May 2011  << Back  

  Can't open the PDF? Click here for help.



Obesity in adults with serious and persistent mental illness: A review of postulated mechanisms and current interventions

James L. Megna, MD, PhD

Department of Psychiatry, SUNY Upstate Medical University, Syracuse, NY, USA

Thomas L. Schwartz, MD

Department of Psychiatry, SUNY Upstate Medical University, Syracuse, NY, USA

Umar A. Siddiqui, MD

Department of Psychiatry, SUNY Upstate Medical University, Syracuse, NY, USA

Mariela Herrera Rojas, MD

Department of Psychiatry, SUNY Upstate Medical University, Syracuse, NY, USA

BACKGROUND: An epidemic of overweight and obesity in the United States has had profound effects on the health of the general population, with consequent development of metabolic syndrome and related morbidity and mortality. However, these effects have been more widespread among adults with serious and persistent mental illness.

METHODS: A literature search was conducted using the PubMed and Ovid databases. Terms used, in varying combinations, were schizophrenia, schizoaffective disorder, bipolar disorder, obesity, atypical antipsychotic, diabetes mellitus, hyperlipidemia, and metabolic syndrome. Of 103 articles generated, 71 were deemed pertinent to the current study. One reference was decided upon based on personal communication.

RESULTS: Both nonpharmacologic and pharmacologic factors contribute to obesity development in adults with serious and persistent mental illness. Consequently, similarly targeted nonpharmacologic and pharmacologic interventions have been used to mitigate against body weight gain. Although the results obtained thus far are promising, effect sizes only in the low to medium range have been realized, with nonpharmacologic interventions demonstrating slight superiority.

CONCLUSIONS: Improved therapeutic methods are needed to address the effects of obesity on individuals with serious and persistent mental illness. Factors that will likely contribute to such advancement are a better understanding of the mechanisms involved, earlier intervention, and adequately powered, randomized controlled trials of sufficient duration, with baseline body weight as a covariate.

KEYWORDS: obesity, serious and persistent mental illness, morbidity and mortality, nonpharmacologic and pharmacologic mechanisms/interventions



There is an epidemic of overweight and obesity in the United States. Results from the 2005-2006 National Health and Nutrition Examination Survey (NHANES) reveal that >72% of adults are either overweight (body mass index [BMI] ≥25 to <30; 32.7%), obese (BMI ≥30; 34.3%), or extremely obese (BMI ≥40; 5.9%).1 However, among adults with serious and persistent mental illness (SPMI) the prevalence of obesity has been reported to be as high as 55%.2 The enormity of this problem cannot be overstated, as there is evidence that it has contributed to a widening of the mortality gap associated with SPMI compared with that of the general population.3 Supporting data include well-established BMI-related health risks of high, very high, and extremely high for BMIs of 30.0 to 34.9, 35.0 to 39.9, and >40.5, respectively.4 Accordingly, there is a growing consensus of concern about obesity-related medical illness in individuals with SPMI.5-7

Obesity is a major factor in the development of numerous medical problems. Increased body weight carries an increased risk for the development of type 2 diabetes mellitus,8 dyslipidemias,9 metabolic syndrome,10 and hypertension.8 The prevalence of these problems is particularly high in the SPMI group, with reported rates approaching 15% for type 2 diabetes mellitus, 69% for dyslipidemias, 63% for metabolic syndrome, and 58% for hypertension.2 These factors contribute to the nearly doubled risk of dying from cardiovascular disease among patients with SPMI.2

Given that some individuals with schizophrenia spectrum illnesses develop type 2 diabetes mellitus and metabolic syndrome without significant weight gain,11-14 obesity is not the only factor contributing to morbidity and mortality in this group; however, it does play a central role. In this article, the authors review postulated mechanisms of obesity development and current interventions for weight reduction among individuals with SPMI.

Mechanisms of weight gain

Mechanisms that account for weight gain among patients with SPMI can be classified as nonpharmacologic and pharmacologic. Nonpharmacologic mechanisms include factors related to the patient’s psychiatric disease and lifestyle, as well as other biologic and nonbiologic factors. Among pharmacologic mechanisms, psychotropic medications interact with nonpharmacologic factors, but they can also directly induce weight gain and obesity.

Nonpharmacologic. In bipolar disorder, nonpharmacologic variables that influence the onset and maintenance of obesity include atypical depression, eating habits and behavior, physical inactivity, and lowering of the basal metabolic rate during depression, with a reduction in spent energy.15

Patients with schizophrenia typically have a sedentary lifestyle and consume an unhealthy diet.16 It has been noted, as well, that patients with schizophrenia often have limited access to foods lower in calories and higher in nutrients.17 Also, these individuals tend to have greater visceral adiposity than healthy individuals.18

Interesting studies have emerged from cohorts of neuroleptic-naïve, first-episode schizophrenia patients (TABLE 1).19-23 The findings suggest that metabolic abnormalities, such as insulin resistance, are present in patients with SPMI even before they begin treatment with antipsychotic medications.20-23 This factor (baseline insulin resistance) likely contributes to the development of type 2 diabetes mellitus and metabolic syndrome in general, as well as when these abnormalities evolve without significant weight gain.11-14

With respect to lifestyle factors, individuals with SPMI are more likely to smoke, and are less likely to engage in exercise.16 Cigarette smoking increases insulin resistance, which in turn decreases glucose tolerance and predisposes to future hyperinsulinemia.24 A survey of outpatients with schizophrenia and affective disorders found that individuals with SPMI are less physically active overall than individuals in the general population (49% vs 22% in the 2005-2006 NHANES sample). Factors associated with inactivity in the SPMI population surveyed included lack of social contacts in the past month and female sex.25

