Long-term maintenance of weight loss is an important, but often elusive,goal. Diet and pharmacological treatments for obesity are generally effective at inducing 8-10% weight reductions by 6 months. Thereafter, weight regain is a common phenomenon. Maintenance-phase medication and individual and group follow-up slow weight regain such that weight reductions at program completion average 2-6, 2-7, and 2-7% greater, respectively, than those in control groups receiving no maintenance contact. Consistent and structured eating, frequent self-weighing, and high levels of physical activity acquired through short bouts of brisk walking are pragmatic recommendations to support weight regain prevention.
Obesity has reached epidemic proportions in the United States. In 2005-2006, > 72 million adults, or 34% of the population > 20 years of age, were obese (BMI ≥ 30 kg/m2).1 Obesity is significantly correlated with hypertension, hyperlipidemia,diabetes, and infertility; yet, a 5% reduction in weight can translate into clinically significant improvements in health.2,3 Lifestyle and pharmacological interventions are effective at inducing clinically significant levels of weight loss.4-6 However, maintaining weight loss over a period of years has proved more challenging and has lead to the classification of obesity as a chronic,relapsing disease.7 Respective systematic reviews of pharmacological-, behavioral-, and diet-based weight loss treatments conclude, however, that continued therapy is essential for minimizing weight regain after weight loss.8-12
This conceptual review of the weight regain phenomenon analyzes the efficacy and clinical utility of formal weight loss and weight loss maintenance programs and highlights factors underlying weight regain and weight loss maintenance. Although weight regain prevention remains a challenge, there is room for optimism; a nationally representative survey estimates that 20% of individuals attempting weight loss are able to achieve and maintain 5% weight reductions for at least 1 year.13 Among successful weight losers, various studies indicate that more than 60, 35, and 19% of individuals are able to maintain 10% weight reductions for 1, 3, and 5 years,respectively.14-16
The first section presents the scope of the weight regain phenomenon by analyzing the long-term efficacy of weight loss programs. Data have been gathered from systematic and meta-analytical reviews of randomized controlled trials of nonsurgical weight loss therapy programs. Weight loss surgery is an option for some patients with obesity; a discussion of the risks and benefits of surgery is beyond the scope of this article and will not be considered here.
The second section presents the efficacy of weight loss maintenance programs to support weight reduction and prevent weight regain. The goal is to identify best practices for weight regain prevention. Data are from randomized controlled trials in which program duration was longer than 6 months and the efficacy of a maintenance program was assessed relative to no follow-up support, placebo, or a comparative maintenance intervention group. Trials were identified through a keyMesh search on PubMed for “weight loss maintenance” and “weight regain prevention” and bibliography reviews of identified articles. The outcomes reported are statistically significant group differences in weight reduction, percentage of initial weight loss maintained, and net effect of the intervention on percentage of weight reduction. If not reported, percentage of weight reduction was calculated from baseline weight and the amount of weight lost; the percentage of weight loss maintained was calculated from initial weight lost and amount of weight loss maintained; and the intervention effect was calculated as the arithmetic mean difference in percentage of weight reduction. Baseline weight was not reported in three studies, and weight loss in kilograms was presented instead.
The third section identifies risk factors for weight regain and strategies used by successful weight loss maintainers. Data are from reports based on population surveys including the National Health and Nutrition Examination Survey (NHANES) for 1999-2002,14,17 Consumer Reports,18 and Consumer and Health Styles 2004;19,20 the National Control Registry for Weight Control (NWCR), a longitudinal database containing information on > 4,000 self-selected individuals who have maintained a weight loss ≥ 13.6 kg for at least 1 year;21 and posthoc analyses of randomized controlled trials.
Long-Term Efficacy of Weight Loss Programs
The mainstays of weight loss therapy include diet, exercise, behavioral therapy, and pharmacological therapy. Although the duration of programs varies, maximal weight loss occurs in the first 6 months of therapy.22,23 The short- and long-term efficacy of weight loss programs is presented in Table 1. With the exception of continuous pharmacological therapy, the pattern of weight regain is evident from 6 months on.
Calorie restriction programs restrict caloric intake by 300-500 kcal/day and fat calories to < 30% of daily caloric consumption. The expected rate of weight loss is 0.5-1.0 lb/week. In a review of 51 diet restriction programs, weight reductions at 6 months averaged 5% of baseline weight.22 By 2 and 3 years, weight reductions averaged 4.4 and 3%, respectively.
