Medicare ACO Use of Palliative Care Consults and Hospice
Topic: Impact of Program Financing
The healthcare sector's pivot toward value-based reimbursement may elevate the profiles of hospice and palliative care, both of which have demonstrated clinical and financial value but nonetheless are still characterized by highly variable utilization. The accountable care organization (ACO) model incentivizes providers to deliver efficient health care services. We examined 2014 data from Medicare's largest ACO model, the Medicare Shared Savings Program (MSSP), to characterize the utilization of palliative care consults and hospice election. Using a 50% sample of 2014 MSSP-aligned beneficiaries, those receiving palliative care consults were identified via ICD-9 code V66.7 in the inpatient, outpatient, and carrier claims files. Beneficiaries electing hospice were identified via the Medicare Beneficiary Summary File. Both the palliative and hospice utilization measures were indexed to the number of decedents to adjust for ACO cohort size and health status and capture intensity of usage at the ACO level. There was substantial variation in both palliative and hospice intensity among MSSPs in 2014, with palliative care intensity showing a mean of 0.26 (# of aligned beneficiaries receiving palliative care/# of aligned decedent beneficiaries) and a range of 0.02-0.48 and hospice election intensity having a mean of 0.47 (# of aligned beneficiaries electing hospice/# of aligned decedent beneficiaries) and a range of 0-0.79.
We then assessed MSSP provider networks to examine the extent to which MSSPs formally including a hospice and palliative care specialist also showed higher rates of hospice and palliative care intensity. Having a hospice or palliative care specialist in an MSSP's network was not associated with the intensity of hospice election in a statistically significant way, but did show a statistically significant association with the intensity of palliative consults. To further characterize MSSP approaches to hospice and palliative care, we examined the association of the care intensity measures with the MSSP quality measures and overall quality score. For both hospice and palliative care, the care intensity measures were not significantly associated with MSSP quality. Collectively, our findings further confirm the variability in how MSSPs are approaching hospice and palliative care, and also raises the concern that end-of-life quality of care is not sufficiently captured in the MSSP quality metrics. The intersection of high-value but underused services, such as hospice and palliative care, with a reimbursement model that rewards such care suggests that these services would be used more intensively in the MSSP model, but we do not yet see evidence of this happening in a systematic way.
- Julia Driessen
- Assistant Professor
- University of Pittsburgh
- 130 De Soto St, Crabtree A614
- Pittsburgh, PA 15261
- Turner West