A number of other nonbiologic factors may be considered factors in weight gain. Inpatients with chronic psychosis who voluntarily completed a questionnaire perceived their obesity status accurately. However, their level of concern about their weight didn’t correlate with their BMI, and the obese patients were reluctant to choose mandatory dieting as a remedy.26 Other important nonbiologic factors included age and sex,27-29 lower baseline BMI,30 lower body weight before initiation of drug treatment,27,28,31,32 rapid initial weight gain after initiating drug treatment,31 decreased cigarette smoking activity,32 and socioeconomic status.27

The most robust biologic factors that contribute to weight gain include serotonin (5-HT2C) and histamine (H1) receptors.27 The interaction between dopamine and serotonin/histamine neurotransmission may affect the regulation of appetite. Dopamine is involved in the primary reinforcing characteristics of natural stimuli such as food intake.16 Antagonistic action in the serotonin system increases appetite,16,27,33 as does histamine antagonism.27,33 The latter also has sedative effects, leading some investigators to speculate that sedation may induce weight gain because of reduced mobility,34 making physical activity problematic.16 Also important is the cholinergic system, ie, the anticholinergic effect of dryness of the mouth stimulates liquid intake, which may cause greater consumption of caloric drinks.15

Pharmacologic. Atypical antipsychotics have been associated with a higher prevalence of obesity.35 Clinical trials indicate that mean weight gain and the incidence of clinically significant weight gain vary by antipsychotic agent, with greater than placebo-level effects on weight observed for all FDA-approved antipsychotic agents in the United States.17

In short-term studies, a definite rank order of weight-gain potential among atypical antipsychotics has been demonstrated: clozapine has the highest risk of weight gain, followed, in decreasing order of magnitude, by olanzapine, quetiapine, risperidone, amisulpride, aripiprazole, and ziprasidone. However, in long-term studies, except for clozapine at one end of the spectrum and ziprasidone at the other, the differences in weight-gain liability displayed by the other atypical antipsychotics are diminished. These findings suggest that differences between short-term and long-term treatment could be due to a complex overlapping of various factors, both drug-specific (eg, relative receptor affinity, timing of weight-change plateau, and drug-specific/dose-dependent weight gain) and patient-specific (eg, genetic vulnerability, sex, age, BMI, weight before starting antipsychotic treatment, type of psychiatric disorder, and individual lifestyle).36

In a study of weight change and its correlates during the treatment of acute mania in a closed-ward hospital setting, patients receiving any kind of atypical antipsychotic showed greater weight gain than those receiving first-generation antipsychotics or those not receiving any antipsychotics.37 Similarly, a study of inpatients during an acute hospitalization found that those receiving atypical antipsychotics gained more weight.32

There is ample evidence that binding to certain neurotransmitter receptors may at least partly explain antipsychotic-associated weight gain.34 Clozapine and olanzapine have the greatest affinity for 5-HT2C and H1 receptors. Accordingly, the combined blockade of H1 and 5-HT2C receptors has been especially associated with weight gain, sometimes profound.27,33 Risperidone has lesser yet substantial 5-HT2C antagonism activity with low affinity for the H1 receptor38 and, thereby (presumably), produces less weight gain. Quetiapine, however, has high affinity for H1 receptors but is a weak antagonist of the 5-HT2C receptor,38 which places it at about the same level of associated weight gain as that of risperidone—at least in short-term studies. In long-term studies, quetiapine appears to produce more weight gain,36 yet lower than that of clozapine and olanzapine. Ziprasidone is a potent agonist of 5-HT1A receptors and a potent 5-HT2C receptor antagonist. It is also a synaptic reuptake inhibitor of serotonin27 with low affinity for the histamine reuptor.38 Aripiprazole has moderate affinity for the histamine receptor, no antagonism at the 5-HT2C receptor, antagonism at the 5-HT2A receptor, and partial 5-HTlA agonism.27

The mixed receptor activity profile of these agents likely contributes to their lowered weight-gain potential. In total, then, knowledge of the variable receptor activity profiles can be applied clinically. For example, if a patient with SPMI has a personal and/or first-degree family history of obesity, diabetes, hyperlipidemia, or hypertension, selection of an agent associated with less weight gain (eg, ziprasidone, aripiprazole) would be prudent. However, if the patient has no such history, the choice of antipsychotic agent can be made from among a greater number of options (influenced, of course, by clinical need). Nevertheless, if >7% of baseline body weight is gained, switching to a drug associated with lower weight gain must be seriously considered.

The high interindividual variability in antipsychotic-induced metabolic abnormalities suggests that genetic makeup is a possible modulating factor. Of particular interest has been the serotonin 2C receptor, encoded by the HTR2C gene. Antipsychotics that have been associated with the largest risk of weight gain, lipid abnormalities, and glucose intolerance have a relatively high affinity for the 5-HT2C receptor compared with other antipsychotics. Furthermore, studies in humans have shown a genetic association of the serotonin 2C receptor, encoded by the HTR2C gene, with obesity and some eating disorders.39 Although several genes have been identified, the 5-HT2C receptor polymorphisms have shown the strongest association with antipsychotic-induced weight gain.15,27,40 A recent study showed that both HTR2C and HTR2A gene polymorphisms seem to be associated with the occurrence of metabolic abnormalities in patients treated with olanzapine or clozapine.41

It has been shown that the–759C/T polymorphism of the 5-HT2C receptor regulatory region affects treatment-induced weight gain in patients with first-episode schizophrenia.42 This–759C/T polymorphism of the 5-HT2C gene has also been identified as an important genetic factor associated with clozapine-induced weight gains in the same patient population.43 The 5-HT2C-receptor promoter (–759C/T) is identified as a true genetic susceptibility factor for antipsychotic-induced weight gain.27 In patients diagnosed primarily with schizophrenia or schizoaffective disorder who are treated with antipsychotics, a combined genotype carrying the variant HTR2C:c.1-142948(GT)n 13 repeat allele, the common allele rs3813929 C (also known as–759C/T), the variant allele rs518147 C (also known as–697G/C), and the variant allele rs1414334 C was significantly related to an increased risk of obesity.44