Diet and exercise
Diet-only programs have proven more effective than exercise-only programs for inducing clinically significant levels of weight loss.22 Yet,exercise can enhance the amount of weight loss achieved from dieting.22,24 A meta-analysis of 17 trials demonstrates that average weight reductions for combined programs are 8.5% at 6 months and 4% at 4 years.22 Nonetheless, reversion to baseline weight is predicted by 5.5 years; a steady rate of regain is observed from 6 months on.25 The pattern of weight regain is slower in patients without diabetes and in those participating in induction programs that recommend fewer calories per day.
Low-calorie diets (LCDs) restrict caloric intake to 800-1600 kcal/day. A meta-analysis of formalized LCD programs estimates average weight loss at 6 months to be 9.7% of baseline weight.26 By 1 and 2 years, weight reductions average 5%. Altering the macronutrient composition of LCDs to be very-low-fat or low-carbohydrate diets has not enhanced long-term weight loss outcomes.27,28
Meal replacement (MR) programs are often prescribed by physicians, but trials examining their efficacy have only emerged in the past 10 years.29 During weight loss, all or the majority of meals are replaced. In the maintenance phase of care, meal replacements are gradually titrated down and replaced with a sensible diet. Although calories are restricted within the range of LCD programs, MRs are thought to improve diet adherence by removing calorie counting and portion control from the patient's control. In a review of MR programs with a duration of 3 months or longer, weight losses at 6 months and 1 year averaged 9.6 and 7.5% of baseline weight,respectively.30 Two meta-analyses have demonstrated that MR programs are more effective than LCDs for maintaining weight loss at 1 year;22,30 the net difference in weight loss maintained at 1 year was 2.6 kg in favor of MR.30
Very-low-calorie diets (VLCDs) are intensive diet programs that must be supervised by a physician.31 Daily caloric intake is restricted to 400-800 kcal/day, often via a liquid diet for up to 4 months. VLCDs are associated with rapid weight loss—16% weight reductions are commonly observed at 6 months—followed by rapid weight regain.22,26,32 Six trials have directly compared the efficacy of VLCD to LCD programs for weight loss. In these programs, continuous use of an LCD was compared to a sequential regimen of a VLCD followed by an LCD. A meta-analysis of these studies demonstrates that the initial net benefit derived from a VLCD diminishes over time such that group differences were not statistically significant at 2 years.26
Behavioral therapy, in combination with diet and exercise, is highly effective at inducing weight loss.6 The goal of behavioral therapy is to promote weight loss through changes in diet and activity level. Standard behavioral therapy (SBT) programs are generally provided in a group setting, led by a dietitian or health psychologist, and focused on stimulus control, self-monitoring of diet, physical activity and weight, and nutrition counseling, among other topics.23 Average duration of SBT programs is 4-6 months. In a review of nine programs, weight reductions averaged 10% at 6 months and 8% at 18 months.23 Follow-up data are sparse. A separate study found that that at 2 and 4 years, weight reductions approached 3.8 and 1.8 kg,respectively.12
Commercial weight loss programs
Weight Watchers is the only commercial program to report data from randomized controlled trials.33 In a multicenter trial, participants randomly assigned to a Weight Watchers program lost significantly more weight than a self-help control group.34 For the Weight Watchers group, weight reductions of 4.6% were maintained at 1 year,and reductions of 3% were maintained at 2 years. Between-group differences in percentage of weight reduction averaged 3% at both time points. In a second study, individualized support from a dietitian greatly enhanced weight outcomes among Weight Watchers participants; 60% of those receiving dietitian support achieved 10% weight reductions at 1 year.35 Follow-up data are not available.
