Impact of Inpatient Palliative Care on Treatment Intensity for Patients with Serious Illness

Background: Palliative care is associated with decreased treatment intensity and improved quality for individual patients at the end of life, but little is known about how hospital-wide outcomes are affected by the diffusion of palliative care principles. Objective: We examined the relationship between presence of palliative care programs and hospitals' average treatment intensity, as indicated by mean intensive care unit (ICU) length of stay (LOS) and days under Medicare hospice coverage, in the last six months of life among Medicare beneficiaries aged 67 and over with serious chronic illness. Methods: We linked hospital-level data from the American Hospital Association Annual Survey, National Palliative Care Registry, and Dartmouth Atlas of Health Care to examine hospital-level treatment intensity for chronically ill Medicare beneficiaries who died in 2010. We used propensity score-adjusted linear regression to estimate the relationship between palliative care programs and hospitals' mean ICU LOS and hospice length of enrollment. Results: Among 974 hospitals meeting inclusion criteria, we compared 295 hospitals with palliative care programs to 679 hospitals without. Hospitals with palliative care programs were higher volume, more likely to be teaching hospitals, and have oncology services and less likely to be located in rural areas. In propensity score weighted analyses, the mean ICU LOS in hospitals with palliative care was shorter by 0.23 days (standard error [SE] = 0.26), but this was not statistically significant (p = 0.76). In addition, the mean length of hospice enrollment among beneficiaries served by hospitals with palliative care was longer by 0.22 days (SE = 0.61), but also was not statistically significant (p = 0.76). Conclusions: Hospital-based palliative care programs alone may not be sufficient to impact ICU LOS or hospice length of enrollment for all chronically ill older adults admitted to hospitals. Future work should measure hospital-wide palliative care outcomes and effects of core palliative knowledge and skills provided by nonpalliative care specialists.