It is hypothesized that 5-HT2C polymorphism status determines levels of circulating leptin27(see below), providing a potential mechanism underlying the genetic association of the 5-HT2C receptor with weight gain.45

A study of the effects of multiple candidate genes on body weight in schizophrenia patients who received risperidone found that of 15 genetic polymorphisms examined, 5-HT2A 102-T/C; 5-HT2C–759-C/T; 5-HT6 267-C/T; BDNF 66-Val/Met; and CYP2D6 188-C/T significantly influenced body weight.28 However, current pharmacogenomic approaches have estimated the number of candidate genes for antipsychotic-induced weight gain to be around 300, a partial listing of which is shown in TABLE 2.27

The role of neuroendocrine factors such as leptin, ghrelin, orexins, and prolactin is unclear, and further research is required to elucidate the mechanisms involved.27 Leptin is a hormone produced by fat cells that binds to the ventral medial nucleus of the hypothalamus (the satiety center), thereby decreasing food intake and increasing energy expenditure.27,46 Serum leptin levels are proportional to adipose tissue mass.47 In a study, leptin levels did not differ between patients with chronic schizophrenia and controls, but levels in patients with schizophrenia correlated with weight gain.46 Among atypical antipsychotics, olanzapine induced a greater increase in leptin levels, independent of weight gain.46 Clozapine and olanzapine appear to upregulate serum leptin levels, but the elevated leptin levels apparently do not reduce food consumption.47 This is likely to occur because of drug blockade at other receptors, such as 5-HT2C, that interacts with the effects of leptin in the hypothalamus,40,48 and with neural processing of leptin-mediated signaling of adipocytes.47 H1-receptor antagonism may also contribute to the “leptin resistance” associated with obesity.49 It is likely that leptin plays a perpetuating role in antipsychotic-induced weight gain, as it also has a role in regulation of insulin sensitivity.27

Ghrelin is a newly-discovered, gut-derived orexigenic peptide with a role in regulation of eating behavior.27,50 Atypical antipsychotics promote the orexigenic effects of ghrelin and at the same time render anorexigenic pathways ineffective.27 Fasting morning ghrelin levels decrease during the first few weeks after initiating antipsychotic treatment but increase with longer-term exposure.48 Conflicting results have been obtained, however.27

Orexins are released by neurons of the lateral hypothalamic/perifornical area and are involved in body-weight regulation and wakefulness. There aren’t many studies evaluating the activating effect of antipsychotics on orexins.27 Hyperprolactinemia can impair insulin sensitivity15,27,51 and affect circulating levels of gonadal hormones by altering the ratio of androgen to estrogen in the body,51 leading to increased adiposity.15,27 Risperidone is known to substantially increase prolactin levels.16

Lastly, weight gain can occur because of indirect effects on glucose and lipid levels. Olanzapine and clozapine have the greatest impact in increasing blood glucose and lipids52 independent of adiposity.27 However, a complex relationship exists among adiposity, blood glucose, and serum lipid levels in relation to antipsychotic treatment.27


Results from studies in neuroleptic-naïve patients with severe and persistent mental illness

Authors Population n Design Variable Results Conclusions
Cohn et al, 200620

Drug-naïve schizophrenia patients

Drug-free schizophrenia patients



9 matched controls

3-hour FSIGTT

49.8% lower MIS in drug-naïve patients

42% lower MIS in all patients

Evidence of BIR and susceptibility to type 2 DM

Spelman et al, 200721

Drug-naïve, first-episode schizophrenia patients

First-degree unaffected relatives



38 matched controls


Fasting plasma glucose, insulin, and cortisol levels


IGT: 10% in patients, 18.2% in relatives, and 0% in control subjects

Higher levels of insulin at baseline and at 2 hours, and presence of insulin resistance in patients and relatives

Possible shared environmental and/or genetic predisposition to IGT and to develop type 2 DM

Venkatasubramanian et al, 200722

Neuroleptic-naïve schizophrenia patients


44 matched controls

Fasting plasma glucose, insulin, IGF-1, and cortisol levels


Higher mean plasma insulin and cortisol levels

Lower IGF-1 in patients

Higher IRS in patients

Deficient IGF-1 might underlie BIR in schizophrenia

van Nimwegen et al, 200823

First-episode, neuroleptic-naïve schizophrenia patients


7 matched controls



No significant difference in IMPGU

Higher EGP in patients

Direct link between schizophrenia and hepatic insulin resistance

BIR: baseline insulin resistance; DM: diabetes mellitus; EGP: endogenous glucose production; FPI: fasting plasma insulin; FSIGTT: frequently sampled intravenous glucose tolerance test; IGF-1: insulin-like growth factor-1; IGT: impaired glucose tolerance; IMPGU: insulin-mediated peripheral glucose uptake; IRS: insulin resistance score; MIS: mean insulin sensitivity; OGTT: oral glucose tolerance test.