Weight loss medication
Orlistat and sibutramine are the only two drugs approved by the Food and Drug Administration (FDA) for long-term treatment of obesity.36 A recent trial demonstrated that the efficacy of pharmacological treatment for obesity is enhanced when medication use is combined with a lifestyle program focused on diet, exercise, and behavioral therapy.37 Published meta-analyses, however, pool data from trials that provided adjunctive lifestyle programs and those that did not. In a meta-analysis of 20 trials, pharmacological therapy ± adjunctive support induced an 8% body weight reduction by 6 months.22 Weight loss hit a plateau at 6 months. With continued medication use, however, 7-11%body weight reductions were maintained for up to 3 years. A second meta-analysis demonstrated that pharmacologically treated patients were three times as likely as those solely relying on lifestyle programs to maintain clinically significant weight losses for 1-2 years.24 Weight regain, however, often occurs after medication discontinuation.36
Weight Loss Maintenance Programs
The goal of weight loss maintenance programs is to prevent weight regain after weight loss. Programs can be classified by approach: sequential medication use, sequential dieting, individualized follow-up support, or group-based follow-up support. This section provides an overview of the long-term outcomes of randomized controlled trials of weight regain prevention programs (Table 2).
Sequential medication use appears to be an effective strategy for delaying rapid weight regain. In each of the eight trials identified, maintenance-phase medication use improved end-of-program weight loss maintenance outcomes over placebo.29,38-44 Weight reductions averaged 2-6% greater among those receiving maintenance-phase medication compared to placebo. Sequential medication use induced additional weight loss in three trials.39,40,44 and slowed the pattern of weight regain in the other five trials.29,38,41-43 Neither the type of induction therapy nor the amount of weight initially lost appears to explain this differential effect. Most troubling, however, is that intervention effects appear to diminish rapidly after medication discontinuation,40 and dose reduction does not appear to be an effective alternative.29 Incidentally, addition of a second weight loss medication during the maintenance phase does not appear to produce a synergistic effect.45
Although sequential dieting is common in the literature—low-calorie diets often follow very-low-calorie diets; partial meal replacements often follow meal replacement diets—few trials have accessed the efficacy of sequential dieting for prevention of weight regain. Findings from one trial suggest that optional food provision is not an effective weight regain prevention strategy.12 Three other trials have assessed the relative efficacy of sequential dieting approaches for preventing weight regain. Within the context of group follow-up support, sequential dieting with a low-carbohydrate diet was equally as effective as a low-fat diet;46 an MR was equally effective as orlistat;47 and an MR was equally effective as an LCD48 for supporting weight regain prevention at program completion. In all of these trials, 90-100% of initial weight losses were maintained at program completion.
Given the multi-component design of these trials, it is difficult to assess the relative contribution of sequential dieting to weight regain prevention. Moreover, follow-up data are only available from one trial, and significant weight regain was evident after discontinuation of maintenance support.48
Individual follow-up support
In terms of individual follow-up support, telephone contact has received the most attention in the literature and appears to be effective, especially after group induction therapy and when provided by a therapist. In all identified trials, the trend was for telephone support to improve weight loss maintenance outcomes over no maintenance contact. Group differences only reached statistical significance in three of the seven study groups identified.49,50 In these programs, end-of-program weight reductions averaged 2-7% greater for those receiving telephone support compared to no maintenance care. Telephone maintenance support after relapse prevention training produced the largest benefit over usual maintenance care, and benefits were sustained 6 months after discontinuation of maintenance contact.49
In contrast to traditional behavioral therapy, relapse prevention training(RPT) acknowledges that lapses in self-care behavior are inevitable.51 The focus of RPT is thus to teach anticipatory planning and coping techniques.52 It is hypothesized that individualized follow-up support enhances outcomes by enabling new skill assimilation.49 Telephone support after group-based induction, including behavioral therapy and nonbehavioral therapy, have also enhanced weight loss maintenance outcomes.49,50 Notably, nontherapist contact12 and telephone contact after a telephone-based induction program53,54 or individualized cognitive behavioral therapy55 have not proven effective for supporting weight regain prevention.
In addition to the telephone, other routes of delivery of individualized follow-up support have been explored in the weight regain prevention literature. One trial assessed the efficacy of an onsite stepped-care approach; after a group-based induction program, participants were eligible to receive individualized problem solving training during the maintenance phase of care if they regained > 1% of baseline weight.56 Although the stepped-care program did prevent weight regain per se, among those who regained > 1%, those who received individualized problem-solving training greatly benefited from the stepped-care approach. Other trials found that an interactive website50 and mail-based maintenance support53,54 were no more effective than a self-directed maintenance program or usual care for supporting weight regain prevention.
