Intentional weight loss in obese persons has been shown to improve health in both epidemiologic and clinical intervention studies. Williamson et al. (16) have shown that intentional weight loss decreases both cardiovascular and overall mortality. Weight loss also decreases morbidity (17).

Short-term studies have demonstrated that weight loss resulting from low-calorie diets, very-low-calorie diets, anorectic drugs, or bypass surgery is associated with increased insulin sensitivity, improved insulin secretion by islet cells, decreased hepatic glucose production, and improved glucose disposal (1,18,19,20,21,22,23). Wing et al. (24) conducted a long-term study in which diabetic patients lost an average of 5.6 ± 4.0 kg during the active intervention phase and maintained a weight loss of 4.5 ± 7.5 kg at 1 year. Those who were most successful in losing weight had significant improvements in fasting blood glucose and glycosylated hemoglobin values at 1 year even though they remained 20% above ideal body weight at the end of the study. Reductions in insulin dose or oral hypoglycemic medication were made for 100% of patients who lost 6.9 to 13.6 kg, 46% of those who lost 2.4% to 6.8 kg, and 40% of those who lost 0 to 2.3 kg. In the United Kingdom Prospective Diabetes Study (UKPDS) study (25), the response to diet was studied in 3,044 patients with newly diagnosed type 2 diabetes. Their mean fasting plasma glucose level was 12.1 ± 3.7 mmol/L. The investigators determined that the weight loss needed to achieve normalization was 16% of ideal body weight in patients with initial fasting plasma glucose levels of 6 to 8 mmol/L, 21% in those with levels of 8 to 10 mmol/L, 28% in those with levels of 10 to 12 mmol/L, 35% in those with levels of 12 to 14 mmol/L, and 41% in those with levels greater than 14.0 mmol/L. At the 15-month assessment, 482 patients maintained normal fasting plasma glucose levels. On average, they had slightly lower fasting plasma glucose levels at study entry and had lost more weight than did other patients (25). After 3 months, 742 patients achieved normal fasting plasma glucose levels. However, it is important to remember that there will be nonresponders, and obese individuals who remain hyperglycemic after a weight loss of 2.3 to 9.1 kg are unlikely to improve with further weight loss and should be considered for treatment with hypoglycemic agents (26).

Weight loss through diet and exercise can reduce the risk of the conversion of impaired glucose tolerance to frank diabetes. This has been shown in studies in China (27), Finland (28), and the United States (29). In the Diabetes Prevention Program in the United States, which studied more than 4,000 patients over 5 years, the progression to diabetes was decreased by 58% (29).

Ross and Rissanen (30) have shown effects of diet and exercise on insulin sensitivity, preferential loss of visceral fat, and blood pressure. The effect of weight loss on hypertension has now been documented in a number of randomized clinical trials. Long-term studies have confirmed the effectiveness of weight reduction in lowering blood pressure and enabling some patients to become normotensive without the use of antihypertensive drugs and for others to reduce the dosage of required drugs or the number of drugs taken (31,32,33,34,35,36,37,38,39,40,41,42). The marked improvement found in many unmedicated hypertensive patients suggests that weight reduction should be the initial treatment for an obese person with hypertension. In patients for whom elimination of antihypertensive drugs is not a realistic goal, weight loss can potentiate the effects of drug therapy (40,41,43).

The dyslipidemia of obesity is characterized by high triglycerides, reduced high-density lipoprotein (HDL) cholesterol, and small dense low-density lipoprotein (LDL) particles (44,45). With weight loss, triglycerides fall, HDL cholesterol rises, and small dense LDL particles become larger and less atherogenic (45). Short-term studies have confirmed the value of behavioral, low-calorie diet, and very-low-calorie diet interventions in decreasing total and LDL cholesterol levels and the HDL/LDL ratio and lowering triglyceride levels (46,47,48,49,50). Additional long-term studies, of greater than 1 year, have also been done and have shown a sustained improvement in the lipid profile of overweight and obese patients (33,36,51,52,53,54). Although some of these lipid changes are relatively small, they should not serve as a disincentive to embarking on a weight-loss program, because even modest improvements in lipid levels are associated with a decreased risk of cardiovascular disease (1,17,55).

