Prevalence of obesity and diabetes in patients with schizophrenia

To compare the prevalence of diabetes in patients with schizophrenia treated at a community mental health center with controls in the same metropolitan area and to examine the effect of antipsychotic exposure on diabetes prevalence in schizophrenia patients.

The study was a comprehensive chart review of psychiatric notes of patients with schizophrenia and schizoaffective disorder treated at a psychosis program in a community mental health center. Data collected included psychiatric diagnoses, diabetes mellitus diagnosis, medications, allergies, primary care status, height, weight, body mass index (BMI), substance use and mental status exam. Local population data was downloaded from the Centers for Disease Control Behavioral Risk Factor Surveillance System. Statistical methods used were χ² test, Student's t test, general linear model procedure and binary logistic regression analysis.

The study sample included 326 patients with schizophrenia and 1899 subjects in the population control group. Demographic data showed control group was on average 7.6 years older (P = 0.000), more Caucasians (78.7% vs 38.3%, P = 0.000), and lower percentage of males (40.7% vs 58.3%, P = 0.000). Patients with schizophrenia had a higher average BMI than the subjects in the population control (32.11, SD = 7.72 vs 27.62, SD = 5.93, P = 0.000). Patients with schizophrenia had a significantly higher percentage of obesity (58.5% vs 27%, P = 0.000) than the population group. The patients with schizophrenia also had a much higher rate of diabetes compared to population control (23.9% vs 12.2%, P = 0.000). After controlling for age sex, and race, having schizophrenia was still associated with increased risk for both obesity (OR = 3.25, P = 0.000) and diabetes (OR = 2.42, P = 0.000). The increased risk for diabetes remained even after controlling for obesity (OR = 1.82, P = 0.001). There was no difference in the distribution of antipsychotic dosage, second generation antipsychotic use or multiple antipsychotic use within different BMI categories or with diabetes status in the schizophrenia group.

This study demonstrates the high prevalence of obesity and diabetes in schizophrenia patients and indicates that antipsychotics may not be the only contributor to this risk.