Group follow-up support
Group follow-up support has been explored after weight loss induction via medication, group behavioral therapy, and very-low-calorie diets. It has additionally been provided to support weight regain prevention among successful weight loss maintainers, that is, individuals that had lost and successfully maintained > 10% weight reductions for at least 1 year. Compared to induction only or usual maintenance care, the trend in all trials was for group follow-up support to enhance weight loss maintenance outcomes at both program completion and thereafter.12,57-65 Group differences reached statistical significance in six of the 11 study groups identified (P > 0.05 for group differences in weight change). In these studies, end-of-program weight reductions averaged 2-7% greater for those receiving group-based maintenance support compared to no support. Peer group meetings with individualized therapist contact60,61 and problem-solving training63 produced the greatest benefit over usual care, although behavioral therapy booster meetings were also effective.12 Programs of comparable duration have yet to determine the relative efficacy of these programs for supporting weight regain prevention.
Maintenance programs based on relapse prevention training and peer group meetings produced a delayed benefit.59.62 Although no effect was demonstrated at program completion, 6-12 months later,weight reductions were significantly greater among those who previously received maintenance support than among those who had not. A weight-focused maintenance program similarly produced a delayed benefit over an exercise-focused maintenance program.66
Internet delivery of group follow-up support has yet to be validated in the literature. In one trial, Internet-based booster meetings delivered via Internet chatrooms appeared to enhance weight loss maintenance outcomes compared to a minimal contact and an onsite booster group.64 Group differences did not reach statistical significance, however. In another trial,group-based Internet support offered to successful weight losers effectively decreased the proportion of participants who regained weight; however, only onsite support proved effective for minimizing the amount of weight regained compared to a newsletter control group (P <0.05).58
The utility of supplemental exercise programming during group-based maintenance support has been assessed in three trials.57,67,68 In each of these trials, induction therapy did not include a specific focus on exercise; rather, exercise was introduced as a key strategy for weight loss maintenance. In addition to group-based maintenance support, participants in these programs were encouraged to meet energy expenditure goals via participation in supervised and independent exercise programs. Exercise expenditure goals were moderate in nature (1,000 kcal/week) and focused on walking, cycling, and/or weight resistance training. Although the trend was for supplemental exercise programming to enhance weight loss maintenance outcomes compared to nonsupplemented maintenance support, group differences did not reach statistical significance in any study.
Given the likelihood of rapid weight regain after medication discontinuation and the desire of many patients to not use weight loss medications indefinitely, there is a need to identify maintenance programs to be used after medication discontinuation. One trial has addressed this question.65 Compared to usual maintenance care, group follow-up support after pharmacological treatment of obesity enhanced weight loss maintenance outcomes at program completion (P < 0.05); weight reductions averaged 3% greater for those who received group follow-up support compared to usual maintenance care.65 Moreover,100% of initial weight losses were maintained by those receiving group-based follow-up support.
The trials identified in this review appear to support the assertion that continued therapy, be it based on maintenance-phase medication, sequential dieting with group follow-up support, individualized telephone support, or group-based follow-up support, is effective for slowing or delaying weight regain. Only one program identified—telephone support after relapse prevention training—supported weight regain prevention during the life of the maintenance program and new weight stabilization thereafter.
The long-term efficacy of other seemingly effective weight regain prevention programs, such as group-based problem solving training or sequential medication, is uncertain, because follow-up data have not been reported. Preliminary evidence, however, suggests that rapid regain occurs after discontinuation of maintenance medication.
In terms of relative efficacy of maintenance approaches, research is limited. One trial demonstrated that within the context of a multi-component program, sequential dieting with an MR was as effective as sequential medication for supporting weight regain prevention. Further research on the relative efficacy of maintenance approaches for supporting weight regain prevention and new weight stabilization are needed.
Factors Underlying Weight Regain and Weight Loss Maintenance
Some individuals are better able than others to maintain weight losses. Factors underlying weight regain and weight loss maintenance have been identified, and the former can be classified along a time axis(Table 3). Factors are clinical, psychological, and behavioral in nature.
