seen by control physicians gained weight (0.56 kg). During the second 6 months, patients seen by experimental and control physicians gained weight (0.94 kg and 0.73 kg, respectively). At the end of the year, experimental group patients lost weight (0.88 kg), whereas control group patients gained weight (1.3 kg). In comparing gainers and losers regardless of group assignment, weight loss was associated with more physician visits (9.3 compared with 6.0 visits) and with a reduction in the number of antihypertensive medications used ( –0.62 compared with +0.08 medications). The authors concluded that "the apparent success was achieved by educating the physicians, increasing the frequency of visits, and taking these opportunities to educate the patients (which) involved little extra expense or effort" and that "frequent visiting is important in encouraging weight loss". They also noted that the family medicine residents were "enthusiastic when they recognized the intervention as a potential means of improving their effectiveness".

Table 2 lists statistics comparing weight change results of dietary intervention (medically supervised) and control (unsupervised) patients enrolled in studies designed to examine the effect of weight loss on health measures such as blood pressure, glucose control, and lipid profile. In addition to the significantly greater weight loss noted among patients in the intervention group, most controls gained weight. Because potential participants must be willing to accept any of the trial interventions, those volunteering for these studies are most likely to be interested in losing weight. Stamler and colleagues [22] specifically excluded persons expressing an unwillingness to change. At baseline, those randomized to the control group could be assumed to have equal motivation to lose weight as those randomized to intervention groups (although, in some cases, controls are specifically told not to diet or to make other lifestyle changes, and in other studies they are considered to be "self-help" patients, particularly if their participation in screening and enrollment has heightened their awareness of the value of weight control in improving health or if they receive educational materials for home use).

In one study, we compared the outcome of 187 obese hypertensive patients using medically supervised VLCD and LCD programs [23]. Those receiving the VLCD (a protein-sparing modified fast) experienced a 17% reduction in initial body weight, whereas those receiving the LCD (a balanced deficit diet) lost 12% of initial body weight (Table 3). During the 6 months of active maintenance (medically supervised), patients regained only 5% (VLCD) and 2% (LCD) of their body weight. For the next 18 months, patients had no clinic contact and came in for milestone visits only. During this period, patients regained weight at a faster rate, maintaining only 41% (VLCD) and 38% (LCD) of their initial weight loss at 18 months.

In a study of 49 overweight hypertensive patients, Ramsay and colleagues [24] compared the effectiveness of physician advice alone, physician advice plus written and oral diet instructions (given by a physician), and physician referral to a dietitian (who was unaware of the study) for instruction on body weight and blood pressure. After this encounter (advice, diet sheet, dietitian), management during the 1 year of follow-up was routine and voluntary until the second and last data collection visits (9 to 15 months after the baseline visit). The results indicated that intervention of any kind had some effect, with one third of all patients losing 6 kg or more 1 year after randomization (Table 4).

Other Evidence That Physician Supervision Makes a Difference

Even if physician supervision is not required for weight control, evidence suggests that physician participation could make a difference in patient outcome. Published reports indicate that physicians can influence patient behavior with regard to smoking cessation [25, 26]. Similarly, patient outcome in the treatment of hypertension improved after a single physician tutorial, which led to more time spent in patient education [27]. Cohen and associates [21], as described above, showed the ability of physicians to make a difference in body weight and clinical outcome.

Recent reviews [28] of the doctor-patient relationship indicate that it provides many opportunities to encourage behavioral change, to reinforce positive changes (for example, improved dietary practices, increased activity, and steady weight loss and control), and to encourage use of community resources through appropriate referrals to outside agencies and facilities (for example, local registered dietitian, acceptable supervised and unsupervised weight-loss programs). Patients appear to be willing to listen to advice from their physicians, with some authorities reporting that people expect such advice on maintaining health [29].

In addition to providing guidance to weight-loss efforts among overweight and obese patients, physicians can help prevent the development of obesity by alerting patients at risk for weight gain during routine medical examinations. The incidence of substantial weight gain during a 18-year period has been shown to be highest among adults 25 to 34 years old (particularly women and those who are already overweight) and to decrease with increasing age [30].

Specific data regarding the success and cost-effectiveness of physician involvement in weight-loss attempts should come from the Trial of Antihypertensive Interventions and Management (TAIM) study. The investigators reported that a weight loss of 4.5 kg or more lowered blood pressure in a manner similar to low-dose drug therapy and potentiated drug effect [31]. These 6-month results and the accompanying summary of other hypertensive trials suggest that physicians can assist in producing the modest amount of weight loss (that is, 10%) required to improve health.

Direction for Future Research

Future work should define the medical criteria to determine when a dieter should be treated under physician supervision. Identification of those persons at high risk for disability, morbidity, and death due to their obesity and of the role for adjunct treatment with weight control drugs and gastric surgery must be part of the physician's responsibility. In addition, treatment standards should be established for programs that are medically supervised (for example, indication for electrocardiograph, frequency of electrolyte monitoring, and required blood tests) [32, 33]. Similarly, guidelines for unsupervised programs should be established to indicate when and to what extent physicians should be involved.