SUBSTITUTE PRESCRIBING WITH ANTIPSYCHOTIC REDUCTION

Introduction The Center for Medicare and Medicaid Services (CMS) has called for a reduction in antipsychotic use in the geriatric population, in large part due to the association of these medications with increased mortality. Antipsychotics have significant side effect profiles, including QTc prolongation, weight gain, diabetes mellitus, hypercholesterolemia, and sedation, and may contribute to overall mortality in ways that are not completely understood. As long term care facilities work to decrease their antipsychotic prescribing rates in line with CMS recommendations, they must contend with the challenge of finding alternative treatments for episodes of acute agitation. This observational study aimed to describe the response to this challenge in a geriatric subacute care facility by examining the Medication Administration Record (MAR) to determine if prescribing rates of other medications commonly used to treat agitation increased when antipsychotics were no longer used for this purpose. Methods The study site was a university-affiliated geriatric care facility, which contains long term chronic hospital units, memory support units, and subacute rehabilitation units. Medications were prescribed by geriatric medicine attendings, nurse practitioners, and physician assistants on each unit, with input at times from consulting psychiatrists. Prescribers were employed as members of the facility's closed medical staff. Nursing and administrative staff were generally stable over the time period study. The total census at both sites ranged from 668-725 patients. Over the course of our study, we examined two long term care units at the facility The study units discontinued the use of prescribing antipsychotics on an as needed (PRN) basis and began an initiative modeled after “Appropriate Use of Antipsychotics,” based on the Canadian Foundation for Healthcare Improvement and Appropriate Use of Antipsychotics collaborative. This initiative included a series of interventions to reduce antipsychotic use overall, including staff education and training in specific behavioral interventions. We collected data on the rates of prescribing medications, including antipsychotics, benzodiazepines and trazodone at four time periods: at baseline, some months prior to the start of the initiative (Jan – March 2017), after PRN antipsychotics were discontinued immediately prior to the start of the initiative (July – Sept 2017), during the initiative (January – March 2018) and later after the start of the initiative in follow-up (July – Sept 2018). Six days during each time period (the 15th and last day of the month for each of the 3 months in the period) were selected during which data were collected from the MAR examining the number of medication administrations of each medication. Further analysis was performed looking specifically at the rates of benzodiazepine and trazodone administrations, since these were found to be commonly prescribed medications. Results Administrations of antipsychotics decreased (from 22.0% of the total number of patients at the facility receiving an antipsychotic at time 1 to 19.0% in time 3; to 16.7% at time 4). Concurrently, rates of trazodone prescription rose significantly both by average number of doses per day prescribed (see figure 1; ANOVA for differences between the four time periods F(3,20)=95.43, P Conclusions As rates of antipsychotic prescribing on the units decreased, we observed a phenomenon of substitute prescribing in the form of a significant increase in the amount of trazodone prescribed. At the same time, there was a slight decrease in the amount of benzodiazepines prescribed. Our findings suggest that trazodone was felt by prescribers to be a safer alternative to antipsychotics. The absence of a similar increase in benzodiazepine prescribing suggests that prescribers remained wary of this class of medication because of its longstanding association with adverse effects such as falls. These data also suggest that an ongoing focus on behavioral interventions for agitation in dementia is still needed in order to minimize the phenomenon of substituting one medication with side effects for another with a different set of side effects. Side effects of trazodone include oversedation (which leads to risks of falls), hypotension, QT prolongation, and cognitive effects. A strength of this study is our access to accurate, detailed prescribing data. An additional strength is that the study took place at a facility with stable nursing and medical staffs, thereby decreasing the possibility that staff changes acted as confounders. Limitations of our study include that our data was observational, did not differentiate between PRN doses of a medication and standing doses, did not investigate the symptoms for which each medication was prescribed, and did not compare the effectiveness or adverse effects of these prescriptions. Future research is needed to describe the clinical impact of the phenomenon of medication substitution as antipsychotic prescribing continues to be reduced. This research was funded by: This study took place at Hebrew Rehabilitation Center/Hebrew SeniorLife and was supported by the Canadian Foundation for Healthcare Improvement, a not-for-profit organization funded by Health Canada.