Pre-treatment indicators of future weight regain include older age;69 Mexican-American ethnicity;70 frequent previous diet attempts;71,72 high baseline or maximum weight;73-75 nonmedical triggers for weight loss induction;76 binge eating;77 dietary disinhibition (loss of control while eating);77,78 “all or nothing”thinking;79 perceived barriers to exercise, including “too tired,” “too hard,” “not enough time,” and “no companion”;perceived barriers to diet, including “high cost of healthy foods”and “eating away from home too often”;19,20 and lacking self-efficacy, motivation, realistic weight loss goals, and a strong body image.80
Post-weight loss indicators of weight regain include weight loss >15-30% of baseline weight;14,81 early regain;82 not responding to early regain;78 perceptions of hunger;83 dissatisfaction with achieved weight loss;80 dietary disinhibition;81 emotional eating;78 binge eating; consuming a diet high in calories, fats, and sugars;14 frequent consumption of fast food (> 2-3 times per week);20 a sedentary lifestyle or decreased frequency and level of physical activity;14 and viewing more than 2-4 hours of television a day.14 Although somewhat counterintuitive, continuation of weight loss efforts is also associated with weight regain.16,73
Weight loss maintenance
Studies of the NWCR of successful weight loss maintainers provide insight into the transient course of successful weight loss maintenance. Among successful weight loss maintainers, modest weight regain and weight fluctuations are common, and recovery from regain > 2.6 kg is difficult.81,84 During the first years of weight loss maintenance, continued effort and attention to weight control is essential. By 2 years, however, maintainers report less reliance on weight loss strategies and reduced attention and effort to weight control, perhaps reflecting new habit assimilation.85 Furthermore, maintenance of weight loss for ≥ 2 years is protective against subsequent regain; by 2 years the likelihood of regaining 2.6 kg in the coming year is only 50%; by 5 years the likelihood drops to 27%.81 Evidence from the NHANES survey similarly supported the contention that among successful weight loss maintainers, years from maximal weight loss is protective against weight regain.14
Studies of the NWCR identify eight behavioral strategies important for weight loss maintenance.21,77,86 These included maintaining high levels of physical activity and limiting television viewing to less than a few hours a day; eating a diet low in calories and fat; regularly consuming breakfast; maintaining a consistent eating pattern throughout the week and year; reigning in emotional eating;frequently monitoring weight; and catching slips before they turn into large-scale weight gains. A Consumer Reports survey additionally identified direct coping, (i.e. with stress and lapses in behavior) as opposed to avoidance or emotional eating and help-seeking as important behaviors for successful weight loss maintenance.18
The obesity literature has identified numerous weight loss induction programs that are effective at inducing clinically significant levels of weight loss. Diet and exercise, low-calorie diets, meal replacement,behavioral therapy, and pharmacological agents are generally effective at inducing 8-10% weight reductions by 6 months. VLCDs have demonstrated the capacity to induce even larger weight losses. For all programs, maximal weight loss occurs in the first 6 months of therapy. For all nonpharmacological programs, weight regain begins shortly thereafter. On average, weight losses at 2 years range between 3 and 6% for nonpharmacological therapies and between 7and 8% for pharmacological therapies. For individuals at risk, these outcomes are within the range known to induce improvements in cardiovascular risk factors and prevent type 2 diabetes.2,3
Weight loss maintenance programs identified in this review demonstrate that sequential medication use and individual and group follow-up support are effective for improving weight loss maintenance outcomes, predominantly by slowing the pattern of weight regain. Only one program, telephone contact after relapse prevention training, produced clinically significant benefits that were sustained well after discontinuation of maintenance contact. Although the ideal outcome of weight loss maintenance programs would be new weight stabilization, delayed onset of weight regain is a second-best outcome. The cardiovascular and metabolic health benefits associated with weight reduction accrue in a dose-dependent fashion. Noncurative therapies are streamlined into the treatment of hypertension and dyslipidemia; acceptance of continuous models of care for obesity may also be necessary.12
Potential points of care to support weight regain prevention are high-lighted in Table 4. Patients often expect to lose 20-30% of body weight.87-89 On average, patients are only achieving half of this goal and are not sustaining it for much longer than 6 months. The impact of unrealistic weight loss goals on weight loss maintenance is uncertain.80.87.89-92 Satisfaction with weight loss attained, however, predicts weight loss maintenance success.79,92,93 Patient education on the health benefits from modest weight reductions may serve to promote patient satisfaction with weight loss achieved, and in turn,support weight loss maintenance.
