Diabetes Prevalence and Risk Factors in Patients With Chronic Mental Illness on Second-Generation Antipsychotics

Abstract

The rates of obesity and type 2 diabetes (T2D) are much higher in patients with chronic mental illness compared to the general population1. Second-generation antipsychotic medications (SGA) are clearly contributory to these adverse metabolic phenotypes2. Thus, annual monitoring of fasting plasma glucose (FPG) levels is recommended to screen for T2D in patients on SGA3. Of note, FPG, and also HbA1c, have poor sensitivities in detecting T2D early, with nearly 50% of cases being missed using either of the above tests when compared to oral Glucose Tolerance Testing (oGTT,4). We thus screened patients with schizophrenia, schizoaffective or bipolar disorder on SGA per oGTT. We hypothesized that we would identify more patients with T2D per oGTT compared to FPG or HbA1c testing alone. To identify risk factors for T2D, we also assessed BMI, waist/hip ratio, blood pressure, dietary habits (UKDDQ = UK Diabetes and Diet Questionnaire), physical activity records (3d pedometer records), measures of psychopathology, cognition (PHQ-9 = Patient Health Questionnaire, SLUMS = Saint Louis University Mental Status exam) as well as HOMA IR (Homeostatic Model Assessment of Insulin Resistance), Matsuda index and lipid profile. Data were analyzed using SPSS, comparing normal and prediabetic patients to patients with T2D (per oGTT criteria). Thus far, we screened 22 patients per oGTT and newly identified 5 patients with T2D (23%), only one patient had T2D per HbA1c criterion, 4 were identified per elevated FPG, 3 patients had elevated 2h plasma glucose levels. Patients with T2D had significantly lower Matsuda indices (p<0.0001); trends towards higher HOMA IR and waist/hip ratios were observed in diabetic compared to non-diabetic patients (p=0.08). Patients with T2D tended to be heavier, had worse dietary habits (higher UKDDQ scores), and surprisingly were less depressed (lower PHQ-9 scores) than non-diabetic patients (p<0.25). In conclusion, oGTT is more sensitive in identifying patients with T2D early compared to FPG or HbA1c alone. Matsuda indices, HOMA IR and waist/hip ratios may serve as additional markers to distinguish non-diabetic from diabetic patients. Since metabolic syndrome and diabetes are all risk factors for an increased cardiovascular mortality in patients with chronic mental illness, identifying and treating T2D early and aggressively should be primary goals of care in this high-risk patient population.