Medicare Hospice Reform and End-of-Life Care

• In Event: Public Health Insurance Programs: Cost Containment, Equity, and Care Burden Among Aging and Disabled Populations

Abstract

Nearly nine in ten caregivers of dying older adults are unpaid family members (Ornstein et al., 2017), with hospice providing an important source of formal support. To limit long hospice stays, Medicare introduced audits targeting hospices with high shares of long-stay patients under the 2014 IMPACT Act and reduced daily payments after day 60 in 2016. Previous studies find that these reforms reduced hospice enrollment among long-stay patients, including those with dementia (Gianattasio et al., 2022; Nandan, 2024). However, little is known about their effects on family caregiving. I examine this question using the Gateway Harmonized HRS End of Life file linked to the RAND HRS, covering 9,321 Medicare decedents who died between 2000 and 2019. Following Nandan (2024), I predict hospice length of stay using pre-reform patients' diagnoses, health care utilization, and demographic characteristics. I then compare changes in outcomes between decedents with longer and shorter predicted stays before and after the reforms. A 30-day increase in predicted hospice stay, approximately one standard deviation, is associated with a 3.2 percentage point (7 percent) reduction in paid care during the last three months of life and a 1.3 percentage point reduction in the probability of dying in a hospice facility. I find no evidence of increases in family caregiving. Estimates for family caregiving hours and intensive caregiving are positive but imprecise. I find no evidence of increased hospital or nursing-home utilization. These preliminary findings will be further examined using linked Medicare claims to measure hospice enrollment and payments directly.

Author

• Soohyun Kim, Syracuse University
  Presenting Author