Researcher unveils new hospice enrollment data
Length-of-stay problem is getting worse
Nicholas A. Christakis, MD, PhD, MPH, co-author of a widely cited study documenting the problem of shrinking length of stay in hospice1, reported on his latest research in this area at the Arlington, VA-based National Hospice Organization’s management meeting in Washington, DC, in May.
Christakis, assistant professor of medicine and sociology at the University of Chicago, has expanded on his published 1990 data by examining 184,000 Medicare patients enrolled in hospice in 1993. Among the results reported, median length of stay in hospice dropped from 36 to 30 days between 1990 and 1993. While the proportion of hospice patients living longer than 365 days on hospice stayed the same, the proportion living only two, seven, and 30 days increased, thus "compressing the front end of the hospice survival curve," he said.
"I have one very tentative idea why. Ten to 15 years ago, hospice was new. People using hospice were those who believed in it and really valued it," he explained. As hospice expanded, it came to include patients who weren’t as attracted to the philosophy and weren’t willing to accept it until much later in the illness.
Perceived optimal length of stay surveys
Christakis is now doing preliminary research to test this hypothesis. He has also surveyed internists and hospice professionals to determine what they feel is the optimal length of stay for taking full advantage of hospice services.
Christakis is now studying the disparity between an ideal of 80 to 90 days and observed reality of 30 days. Given that "the time of enrollment is picked by human decision making, it might be possible to modify the survival curve of enrolled patients by changing patient, physician, and hospice provider behavior."
Prognosis is regularly identified as a major barrier to hospice referral by both physicians and hospices, so Christakis is examining prognostic accuracy in five Chicago hospice programs. There are different ways of defining a six-month prognosis, which produce very different results. He also challenges the audits of long-stay hospice patients by the federal Office of Inspector General (OIG) on methodologic grounds. OIG auditors’ conclusion that many of these patients failed to have an appropriate, documented terminal prognosis was not justified because they failed to compare them to the charts of short-stay patients, Christakis said.
"There are only three ways you can live too long after hospice enrollment," he said. These ways are: fraud, receiving excellent life-prolonging care, and prognostic error. "How do we separate rectifiable prognostic error from prognostic error that’s unrectifiable?" The government has also failed to specify what is an appropriate threshold for accurate prognostication, Christakis charged.
Christakis recommended a few things hospices can do to encourage earlier referrals, such as recognizing that palliative and curative care are not mutually exclusive, clarifying patient goals, increasing physician education and training, developing and using prognostic systems, and establishing new financial incentives under managed care.
Reference
1. Christakis NA, Escarce J J. Survival of Medicare patients after enrollment in hospice programs. NEJM 1996; 335(3): 172-178.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content