Effects of an antipsychotic restriction policy in a veteran population

Abstract

Introduction: Recent trials have failed to demonstrate differences in efficacy between first generation antipsychotics (FGAs) and second generation antipsychotics (SGAs). To reduce costs, many health care systems have restricted the availability of SGAs through use of prior authorizations. Restrictions for the off-label use of SGAs and the use of dual-antipsychotic therapy have also been implemented in many health care systems. At the South Texas Veterans Health Care System (STVHCS), a restricted drug request (RDR) method has been implemented to manage costs and improve patient safety. Risperidone, due to its lower cost and equal efficacy, is the first-line option of SGAs. If one wishes to prescribe an SGA other than risperidone, an RDR is submitted and reviewed by Veterans Integrated Service Network (VISN) pharmacists. Since the introduction of these policies at the STVHCS, the impact of the RDR has not been assessed. Rationale: The primary aim of this study was to determine the effects of the RDR policy on the care of STVHCS veterans as evidenced by changes in hospitalization rates of veterans with a denied request for an SGA due to initial criterion failure. Secondary outcomes included: impact of antipsychotic RDR denial on mental health as evidenced by changes in no-shows and cancellations for follow-up psychiatric appointments, psychiatric emergency department visits, presence of suicidal ideation, change in weight, hemoglobin A1c, number of psychotropic medications prescribed, and extrapyramidal symptoms. Methods: A retrospective chart review of veterans denied an initial SGA request was conducted from 3 months prior to denial to 3 months post request denial (index date). Data collected included: patient demographics, indication for SGA request, reason for SGA denial, length of time for request evaluation, number of psychiatric hospitalizations, number of no-shows and cancellations for mental health appointments, number of psychiatric emergency department visits, number of reports of suicidal ideation or attempts, weight, hemoglobin A1c lab results, presence of extrapyramidal symptoms, and number of prescribed psychotropic medications. The health care utilization data collected pre- and post-index date, were compared. Results were analyzed using Fisher's Exact, 2-tailed standardized t-tests, and descriptive statistics appropriately matched to data type. Results: Results for both primary and secondary outcomes were not statistically significant. No differences were found in the number of veterans hospitalized pre- versus post-index date [0/33 (0%) versus 2/33 (6%), p=0.492.] The most requested indication for an SGA was PTSD [22/33 (66.7%)] and the most frequently denied SGA was quetiapine [16/33 (48.5%)]. Conclusions: Although outcomes were not statistically significant, several valuable conclusions were drawn from this research. Positive outcomes from a RDR policy were seen by the limitations placed on inappropriate medication prescribing. Also, it was observed that the number of approvals for SGAs was almost three times higher than denials. A subsequent finding from this research is the apparent lack of metabolic monitoring for veterans prescribed SGAs. Further research on these observations, as well as conducting a pharmacoeconomic analysis on the RDR policy, would also be beneficial information for health care providers.