Candidate genes for antipsychotic-induced weight gain

Genes coding for substances with hypothalamic-related regulation of food intake Genes coding for substances with adiposity-related regulation of food intake

Pro-melanin-concentrating hormone gene

Adrenergic alpha-1A, alpha-2A, beta-2, and beta-3 receptor genes

Cannabinoid receptor 1 gene

Neuropeptide Y gene

Neuropeptide Y receptor Y5 gene

Ghrelin precursor gene

Galanin gene

Histamine receptor gene

Apolipoproteins A1, A2, A4, B, and E genes

Paraoxonase 1 gene

Scavenger receptor class B1 and B2 genes

Steroid regulatory element-binding protein F1 and F2 genes

Insulin-induced gene 1 and 2

Proopiomelanocortin gene

Adiponectin gene


The management of overweight and obesity in patients with SPMI has included nonpharmacologic interventions, such as monitoring, dietary suggestions, and behavioral strategies, and pharmacologic interventions, which are primarily directed at the metabolic/weight gain effects of antipsychotic medications.53

Nonpharmacologic interventions. The metabolic/weight gain effects of antipsychotic medication should be of concern when planning the treatment strategy for any patient with SPMI. Baseline screening and regular follow-up monitoring, with intervals titrated for the individual patient’s predisposition, are advised.18

In addressing this topic in early 2004, a consensus development conference convened by 4 key organizations—the American Diabetes Association (ADA), American Psychiatric Association, American Association of Clinical Endocrinologists, and North American Association for the Study of Obesity—recommended that all patients receiving second-generation antipsychotics, regardless of diagnosis, “should receive appropriate baseline screening and ongoing monitoring.”5

The response to the consensus guidelines was subsequently assessed. A historical prospective study was conducted using data for 2000 to 2006 from the PharMetrics database of patients prescribed a second-generation antipsychotic. The rates of lipid and glucose testing showed small, statistically significant increases in the post-guideline cohort compared with the pre-guideline cohort at both baseline and week 12 (all P values <.001). Glucose monitoring rates were higher than lipid monitoring rates, baseline monitoring rates were higher than week 12 rates, and there was a small increase in the proportion of patients who were monitored for metabolic effects in the post- vs pre-guideline periods. Stratification of lipid and glucose testing rates by age showed that older patients were more frequently tested for metabolic effects in both the pre- and post-guideline periods. Up to age 34, increasing age consistently increased the likelihood of lipid or glucose testing at baseline and week 12, before and after the guidelines were issued. Sex had no impact on lipid testing, but males were significantly less likely than females to receive glucose testing in the pre- and post-guideline periods, as well as at baseline and week 12. Nevertheless, post-guideline (baseline and 12 weeks) testing rates remained low.54

The New York State Department of Health and Office of Mental Health (OMH) have responded to these concerns by collaborating on an initiative designed to improve the quality and safety of psychotropic prescribing in the Medicaid population. The central vehicle of this initiative is the Psychiatric Services and Clinical Knowledge Enhancement System (PSYCKES), an award-winning Web-based computer program developed by OMH. PSYCKES is currently used by clinicians to alert prescribers and other health care professionals when one of their patients should receive a treatment review and may need a change of antipsychotic medication.55 Because implementation of this program was recently completed, results are not yet available.

Monitoring has also been conducted at a more remote level using physician surveys and cohort studies. For example, a survey assessed psychiatric clinicians’ perceptions about the impact of pharmacotherapy on weight gain and adverse metabolic effects in patients with bipolar disorder. The survey, a self-administered questionnaire, was sent to 7000 psychiatrists in November 2005. In January 2006 an additional 7000 psychiatrists who treat bipolar disorder in their practices were also surveyed. The first 298 completed surveys were analyzed. About two-thirds of respondents reported that 30% to 60% of their bipolar disorder patients were overweight, and 38% of respondents reported the presence of metabolic syndrome in 20% to 40% of their patients. Almost all respondents (96%) reported a 20-lb increase in a patient’s weight as a troublesome potential adverse event associated with the use of some agents. After initiating a new medication, >80% of respondents monitored their patients’ weights, fasting plasma glucose levels, and fasting lipid profiles at regular intervals. However, 80% did not monitor waist circumference. Furthermore, the survey response rate was <3%.56 Similarly, another survey showed that <30% of psychiatrists measured lipid or glucose levels in most patients prior to initiating treatment with a second-generation antipsychotic.57

Recently, a retrospective, multistate Medicaid cohort study evaluated testing rates from 1998 to 2003. It was observed that lipid and glucose testing was underused in patients receiving second-generation antipsychotics before the ADA guidelines were released. On average, <20% of individuals starting therapy with a second-generation antipsychotic received baseline glucose testing, and <10% received lipid testing.58

Despite the low-level effect—at best—of monitoring on metabolic effects/weight gain, dietary suggestions and behavioral strategies have produced some promising results. For example, a meta-analysis was conducted of all relevant randomized controlled trials (RCTs) of nonpharmacologic interventions and specific treatment approaches to control antipsychotic-induced weight gain in patients with first-episode or chronic schizophrenia. Trials included in the meta-analysis investigated cognitive-behavioral intervention strategies, nutritional counseling interventions, and combined nutritional and exercise interventions. A statistically significant reduction in mean body weight was demonstrated for those in the nonpharmacologic intervention groups compared with those in treatment-as-usual groups (mean weight loss, 2.56 kg).59 Overall, this was a well-done meta-analysis that, like any such study, depends on control of bias in each of the original studies reviewed.

Another study examined the impact of an intensive program of diet, exercise, and counseling on antipsychotic-treated patients with a weight gain of >4.5 kg and an increase in BMI of >5% since the start of antipsychotic treatment. During 24 weeks, the mean weight loss was 6.0 kg, and BMI was reduced by 5.7%.18 However, there was no comparison group.

The charts of patients who gained at least 9 kg after initiation of an antipsychotic agent were reviewed. This study concluded that approximately one-quarter of the patients experiencing weight gain may have been able to reverse the process eventually, particularly with use of behavioral interventions, eg, dietitian counseling and self-directed diet.60 Due to the retrospective study design and nonrandom allocation of patients to weight-loss interventions, however, the results should be applied cautiously.

Another study collected data from 53 patients whose BMI had increased by >7% after starting antipsychotic therapy. The study participants consented to participate in a 12-week educational intervention study aimed at preventing further weight gain and, when possible, inducing weight loss. Body weight and BMI were measured at baseline, at each of the monthly follow-up visits, and at study completion 12 weeks later. Completers showed a significant mean body weight decrease of 3.15 kg, with a mean BMI reduction of 1.2 kg/m2 at the end of the study period.53 The dropout rate was high (>50%) and concomitant medication was not taken into account. Nevertheless, the study was prospective in nature and merits replication with a larger, RCT design.

