Analysis of the Appropriateness of Off-Label Antipsychotic Use for Mental Health Indications in a Veteran Population.

STUDY OBJECTIVE: A substantial proportion of antipsychotic (AP) use in veterans is for nonapproved indications (i.e., off-label prescribing). Not all off-label use is necessarily detrimental to patients, however, and in certain situations, off-label prescribing could be considered justifiable. The objective of this study was to determine the extent to which off-label AP prescribing in a veteran population was potentially appropriate.

DESIGN: Expert panel and retrospective analysis.

DATA SOURCE: Veterans Health Administration (VHA) Corporate Data Warehouse.

PATIENTS: A total of 69,823 veterans who had at least one pharmacy record for an AP medication during fiscal years 2005-2012.

MEASUREMENTS AND MAIN RESULTS: An expert panel was convened to determine if agreement exists on the appropriateness of AP use in various scenarios. The panel consisted of 10 experts in the field of psychiatry: nine physicians with various specialties, and one pharmacist. We used a modified RAND appropriateness method approach to identify potentially appropriate, uncertain, and inappropriate cases of AP use. The use of six second-generation APs was examined individually, and the use of first-generation APs was examined as a class. Based on data previously collected quantifying VHA AP use, the panel was given disease state scenarios for the most commonly occurring off-label diagnoses for AP prescriptions. Disease states were coupled with scenario modifiers that the expert panel considered potentially significant clinical factors. Among the four disease states-anxiety, dementia, insomnia, and posttraumatic stress disorder-29 scenarios were investigated for each AP. None of the scenarios were judged by the expert panel to be appropriate for the use of APs. Of the 203 scenarios for all APs, 60% were judged to be inappropriate by the expert panel, and the remaining 40% were considered uncertain. Of the AP medications, risperidone (72%) and olanzapine (62%) were the most likely to be seen as uncertain, whereas first-generation APs (86%) were the most likely to be considered inappropriate in a given scenario. Widespread off-label use of APs outside of the approved indications of treatment of schizophrenia and bipolar disorder, or adjunctive treatment of major depressive disorder, may not be an appropriate treatment option. According to this expert panel, no examined situations were considered appropriate for the use of APs.

CONCLUSION: The consensus of our expert panel was that off-label AP use is uncertain at best and more likely, even in complicated cases, inappropriate. These findings strengthen the case for stronger control of APs in integrated health care settings such as the VHA, as well as better education and information for practitioners who provide care for patients with anxiety, dementia, insomnia, or posttraumatic stress disorder.