Major Finding: Food and Drug Administration warnings and consensus recommendations have not increased testing for glucose levels and lipid profiles in Medicaid patients who are starting second-generation antipsychotics.
Data Source: Retrospective, population-based case-control study of data on 312,978 patients receiving Medicaid in three states.
Disclosures: The study was funded by Pfizer, which makes ziprasidone. The investigator has received research funds from Eli Lilly and Co., maker of olanzapine. Several other investigators have received funding from or been consultants to those and other companies that make second-generation antipsychotics.
Clinicians reduced prescriptions for one second-generation antipsychotic medication associated with metabolic problems after the Food and Drug Administration required warnings in 2003 about increased risk for diabetes and hyperlipidemia with that class of drugs, a new study shows.
Clinicians did not increase metabolic testing, however, despite recommendations issued at that time by the American Diabetes Association (ADA) and the American Psychiatric Association (APA) to test glucose and lipid levels in all patients who start a second-generation antipsychotic, a large controlled study found.
The study looked at 109,451 Medicaid recipients who started second-generation antipsychotic medications, and a control cohort of 203,527 patients who began taking albuterol but did not receive a second-generation antipsychotic.
Before the FDA warnings and the ADA/APA recommendations, 27% of patients who started a second-generation antipsychotic underwent baseline serum glucose testing, compared with 26% of control patients. Clinicians got baseline lipid tests in 10% of patients starting a second-generation antipsychotic and 11% of controls.
The only statistically significant change in testing rates during or after issuance of the warnings and guidelines was a clinically insignificant 2% increase in baseline lipid testing in patients starting a second-generation antipsychotic, Elaine H. Morrato, Dr. P.H., and her associates reported (Arch. Gen. Psychiatry 2010;67:17–24).
Previous studies have shown that the warnings produced no clinically meaningful change in glucose and lipid monitoring in commercially insured patients. The current retrospective cohort study is the first to find a similar pattern in a Medicaid population, reported Dr. Morrato of the University of Colorado, Denver.
The FDA compelled drug manufacturers to change labels for second-generation antipsychotics starting in December 2003 and to mail “Dear healthcare professional” letters to neuropsychiatric health care providers through August 2004 to warn of increased risk for hyperglycemia and diabetes with use of second-generation antipsychotics and to recommend monitoring of glucose levels in patients with diabetes, risk factors for diabetes, or hyperglycemia.
Concurrently, ADA and APA consensus statements described the metabolic risks of second-generation antipsychotics and provided a monitoring protocol that included baseline tests of glucose levels and lipid profiles.
The current study used data from patients in California, Missouri, and Oregon during a prewarning period (January 2002 through November 2003), a warning period during which the letters were mailed and the ADA/APA recommendations came out (December 2003 through August 2004), and a postwarning period (September 2004 through December 2005).
Prescriptions for clozapine were excluded from the study because of the drug's unique requirement for neutropenia-related testing.
Although the warnings and recommendations did not increase rates of metabolic testing in patients starting second-generation antipsychotics, clinicians did reduce their use of olanzapine, which would be consistent with intent to reduce metabolic risk, the investigators noted.
Before the FDA warnings, olanzapine use was already declining by 5% per year, compared with other second-generation antipsychotics; during the warning period, olanzapine use declined by 20% per year.
Use of a second-generation antipsychotic with lower metabolic risk—aripiprazole—increased significantly during the warning period but not the postwarning period, a change that might have been attributable more to California's elimination of the need for prior authorization for aripiprazole prescriptions during that time than to concerns about diabetes and hyperglycemia, Dr. Morrato and her associates suggested. The warnings by the FDA and the ADA/APA recommendations did not affect the use of quetiapine, ziprasidone, or risperidone.
People with serious mental illness are 1.5-2 times more likely to develop dyslipidemia, hypertension, obesity, and type 2 diabetes, compared with the general population.
More effort is needed to increase screening for diabetes and dyslipidemia in patients receiving second-generation antipsychotics and monitoring for metabolic side effects of these drugs, the authors suggested.
In some previous studies, 60%-80% of psychiatrists reported that they do regularly monitor glucose and lipid levels in patients taking second-generation antipsychotics, and two-thirds of community mental health centers reported having protocols or procedures to screen patients for diabetes and dyslipidemia.
More research is needed to understand the discrepancy between these reports and the low rates of screening found in the current study.