Pharmacologic interventions. Various pharmacologic agents and switching strategies have also been studied in an effort to control antipsychotic-induced metabolic effects/weight gain.

Two recent placebo-controlled RCTs provide encouraging data. In the first, 15 patients with a diagnosis of schizophrenia, on stable olanzapine regimens (mean duration, 71 months), were randomized to 4 weeks of double-blind, add-on aripiprazole, 15 mg/d, or placebo.61 After a subsequent 2-week washout period, patients were crossed over to the alternative treatment. Baseline eligibility criteria included a BMI ≥30 or a BMI >26 along with other metabolic risk factors (eg, hypertension, lipid abnormalities, or elevated fasting glucose). During aripiprazole treatment, there was a mean loss of 3 lb compared with a gain of 2 lb with placebo. The mean BMI of 33 was reduced by active treatment but increased on placebo (–0.4 vs +0.3, respectively). Levels of triglycerides and very-low-density lipoprotein (VLDL) cholesterol were significantly reduced from baseline, but there was no significant change in total, high-density lipoprotein (HDL) and low-density lipoprotein (LDL) cholesterol levels, or in blood pressure. Symptoms of schizophrenia were unaffected. Overall, this was a methodologically sound study in which baseline body weight/metabolic factors were controlled for using analysis of covariance. However, because the study was exploratory in nature, Bonferroni corrections were not made for the multiple statistical tests performed, calling into question the significance of the results.

In the second study, 61 patients diagnosed with schizophrenia, schizophreniform disorder, or bipolar disorder who were on stable clozapine regimens were randomly allocated to 1000 mg/d metformin or placebo.62 The patients had received clozapine for a mean duration of 6.5 to 8 years. This was a 14-week study, with body weight as the primary outcome measure. However, baseline obesity was more frequent in the treatment group than in the placebo group (50% vs 33%). Body weight decreased in the metformin group by an average of 4 lb, whereas it was unchanged with placebo. In the treatment group, BMI, glycosylated hemoglobin (HbA1C), and leptin levels decreased slightly, and HDL cholesterol increased slightly. Both groups exhibited diminished serum glucose over time that did not separate statistically. More importantly, metabolic syndrome decreased from 25% at baseline to 14% at 14 weeks in the metformin group but remained stable in the placebo group. However, 7 patients in the treatment group withdrew because of nonspecific side effects, and study participants had a wide range of body weight and metabolic status, biasing the results.

Before its withdrawal from the US market in October 2010, sibutramine (a serotonin-norepinephrine-dopamine reuptake inhibitor) was an effective and well-tolerated adjunct to behavior modification for weight loss in patients with schizophrenia and schizoaffective disorder who were being treated with olanzapine.63 In a double-blind RCT, the sibutramine and placebo groups demonstrated no significant baseline differences in age, sex, education, ethnicity, diagnosis, body weight, BMI, and blood pressure. At week 12, the sibutramine group had significantly greater losses than the placebo group in body weight (mean, 8.3 lb vs mean, 1.8 lb), waist circumference, BMI, and HbA1c. This was a well-designed study. Weekly group sessions that focused on nutrition and behavioral modification were provided to both groups for the first 8 weeks. Unfortunately, no assessment was performed to determine if an interaction effect (sibutramine and counseling) occurred.

In a different RCT of sibutramine, for clozapine-associated weight gain in patients with schizophrenia or schizoaffective disorder, no significant differences were found in changes in body weight, BMI, abdominal and waist circumferences, or waist/hip ratio. The range of body weight change was–15 lb to +10 lb for the sibutramine group and–9 lb to +20 lb for the placebo group. The 2 treatment groups did not show significant differences in HbA1c, fasting glucose, and cholesterol levels.64 However, underpowering and short study duration may have contributed to the nonsignificant outcomes.

Four other RCTs have generally shown more consistent outcomes. In a 12-week study, 66 overweight, hospitalized patients with schizophrenia received topiramate at doses of 100 mg/d or 200 mg/d, or placebo. In the 200-mg/d topiramate group, body weight, BMI, and waist and hip measurement decreased significantly compared with the 100-mg/d topiramate and placebo groups over 12 weeks.65 Of note, baseline BMI was marginally larger (near statistical significance) in the placebo and 200-mg/d groups.

Results from another RCT demonstrated that patients given olanzapine and reboxetine (a selective norepinephrine reuptake inhibitor, which is not FDA-approved for use in the United States), demonstrated a significantly lower increase in body weight (mean, 2.5 kg) than patients given olanzapine and placebo (mean, 5.5 kg).66 Disparities in baseline BMI, short study duration, and small sample size were methodological flaws.

A post hoc analysis of data from an RCT involving prescription of antipsychotic medication to nevertreated, first-onset psychosis patients compared the weight changes after 6 weeks of olanzapine treatment (standard vs disintegrating formulation). Results showed a significantly greater weight gain among patients taking standard tablets compared with those taking orally disintegrating olanzapine. Interestingly, baseline BMI was higher in the olanzapine tablet group, potentially biasing the results in its favor.67 However, a recent controlled study by the manufacturer of olanzapine rebuts the above study finding.68 Further investigation into the weight-change side effects of the orally disintegrating formulation appears warranted.

Lastly, in an RCT with modest results, a mean weight loss of 0.19 kg (±4.58) was observed after 16 weeks in patients with schizophrenia, schizoaffective, schizophreniform, or bipolar I disorder who received amantadine in addition to their olanzapine regimens.69 Those patients receiving placebo gained an average of 1.3 kg (±4.3). However, the mean BMI of the placebo group at baseline was 1.4 kg/m2 higher, biasing the outcome in its favor.