Few trials have explicitly addressed the relationship between weight loss and cardiovascular disease per se. A 10-year study of 2,500 patients from the Framingham Study cohort conducted by Higgins et al. (36) found a positive association between weight loss and cardiovascular disease, although this finding almost certainly results from the fact that the study does not distinguish disease-related involuntary weight loss from voluntary weight reduction.

Respiratory function is often impaired in obese individuals because of reduced lung volume, altered respiratory patterns, and decreased respiratory system compliance (7). Obesity-hyperventilation syndrome and sleep apnea are the most common respiratory disorders associated with severe obesity (8). Short-term studies of sleep apnea have indicated that weight loss resulting from low-calorie diets or gastric bypass surgery is associated with a marked reduction in apneic episodes and night awakenings and with a general improvement in sleep patterns (56,57,58,59,60,61).

Obesity is often associated with an increased risk of gout (62), as is increased central fat distribution (63). When persons lose weight, their uric acid levels may initially increase (1,64), but precipitation of gout by weight-loss therapy in asymptomatic patients is uncommon (65). Uric acid levels rarely become high enough to necessitate medical treatment (66). Occasionally, large amounts of uricosuria and a predisposition to renal uric acid stone formation occur. When the appropriate diagnosis is made, urine alkalinization is the treatment of choice.

Cross-sectional studies have repeatedly found an association between increasing weight and increased prevalence of osteoarthritis of the knee (67,68,69) and, to a much lesser extent, of the hip (70). There have been just a few studies that have related weight reduction to the reduction of the onset of arthritis (71) or to the improvement of arthritic symptoms (72).

Obese persons are at a greater risk for developing gallstones than are persons of normal weight, probably because their bile is more highly saturated with cholesterol, their gallbladders are larger and less contractile, and their triglyceride levels tend to be higher (7). An association between loss of large amounts of weight and gallstone formation has been reported, although this finding has not been observed consistently (73,74,75,76). However, once individuals lose weight, their risk of gallstone formation and of cholecystitis decreases because the risk factors previously cited improve.

Available data suggest that obese persons may have a greater tendency to depression than do lean persons, but there are no data to suggest that obese persons are more prone to psychosis (77,78). The depression generally improves with weight loss (79).

Obesity is associated with increased health risk, and this is especially so in patients with diabetes (80). The goals of therapy are to improve health risk. There are two ways to approach the
risk—one is to modify weight and the other is to focus on and try to modify other health-related variables (81). Focusing on other health-related variables can be done with individuals for whom the initiation of a weight-loss program seems impossible or impracticable.

It is common for patients who are beginning a weight-loss program to have faulty and unrealistic beliefs about how much and how rapidly they can lose weight. It is important to counsel them in this regard to prevent disappointment and attrition.

Most obese persons wish to lose all of their excess weight and return to normal weight (82). While this may be realistic for someone whose BMI is close to normal (26 to 30), it is not realistic for individuals whose BMIs are higher. For these persons, a weight loss goal should be set for a reduction of 10% to 15% from their present baseline weight (2) for a number of reasons. First, it is very difficult for most persons to lose more than this, and it is consequently damaging for their self-image, sense of success, and satisfaction when they set goals that are unreasonable and generally unattainable. Second, if a patient is pushed to lose more, such as can occur with very-low-calorie (300 to 500 kcal) diets, experience has shown that, while persons can lose weight quickly, they will generally regain it almost as quickly (83).

The traditional technique of handing a patient a printed description of a 1,200- or a 1,500-kcal diet, complete with specific menus and specific portion sizes, and telling him or her to increase exercise without further instruction, has generally been unsuccessful. With inadequate education and support, patients quickly abandon such programs and never increase their activity.

In an effort to prevent such failure, the use of behavioral therapy has increased. Approaches to changing behavior began with Ferster et al. (84) and were greatly popularized by Stuart (85). The goal of such therapy was to modify maladaptive eating by improving environmental control. The aim was to change eating behavior (86). This approach was found to be too narrow, and more recent adaptations have recognized obesity as much more complex than just maladaptive eating, with influences from genetic, physiologic, psychological, and social factors (87). The newer approaches have targeted not only eating but also physical activity. In the 1970s, cognitive-behavioral techniques were initiated (88,89,90). A number of psychologists have written about these treatment techniques (91,92,93,94,95).