Obesity rates have reached epidemic proportions, and are continuing to escalate worldwide (1). In the United States, 68% of adults are overweight; 33.8% are obese (2). Obesity is not a cosmetic issue. It is an important health problem that opens the door to a myriad of diseases, primarily diabetes and cardiovascular disease (CVD), leading to significant reductions in quality of life and life expectancy. In 2008, the annual total healthcare costs attributable to obesity were approximately $147 billion (3). Body mass index (BMI), calculated as weight in kilograms divided by height in meters squared (kg/m²), is the most common measure used to estimate body fat. The World Health Organization (WHO (4); ) and the National Institutes of Health/National Heart, Lung, and Blood Institute (NIH/NHLBI (5); ) have defined obesity as a BMI ≥ 30 kg/m², and overweight as a BMI of 25.0 to 29.9 kg/m². There is a strong positive relationship between BMI and mortality. The risk of mortality increases by 30% for individuals with a BMI ≥ 30 kg/m² and by 100% or more for those with a BMI ≥40 kg/m². This excess mortality is primarily due to the strong association between abdominal or visceral obesity and CVD and type 2 diabetes (6). Obesity is rapidly approaching smoking as the leading cause of preventable deaths in the United States (7), and if this trend continues, obesity may even decrease life expectancy (8).

Body weight is regulated by a delicate balance between energy intake and energy expenditure. Over the last half century, human behavior has adapted to a lifestyle that favors less physical activity and greater consumption of calorie dense, manufactured foods. Therefore, lifestyle change interventions that aim to change eating and activity patterns are one of the most effective treatments for obesity (9, 10). These interventions are commonly based on cognitive behavioral therapy principles and learning theory (11). Lifestyle interventions are shown to prevent the development of health issues (e.g. diabetes) with modest weight loss and provide lasting change (9).

Individuals with a diagnosis of mental illness have been disproportionately affected by the obesity epidemic. Many psychiatric medications–specifically antipsychotics–increase appetite and food intake (12). This is especially true of individuals with severe mental illness (SMI) who, due to downward social drift and poverty, have difficulty finding affordable options for food and physical activity. Individuals with SMI die 25 years earlier than the general population, mostly due to obesity related conditions such as diabetes and CVD (13). Obesity also has psychiatric and social consequences including depression, increased stigma, and decreased social functioning (14, 15). It is imperative that the question of obesity be addressed during psychiatric care. However, cognitive impairment, psychiatric symptoms, and motivation issues, in combination with the obesogenic psychotropics these patients are prescribed,(i.e. antipsychotics, mood stabilizers, and antidepressants) raises the issue of whether lifestyle interventions would work in these patients. In this paper, we review the available evidence on the efficacy of behavioral treatments for obesity in SMI populations. Furthermore, several of these lifestyle intervention concepts can easily be incorporated by individual practitioners in psychiatric practice. Implementation of these techniques could be helpful in controlling weight and, ideally, improving patients’ self confidence, health, and quality of life.

Several studies have addressed the prevalence of obesity in patients with SMI such as schizophrenia or bipolar disorder. Allison et al (16) found that 42% of a group of individuals with schizophrenia had a BMI of 27 or greater, compared with 27% of the general population. Keck et al. (17) reviewed 45 studies of patients with bipolar disorder and found that the overall prevalence of overweight and obesity was higher in these patients than in control populations. Excessive carbohydrate consumption, low rates of exercise, and treatment with medications associated with weight gain were identified as prominent risk factors for obesity in these patients. In the 1460 patient CATIE study, the prevalence of metabolic syndrome among patients with chronic schizophrenia was 36% for males and 52% for females; in the general population, the rates are 20% for males and 25% for females (18). In the CATIE study, obesity was associated with a 25% increase in outpatient medical costs, even after controlling for demographic characteristics and medical comorbidity (19). When we reviewed 100 consecutive charts in the Connecticut Mental Health Center Psychosis Program, we found that 78% were overweight and 59% were obese, whereas 58.2% of adults without mental illness in Connecticut are overweight and only 20.1% are obese. The relative importance of various risk factors for obesity in psychiatric patients like sedentary lifestyle, low levels of physical activity, low socioeconomic status, and genetic polymorphisms, have not been adequately studied. However, a vast body of literature has emphasized the impact of antipsychotic medications.

