Medicating distress: Use of psychotropic medications among patients with dementia receiving hospice care

ABSTRACT The goal of this Mentored Career Development Award is to prepare Dr. Lauren Gerlach for a career as an independent investigator focused on measuring and improving the quality of end-of-life care for older adults with psychiatric disorders and dementia. Dr. Gerlach’s long-term career goal is to establish herself as an independent investigator in geriatric mental health services and mixed methods research. Behavioral symptoms in advanced dementia are common, and despite limited evidence to support their use, the mainstay of such symptom management in hospice is off-label use of psychotropic medications such as antipsychotics. However, recent clinical trial evidence suggests that use of antipsychotics at end of life may be associated with worse outcomes for patients—raising concerns that the very medications prescribed to alleviate distress in fact exacerbate it. Little is known about how widely psychotropics are prescribed during end-of-life care for dementia, the factors that influence this prescribing, or the outcomes of such use. These critical knowledge gaps—including which patients are at greatest risk of potential harm—limit the ability to develop a consensus on best practices to address distressing behavioral symptoms in hospice care for dementia. Dr. Gerlach’s clinical training as a geriatric psychiatrist and prior research in pharmacoepidemiology provide the critical foundation for this work, but her transition to independence requires new skills and knowledge that will be gained through a coordinated program of research, mentorship, and coursework during the 5-year award period. This will include training in: 1) hospice-based dementia care, 2) use of Medicare and Minimum Data Set (MDS) data to examine end-of-life care, 3) advanced statistical methods for pharmacoepidemiology research, and 4) qualitative research methods. The proposed study will use Medicare data for all older adults with dementia enrolled in hospice in 2015 to determine the array of patient, provider, and hospice characteristics that influence psychotropic prescribing (Aim 1). In partnership with a local hospice organization, an exploratory sub-aim will examine the indications for psychotropic prescribing at end of life—information not available in claims data. Aim 2 will use linked Medicare and MDS data to examine outcomes associated with psychotropic prescribing (e.g., functional impairment, behavioral symptoms, mortality). Finally, Aim 3 will include interviews with patients with early-stage dementia, caregivers, and hospice providers to elicit stakeholder perspectives on the goals of hospice care. Study findings will inform how to best manage behavioral symptoms at the end of life. Given that not all patients with dementia utilize hospice, this K23 is designed to lay the foundation for a subsequent R01 to examine broader end-of-life care among all patients with dementia and to determine how hospice enrollment impacts outcomes. The R01 study will use Medicare and linked survey data to determine factors that impact the care patients receive in the last year of life and to identify the extent of potentially inappropriate treatments and impact on caregiver outcomes to better inform hospice care models for dementia.