A somewhat less methodologically sound study included patients who were age ≥18 who received ≥2 prescriptions for risperidone or olanzapine for at least 60 days. The study participants were then switched to the other atypical antipsychotic and were dispensed 2 more prescriptions for at least 60 days after the index date. Age, body weight, and BMI were retrospectively abstracted from automated databases containing patient-specific prescriptions and vital-sign information. The average body weight change in patients switched to olanzapine was +2.3 kg, and the BMI change was +0.8 kg/m2. The average weight change after patients were switched to risperidone was–0.45 kg, with a change of–0.2 kg/m2 in BMI.70 Study limitations include a retrospective naturalistic design, unknown diagnoses, and variable postmedication switch time intervals.

Finally, a case series investigated the use of orlistat (reversible lipase inhibitor) in 13 patients with a mean weight gain of 16.4 kg secondary to antipsychotic drug use. Results showed that this agent, administered in 3 daily doses with meals, was safe, well tolerated, and effective, resulting in an average weight loss of 35% during an acute treatment period of 3 months.18 However, the absence of a control group limits the internal validity of this study and calls into question the statistical significance of its outcome.


Obesity and its medical sequelae are significant clinical challenges in the treatment of patients with SPMI. Fortunately, some progress has been made using nonpharmacologic and pharmacologic interventions, albeit with low to medium effect sizes. Accordingly, improved therapeutic methods with more substantial containment of weight gain are needed, ideally before the associated morbidity is manifest. Factors that will likely contribute to such advancement include a better understanding of the mechanisms involved, earlier intervention, and design of adequately powered RCTs of sufficient duration that control for baseline body weight/BMI. Indeed, this last factor has been shown to predict both body weight gain32 and loss.71,72 Whether or not heightened monitoring, such as that used in the PSYCKES program, contributes to body weight control remains to be determined. Nevertheless, care for patients with SPMI will benefit from progress in one or a combination of improved interventional factors that, it is hoped, carry a minimal side effect burden.

ACKNOWLEDGMENT: The authors thank Ms. Diane DeLuca for her technical assistance.

DISCLOSURES: Dr. Schwartz has been a paid researcher for Bristol-Myers Squibb, Cephalon, Cyberonics, Inc., Forest Pharmaceuticals, Pfizer Inc, and Wyeth Pharmaceuticals; a paid corporate speaker for AstraZeneca, Bristol-Myers Squibb, Cephalon, Cyberonics, Inc., PamLab, L.L.C., Pfizer Inc, and Wyeth Pharmaceuticals; and a consultant for Cephalon, PamLab, L.L.C., and Wyeth Pharmaceuticals. Drs. Megna, Siddiqui, and Herrera Rojas report no financial/personal relationships with other individuals/commercial entities (eg, pharmaceutical companies) whose products are mentioned in this article or with manufacturers of competing products.