A comprehensive review of psychotropic induced weight gain and their metabolic effects is beyond the scope of this article. Antipsychotic medications have been associated with weight gain since their introduction in the 1950s (20). The newer atypical antipsychotics (i.e. clozapine, olanzapine, risperidone, ziprasidone, quetiapine, and aripiprazole) have become agents of choice because of their favorable extrapyramidal side effect profile. However, over the last decade, atypical antipsychotic-induced weight gain has become a major health concern in patients with schizophrenia or schizoaffective disorder. In a landmark study, Allison et al. (20) reviewed available research data and found that after ten weeks of treatment, patients taking clozapine had a mean weight gain of 3.99 kg, followed by a mean weight gain of 3.51 kg for olanzapine, 2.10 kg for chlorpromazine, 2.00 kg for risperidone, 0.48 kg for haloperidol, 0.43 kg for fluphenazine, and 0.04 kg for ziprasidone treated patients. The mean weight gain for aripiprazole was 0.71 kg; the mean weight gain for quetiapine was 2.00 kg (21). Even medications with more favorable weight gain profiles, such as ziprasidone and aripiprazole, can still produce significant weight gain (22). Also, first episode patients appear to be more susceptible to weight gain, even with lower weight liability antipsychotic agents (23). In a large first episode study, patients gained an average of 16 kg on olanzapine, and 7.5 kg on haloperidol; with BMI increases of 4.7 and 2.7 kg/m², respectively, over a two year period (24).

Psychotropic medications such as mood stabilizers, valproate derivatives, and lithium; as well as antidepressants, like mirtazapine; often have overlooked weight gain liabilities (25). Since many of these medications are combined in daily practice, the weight gain effects may be additive; however, this is not well studied. Since individual effects vary, any significant weight gain in the month following the start of any psychotropic should be considered psychotropic-induced. Our clinic defines significant weight gain very conservatively as 3% to 4% of previous total body weight. It should be noted that weight gain trajectories for any psychotropic have not been well studied. Since it is much more difficult to lose weight gained after the fact, the focus should be on early intervention and prevention of weight gain.

The goal for risk reduction is a loss of 5% to 10% of initial weight. Patients may not ever reach their ideal body weight; however, a weight loss of 5% to 10% has been shown to be sufficient for improving health complications from obesity, including type 2 diabetes, hypertension, and hypercholesterolemia (26, 27). There is also evidence to suggest that intentional weight loss is associated with reduced mortality (28). Moderate caloric restriction remains the principle method for achieving weight loss in overweight and obese individuals. Several approaches have been used to promote weight loss: behavioral therapy, self-help books, commercial and self-help programs, internet-based programs, pharmacotherapy, and surgery. It is generally accepted that lifestyle change is necessary no matter which approach for weight control is used. “Behavioral packages” of lifestyle modification incorporate components of behavioral therapy, cognitive behavioral therapy, caloric reduction through changes in diet, and increased energy expenditure through changes in lifestyle that enhance physical activity (9, 11, 29). The terms lifestyle modification, behavioral treatment, and behavioral weight control are often used interchangeably, but refer to similar programs (9).

Behavioral interventions have become the cornerstone for many approaches to obesity treatment. Lifestyle modification programs are typically heavily influenced by learning theory: applying principles of classical and operant conditioning (29). They manage to remain pragmatic by often incorporating strategies like cognitive restructuring, to induce behavioral change (9, 11, 30). The key elements of behavioral approaches are: giving participants a set of structured methods for gradual lifestyle change in order to modify diet and physical activity, utilizing cognitive techniques to change attitude, and implementing strategies that increase social support. Thus, these interventions attempt to both provide self-knowledge of where problem behaviors originate and teach the skills for changing them (30, 31). In our clinical opinion, self-monitoring may be one of the most important techniques taught (32, 33). Since people tend to underestimate energy intake and overestimate energy expenditure, participants are taught to use measurement tools (e.g. cups, spoons) and read food labels to improve diet, and how to use pedometers or other similar instruments to monitor physical activity. Participants are also instructed to keep records of calorie intake throughout the day including a record of what they eat, how much they eat, and when they eat it. Antecedents of eating can be recorded by rating hunger and emotions in order to identify targets for change like nonhunger related eating. Similarly, physical activity is recorded daily, and pedometers and actometers can be used to provide feedback and encouragement (9).Goal setting is a significant component as well, since people tend to have unrealistic expectations of weight management interventions or set vague goals (34). Modest weight loss through gradual change is emphasized by setting short term goals to change specific problem behaviors in a realistic, but moderately challenging manner, which increases both the effort and sense of accomplishment when goals are achieved. Stimulus control is based on both classical and operant conditioning principles. Associations between non-food cues and eating behaviors need to be broken (e.g. watching movies and eating popcorn). Participants may modify their environment to reduce the availability of unhealthy behavioral cues and to increase healthy behavioral cues. Although weight loss itself is a reward, to reinforce positive behavior, other rewards can be used by the program or participant when behavioral changes are made. Behavioral substitution attempts to control the emotional cues that cannot be altered through environmental modification by replacing problem behaviors with other activities that are not conducive to eating. Problem solving is one of the cognitive skills emphasized in programs. This skill teaches people to identify problem behaviors and their antecedent events, brainstorm potential solutions, choose the solution with the most favorable risk/benefit profile, and then finally apply that solution. Cognitive restructuring works by identifying negative or distorted thoughts that impede behavioral change and then encouraging replacement of these thoughts with opposite alternatives in order to change behavioral outcomes. Relapse management encourages participants to plan in advance for lapses and then prevent them from occurring by managing cravings and avoiding high risk situations. These major techniques are used by most lifestyle modification programs, along with other strategies, such as eating slowly, controlling portions, and lifestyle activity (incorporating short bouts of physical activity into daily life, e.g. avoiding elevators, use of public transportation, parking farther away from the store) (9, 11, 30, 31, 35, 36).

Intervention characteristics

Several diagnoses were used as inclusion criteria for these studies: schizophrenia (37–53), schizoaffective disorder (38–47, 50, 52), or SMI and taking an atypical antipsychotic medication (54–61). One study (53) limited participants to postmenopausal women and one study (44) only recruited participants with type 2 diabetes. The weighted mean BMI for the intervention group at baseline was 31.3; it was 28.5 for the control group.

Intervention personnel included clinical psychologists, psychiatrists, clinicians, nurses or nurse practitioners, nutritionists or dieticians, certified diabetes or health educators, and exercise physiologists.

Dietary strategies

Interventions utilized several educational components and strategies to improve nutrition and facilitate weight loss. These components included keeping a food diary, prescribing an individualized calorie restricted diet, providing healthy eating and dietary guideline education, teaching label reading and appropriate serving sizes for portion control, meal planning, healthy food shopping and preparation on a budget, and how to distinguish between “head” hunger and physical hunger.

Exercise strategies

Interventions primarily focused on increasing lifestyle physical activity and walking, while decreasing time spent in sedentary activities. Participants were encouraged to monitor their physical activity by using pedometers and exercise diaries. Several interventions educated participants on the benefits of regular exercise and modeled low cost exercises that participants could perform at home. Participants also received guidance on how to initiate and maintain a physical activity plan that included choosing appropriate exercises for all fitness levels.

Behavioral strategies

Behavioral modification strategies included goal setting, problem solving, social support, motivational counseling, stress management, stimulus control, behavior modeling and skills training, rewards and reinforcements, assertiveness training, and relapse prevention. Group sessions utilized visual aids, handouts, games, mnemonic devices, and repetition to help participants understand educational concepts.

Prevention of antipsychotic-induced weight gain

Four randomized controlled trials (RCT) (38, 45, 54, 56) assessed the efficacy of behavioral interventions on preventing weight gain for individuals starting an atypical antipsychotic medication. Interventions ranged in duration from 12 to 24 weeks and subjects were either switched to olanzapine before the start of the study (38, 45, 54)or were first episode psychosis patients taking an atypical antipsychotic medication (56). Three studies were conducted on an individual basis (45, 54, 56) whereas the other study (38) utilized group sessions. Scocco et al. (45) investigated a psycho-educational approach to prevent weight gain over 24 weeks; however, only the first eight weeks involved a treatment as usual comparison and these are the results presented in Table 1. All studies provided a nutritional education component, along with integrating lifestyle modification and exercise. Participants in the intervention groups gained significantly less weight (mean = 2 kg) than participants in the control groups (mean = 6.7 kg). These results demonstrate that behavioral interventions can attenuate weight gain upon initiation of an atypical antipsychotic medication.

Treatment of antipsychotic-induced weight gain

Twenty-one prospective studies, twelve non-RCTs (37, 39, 40, 42, 49, 50, 52, 53, 55, 57, 60, 61) and nine RCTs (41, 43, 44, 46–48, 51, 58, 59), evaluated the efficacy of behavioral interventions on weight loss for individuals receiving atypical antipsychotic therapy. Interventions ranged in duration from eight weeks to six months. Eight interventions (41, 42, 44, 46, 50, 52, 55, 58) were group based and five interventions (43, 49, 53, 57, 60) were provided on an individual basis, whereas eight studies (37, 39, 40, 47, 48, 51, 59, 61) utilized a mixture of individual and group-based instruction. Wu et al. (51) evaluated the effectiveness of lifestyle intervention and metformin for the treatment of antipsychotic-induced weight gain, a comparison of the lifestyle only and control groups are the results presented in Table 1.

BMI and weight gain inclusion criteria differed across studies. Ten studies (37, 39, 40, 42, 43, 51, 57–59, 61) included participants with significant atypical antipsychotic-induced weight gain, five studies (41, 46, 48, 50, 52) included overweight individuals, four studies (47, 53, 55, 60) included obese individuals, and two studies (44) (49) included individuals who were taking atypical antipsychotic medications without specific weight or weight gain criteria.

For non-RCT studies, 358 individuals completed the active interventions and 86 individuals completed the control interventions. Participants in the active groups lost an average of 2.5 kg; participants in the control groups gained an average of 2 kg. Seven studies (42, 52, 53, 55, 57, 60, 61) found significant post study mean weight loss within the active groups, three studies (39, 40, 49) found significant differences in mean weight loss between active and control groups, and two studies (37, 50) did not find differences in mean weight loss either between active and control groups or pre-post within the active group.

For RCT studies, 214 individuals completed the active interventions and 196 individuals completed the control interventions. Participants in the active groups lost an average of 2.9 kg; individuals in the control groups lost an average of 0.7 kg. Six studies (43, 44, 47, 48, 51, 59) found significant differences in mean weight loss between the active and control groups, while three studies (41, 46, 58) found no difference in mean weight loss between groups.

Achieved weight loss differed depending on the format of the intervention. Group based interventions produced a mean weight loss difference of 2.8 kg, individual based interventions produced a mean weight loss difference of 3.5 kg, and interventions that combined individual and group instruction produced a mean weight loss difference of 4.5 kg. These results demonstrate that individual or group lifestyle interventions are more effective in reducing or attenuating antipsychotic-induced weight gain than usual care and suggest that a combination of individual and group instruction may provide enhanced results.

In our clinic, the first rule of weight control is rational use of psychotropics at the beginning of treatment. Weight producing medications should be avoided if possible, and weight change after starting any medication should be monitored. For obese patients or those who experienced significant weight gain, in addition to switching medications, there are several simple changes that can be addressed during routine office visits. Every psychiatrist office should have a scale. Electronic office scales are accurate and inexpensive; many can also calculate BMI. Clinicians should weigh patients often and examine weight gain trajectory closely. This also provides an opportunity to discuss weight issues. It is important for patients to learn self-monitoring techniques in order to be successful. This includes monitoring weight and recording food intake. Keeping a food diary can help patients increase their awareness of what and how much they are eating, and identify patterns of consumption that can be targeted for change. Every visit can include brief wellness education; the consequences of being overweight, using moderate caloric restriction rather than intensive dieting to control weight, prevention of emotional or nonhunger eating, utilizing portion control, replacing regular soda consumption with diet to limit unnecessary calories, and stimulus reduction by decreasing the availability of snacks at home. Increasing levels of physical activity is crucial for managing weight gain. The minimum amount of moderate physical activity necessary is 30 minutes daily (62). These 30 minutes can be accumulated over the course of the day and do not need to be done consecutively. There are several ways patients can achieve their 30 minutes without joining a gym. Lifestyle physical activities, like taking the stairs instead of the elevator or getting on and off the bus one stop farther away, provide no cost opportunities for exercise. Additionally, time spent on sedentary activities like watching television must be limited.

Psychiatric patients suffer from obesity more often than the general population. Since psychotropic medications play an important role in overweight and obesity among the mentally ill, weight issues need to be addressed during psychiatric care. Behavioral strategies for weight control are effective in this population, and these strategies can be incorporated into routine office visits without consuming too much time.