Presenting the phenomenon of weight regain as a likely event in the course of weight loss maintenance may also support weight loss maintenance. Those at greater risk of weight regain—those who lost a large amount of weight initially, have a history of dieting, are older, are not exercising, and tend to eat in response to their emotions—should be made aware of their risk profile and encouraged to seek help when behavior reversion or early regains occur. It may be necessary to emphasize that, although they are no longer dieting per se, continued attention to weight, diet, and exercise is essential for successful weight loss maintenance. Given that the risk of regain declines with time, 2-5 years should be presented as a benchmark for successful weight loss maintenance.
In support of patient efforts, physicians should emphasize the importance of weight control, diet, and exercise. The NWCR and the Stop Regain program demonstrate that frequent self-weighing is associated with weight loss maintenance.21,58 Moreover, a decline in the frequency of self-weighing is independently associated with weight regain.94 Self-weighing whether daily or weekly enables pattern recognition, provides an opportunity for positive reinforcement, and is thought to promote self-regulation of behavior. Although some have expressed concerns that self-weighing may negatively affect the psyche, a follow-up study to the Stop Regain program found that frequency of self-weighing was actually associated with a decline in depressive symptoms, increased dietary restraint, and decreased dietary disinhibition.95
A balanced and consistent diet is essential for weight loss maintenance. In support of good habit formation, diets should obtain limited calories from fat and sugar,17,21 and be consistent throughout the day, week, and year.21 Often,successful weight losers report consuming less variety in all food groups with the exception of fruits.96 However,flexible control, as opposed to rigid diet restriction of certain food groups,is recommended as a means to support weight loss maintenance.97 Other dietary strategies used by successful weight loss maintainers include self-recording of intake, planning meals, regulating portion size, consuming more than five servings of fruits and vegetables a day, and limiting intake of fast food.19,20,28
Exercise is central to weight loss maintenance. Correlational studies demonstrate that frequent exercisers are more likely to maintain weight losses. Randomized control trials support this finding as well.98-100 Although the Centers for Disease Control and Prevention recommends 150 minutes/week for weight control.101 studies of successful weight loss maintainers suggest that higher levels of physical activity are necessary for weight regain prevention. The American College of Sports Medicine recommends 60-90 minutes of moderate activity per day.102
Exercise prescriptions do not have to be daunting. The relationship between activity levels and weight loss maintenance appears to be dose-dependent.99,103,104 Randomized trials suggest that total energy expenditure and not intensity or duration of physical activity is most important for weight loss maintenance.104,105 Home-based exercise programs with or without the provision of home exercise equipment and/or focusing on short bouts of exercise are more effective for improving exercise adherence than is supervised group exercise, providing personal trainers, or emphasizing long bouts of exercise.12,106 Moreover, brisk walking is the most common form of exercise reported by successful weight loss maintainers in the NWCR,21,107 and research shows that an increase in leisure activity and steps per day are associated with weight loss maintenance.19,108,109 Moderate physical activity achieved through lifestyle activity, intermittent bouts of exercise, and increased steps taken per day are easy recommendations for exercise adherence and weight loss maintenance.105
Much attention in the literature has been given to identifying effective treatments for obesity. The strength of the present review is its focus on weight regain prevention and analysis of data from randomized controlled trials. It is limited by the fact that the review was nonexhaustive and based on reported and calculated data, including the use of arithmetic mean group differences to define net effect. Nonetheless, the review provides an important overview of the weight regain phenomenon and identifies translatable points of care for primary care physicians.
Long-term maintenance of weight loss is an important, but often elusive,goal.
Evidence from this review and others cited in the literature find that continued therapy from a health professional is essential for weight loss maintenance success. In support of weight loss maintenance, physician-patient interactions should include a discussion of patient goals, the weight regain phenomenon, decision points in self-care and help-seeking behavior, and conclude with advice about self-monitoring, diet, and exercise.
Christina Garcia Ulen is third-year student at the University of Michigan School of Medicine. Mary Margaret Huizinga, MD, MPH, is an assistant professor at Johns Hopkins University School of Medicine. Bettina Beech, DrPH,is an associate professor at the Diabetes Research and Training Center at Vanderbilt University Medical Center. Tom A. Elasy, MD, MPH, is director of the Prevention and Control Core of the Vanderbilt Diabetes Research and Training Center at Vanderbilt University Medical Center and medical director of the Vanderbilt Eskind Diabetes Center. He is also editor-in-chief ofClinical Diabetes.