  1.  Centers for Disease Control and Prevention. NCHS Health E-Stat. Prevalence of overweight obesity and extreme obesity among adults: United States, trends 1960-62 through 2005-2006. Accessed February 10, 2011.
  2. De Hert M, Schreurs V, Vancampfort D, et al Metabolic syndrome in people with schizophrenia: a review. World Psychiatry. 2009;8:15–22.
  3. Saha S, Chant D, McGrath J. A systematic review of mortality in schizophrenia: is the differential mortality gap worsening over time? Arch Gen Psychiatry. 2007;64:1123–1131.
  4. Citrome L, Blonde L, Damatarca C. Metabolic issues in patients with severe mental illness. South Med J. 2005;98:714–719.
  5.  American Diabetes Association, American Psychiatric Association, American Association of Clinical Endocrinologists, et al. Consensus development conference on antipsychotic drugs and obesity and diabetes. Diabetes Care. 2004;27:596–601.
  6. Leucht S, Burkard T, Henderson J, et al Physical illness and schizophrenia: a review of the literature. Acta Psychiatr Scand. 2007;116:317–333.
  7. Meyer JM, Nasrallah HA. Medical illness and schizophrenia. Washington DC: American Psychiatric Publishing, Inc.; 2003.
  8. Must A, Spadano J, Coakley EH, et al The disease burden associated with overweight and obesity. JAMA. 1999;282:1523–1529.
  9. Osborn DP, Nazareth I, King MB. Risk for coronary heart disease in people with severe mental illness: cross-sectional comparative study in primary care. Br J Psychiatry. 2006;188:271–277.
  10. Thakore JH, Mann JN, Vlahos I, et al Increased visceral fat distribution in drug-naïve and drug-free patients with schizophrenia. Int J Obes Relat Metab Disord. 2002;26:137–141.
  11. Kohen D. Diabetes mellitus and schizophrenia: historical perspective. Br J Psychiatry Suppl. 2004; 184:S64–S66.
  12. Newcomer JW. Second generation (atypical) antipsychotics and metabolic effects: a comprehensive literature review. CNS Drugs. 2005;19(suppl 1):1–93.
  13. Thakore JH. Metabolic syndrome and schizophrenia. Br J Psychiatry. 2005;186:455–456.
  14. van Winkel R, van Os J, Celic I, et al Psychiatric diagnosis as an independent risk factor for metabolic disturbances: results from a comprehensive, naturalistic screening program. J Clin Psychiatry. 2008;69:1319–1327.
  15. Torrent C, Amann B, Sánchez-Moreno J, et al Weight gain in bipolar disorder: pharmacological treatment as a contributing factor. Acta Psychiatr Scand. 2008;118:4–18.
  16. Jones M, Jones A. The effect of antipsychotic medication on metabolic syndrome. Nurs Stand. 2008; 22:43–48.
  17. Allison DB, Newcomer JW, Dunn AL, et al Obesity among those with mental disorders: a National Institute of Mental Health meeting report. Am J Prev Med. 2009;36:341–350.
  18. Tschoner A, Engl J, Laimer M, et al Metabolic side effects of antipsychotic medication. Int J Clin Pract. 2007;61:1356–1370.
  19. Meyer JM, Stahl SM. The metabolic syndrome and schizophrenia. Acta Psychiatr Scand. 2009;119:4–14.
  20. Cohn TA, Remington G, Zipursky RB, et al Insulin resistance and adiponectin levels in drug-free patients with schizophrenia: a preliminary report. Can J Psychiatry. 2006;51:382–386.
  21. Spelman LM, Walsh PI, Sharifi N, et al Impaired glucose tolerance in first-episode drug-naïve patients with schizophrenia. Diabet Med. 2007;24:481–485.
  22. Venkatasubramanian G, Chittiprol S, Neelakantachar N, et al Insulin and insulin-like growth factor-1 abnormalities in antipsychotic-naïve schizophrenia. Am J Psychiatry. 2007;164:1557–1560.
  23. van Nimwegen LJ, Storosum JG, Blumer RM, et al Hepatic insulin resistance in antipsychotic naive schizophrenic patients: stable isotope studies of glucose metabolism. J Clin Endocrinol Metab. 2008;93:572–577.
  24. Bergman RN, Ader M. Atypical antipsychotics and glucose homeostasis. J Clin Psychiatry. 2005;66:504–514.
  25. Daumit GL, Goldberg RW, Anthony C, et al Physical activity patterns in adults with severe mental illness. J Nerv Ment Dis. 2005;193:641–646.
  26. Meyer JM. Awareness of obesity and weight issues among chronically mentally ill inpatients: a pilot study. Ann Clin Psychiatry. 2002;14:39–45.
  27. Rege S. Antipsychotic induced weight gain in schizophrenia: mechanisms and management. Aust N Z J Psychiatry. 2008;42:369–381.
  28. Lane HY, Liu YC, Huang CL, et al Risperidone-related weight gain: genetic and nongenetic predictors. J Clin Psychopharmacol. 2006;26:128–134.
  29. Aichhorn W, Whitworth AB, Weiss EM, et al Second-generation antipsychotics: is there evidence for sex differences in pharmacokinetic and adverse effect profiles? Drug Saf. 2006;29:587–598.
  30. Saddichha S, Ameen S, Akhtar S. Predictors of antipsychotic-induced weight gain in first-episode psychosis: conclusions from a randomized double-blind, controlled prospective study of olanzapine, risperidone, and haloperidol. J Clin Psychopharmacol. 2008;28:27–31.
  31. Haddad P. Weight change with atypical antipsychotics in the treatment of schizophrenia. J Psychopharmacol. 2005;19(6 suppl):16–27.
  32. Megna JL, Raj Kunwar A, Wade MJ. A retrospective study of weight changes and the contributing factors in short term adult psychiatric inpatients. Ann Clin Psychiatry. 2006;18:163–167.
  33. Stahl S, Mignon L, Meyer JM. Which comes first: atypical treatment or cardiometabolic risk? Acta Psychiatr Scand. 2009;119:171–179.
  34. Nasrallah HA. Atypical antipsychotic-induced metabolic side effects: insights from receptor-binding profiles. Mol Psychiatry. 2008;13:27–35.
  35. Daumit GL, Clark JM, Steinwachs DM, et al Prevalence and correlates of obesity in a community sample of individuals with severe and persistent mental illness. J Nerv Ment Dis. 2003;191:799–805.
  36. Gentile S. Long-term treatment with atypical antipsychotics and the risk of weight gain: a literature analysis. Drug Saf. 2006;29:303–319.
  37. Kim B, Kim SJ, Son JI, et al Weight change in the acute treatment of bipolar I disorder: a naturalistic observational study of psychiatric inpatients. J Affect Disord. 2008;105:45–52.
  38. Henderson DC. Schizophrenia and comorbid metabolic disorders. J Clin Psychiatry. 2005;66(suppl 6): 11–20.
  39. Mulder H, Franke B, van der-Beek van der AA, et al The association between HTR2C gene polymorphisms and the metabolic syndrome in patients with schizophrenia. J Clin Psychopharmacol. 2007;27:338–343.
  40. Buchholz S, Morrow AF, Coleman PL. Atypical antipsychotic-induced diabetes mellitus: an update on epidemiology and postulated mechanisms. Intern Med J. 2008;38:602–606.
