Prediabetes assessment and follow-up in older veterans with schizophrenia

Assess glycemic monitoring and follow-up.

Retrospective study using administrative data (October 2001-September 2005).

Veterans Health Administration.

A nationwide sample of 39,226 outpatients aged 50 years or older with schizophrenia. Patients had no diagnosis or medications for diabetes at baseline.

Hemoglobin A1c tests; blood glucose tests with same-day low-density lipoprotein to approximate fasting glucose. Glycemic tests were combined to indicate a) prediabetic dysglycemia (100-125 mg/dL proxy fasting glucose or 5.8%-6.4% hemoglobin A1c) and b) diabetic dysglycemia (≥126 proxy fasting glucose or ≥6.5% A1c).

Approximately one-third of patients (32%; 12,587) had proxy fasting blood glucose or A1c tests in 2002; multiple tests were rare. The proportion tested increased to 40% by 2005. Test results suggested prediabetic dysglycemia for 5,345 tested patients (42% of those tested) and diabetic dysglycemia for 1,287 tested patients (10%) at baseline. In multivariate regression models, glycemic testing was associated with dyslipidemia, hypertension, and younger age. Dysglycemia was associated with hypertension, dyslipidemia, and older age. Follow-up treatment/diagnosis of diabetes occurred for 8% of patients (11% of those tested) and was associated with baseline dysglycemia, hypertension, and younger age. Mortality (15% during the 4-year study) was higher among untested and untreated patients.

Dysglycemia was prevalent among older patients with schizophrenia, although monitoring and follow-up were uncommon. Follow-up treatment correlated with survival. Despite evident utility of testing, few at-risk patients with schizophrenia were adequately monitored, diagnosed, or treated for dysglycemia.