  41. Gunes A, Melkersson KI, Scordo MG, et al Association between HTR2C and HTR2A polymorphisms and metabolic abnormalities in patients treated with olanzapine or clozapine. J Clin Psychopharmacol. 2009;29:65–68.
  42. Reynolds GP, Zhang ZJ, Zhang XB. Association of antipsychotic drug-induced weight gain with a 5-HT2C receptor gene polymorphism. Lancet. 2002; 359:2086–2087.
  43. Reynolds GP, Zhang Z, Zhang X. Polymorphism of the promoter region of the serotonin 5-HT(2C) receptor gene and clozapine-induced weight gain. Am J Psychiatry. 2003;160:677–679.
  44. Mulder H, Franke B, van der-Beek van der AA, et al The association between HTR2C polymorphisms and obesity in psychiatric patients using antipsychotics: a cross-sectional study. Pharmacogenomics J. 2007;7:318–324.
  45. Reynolds GP, Hill MJ, Kirk SL. The 5-HT2C receptor and antipsychotic-induced weight gain—mechanisms and genetics. J Psychopharmacol. 2006;20(4 suppl):15–18.
  46. Herrán A, García-Unzueta MT, Amado JA, et al Effects of long-term treatment with antipsychotics on serum leptin levels. Br J Psychiatry. 2001;179:59–62.
  47. Sussman N. The implications of weight changes with antipsychotic treatment. J Clin Psychopharmacol. 2003;23(3 suppl 1):S21–S26.
  48. Zhang ZJ, Yao ZJ, Liu W, et al Effects of antipsychotics on fat deposition and changes in leptin and insulin levels. Magnetic resonance imaging study of previously untreated people with schizophrenia. Br J Psychiatry. 2004;184:58–62.
  49. Starrenburg FC, Bogers JP. How can antipsychotics cause diabetes mellitus? Insights based on receptor-binding profiles humoral factors and transporter proteins. Eur Psychiatry. 2009;24:164–170.
  50. Sentissi O, Epelbaum J, Olié JP, et al Leptin and ghrelin levels in patients with schizophrenia during different antipsychotics treatment: a review. Schizophr Bull. 2008;34:1189–1199.
  51. Henderson DC, Doraiswamy PM. Prolactin-related and metabolic adverse effects of atypical antipsychotic agents. J Clin Psychiatry. 2008;69(suppl 1):32–44.
  52. Wu RR, Zhao JP, Zhai JG, et al Sex difference in effects of typical and atypical antipsychotics on glucose-insulin homeostasis and lipid metabolism in first-episode schizophrenia. J Clin Psychopharmacol. 2007;27:374–379.
  53. Mauri M, Castrogiovanni S, Simoncini M, et al Effects of an educational intervention on weight gain in patients treated with antipsychotics. J Clin Psychopharmacol. 2006;26:462–466.
  54. Haupt DW, Rosenblatt LC, Kim E, et al Prevalence and predictors of lipid and glucose monitoring in commercially insured patients treated with second-generation antipsychotic agents. Am J Psychiatry. 2009; 166:345–353.
  55.  The Coalition of Behavioral Health Agencies, Inc. Coalition Briefs. OMH and DOH partner to improve quality and safety of psychotropic treatment in NYS. December 23, 2008. Accessed March 7, 2011.
  56. Ketter TA, Haupt DW. Perceptions of weight gain and bipolar pharmacotherapy: results of a 2005 survey of physicians in clinical practice. Curr Med Res Opin. 2006;22:2345–2353.
  57. Buckley PF, Miller DD, Singer B, et al Clinicians’ recognition of the metabolic adverse effects of antipsychotic medications. Schizophr Res. 2005;79:281–288.
  58. Morrato EH, Newcomer JW, Allen RR, et al Prevalence of baseline serum glucose and lipid testing in users of second-generation antipsychotic drugs: a retrospective, population-based study of Medicaid claims data. J Clin Psychiatry. 2008;69:316–322.
  59. Alvarez-Jiménez M, Hetrick SE, González-Blanch C, et al Non-pharmacological management of antipsychotic-induced weight gain: systematic review and meta-analysis of randomized controlled trials. Br J Psychiatry. 2008;193:101–107.
  60. O’Keefe CD, Noordsy DL, Liss TB, et al Reversal of antipsychotic-associated weight gain. J Clin Psychiatry. 2003;64:907–912.
  61. Henderson DC, Fan X, Copeland PM, et al Aripiprazole added to overweight and obese olanzapine-treated patients. J Clin Psychopharmacol. 2009; 29:165–169.
  62. Carrizo E, Fernández V, Connell L, et al Extended release metformin for metabolic control assistance during prolonged clozapine administration: a 14 week, double-blind, parallel group, placebo-controlled study. Schizophr Res. 2009;113:19–26.
  63. Henderson DC, Copeland PM, Daley TB, et al A double-blind, placebo-controlled trial of sibutramine for olanzapine-associated weight gain. Am J Psychiatry. 2005;162:954–962.
  64. Henderson DC, Fan X, Copeland PM, et al A double-blind, placebo-controlled trial of sibutramine for clozapine-associated weight gain. Acta Psychiatr Scand. 2007;115:101–105.
  65. Ko YH, Joe SH, Jung IK, et al Topiramate as an adjuvant treatment with atypical antipsychotics in schizophrenic patients experiencing weight gain. Clin Neuropharmacol. 2005;28:169–175.
  66. Poyurovsky M, Isaacs I, Fuchs C, et al Attenuation of olanzapine-induced weight gain with reboxetine in patients with schizophrenia: a double-blind, placebo-controlled study. Am J Psychiatry. 2003;160:297–302.
  67. Arranz B, San L, Dueñas RM, et al Lower weight gain with the orally disintegrating olanzapine than with standard tablets in first-episode never treated psychotic patients. Hum Psychopharmacol. 2007;22:11–15.
  68. Karagianis J, Grossman L, Landry J, et al A randomized controlled trial of the effect of sublingual orally disintegrating olanzapine versus oral olanzapine on body mass index: the PLATYPUS Study. Schizophr Res. 2009;113:41–48.
  69. Deberdt W, Winokur A, Cavazzoni PA, et al Amantadine for weight gain associated with olanzapine treatment. Eur Neuropsychopharmacol. 2005;15:13–21.
  70. Ried LD, Renner BT, Bengtson MA, et al Weight change after an atypical antipsychotic switch. Ann Pharmacother. 2003;37:1381–1386.
  71. Black DW, Goldstein RB, Mason EE. Psychiatric diagnosis and weight loss following gastric surgery for obesity. Obes Surg. 2003;13:746–751.
  72. Hansen D, Astrup A, Toubro S, et al; STORM Study Group. Predictors of weight loss and maintenance during 2 years of treatment by sibutramine in obesity. Results from the European multi-centre STORM trial. Sibutramine Trial of Obesity Reduction and Maintenance. Int J Obes Relat Metab Disord. 2001;25:496–501.

CORRESPONDENCE: James L. Megna, MD, PhD Department of Psychiatry SUNY Upstate Medical University 750 E. Adams Street Syracuse, NY 13210 USA E-MAIL: