The value of hospice care in improving symptom management, quality of life, and satisfaction with care at the end of life has been shown in several studies. However, data indicate that most people with advanced cancer likely to benefit from hospice care do not utilize hospice services, and those who do typically begin hospice care too close to the time of death to receive the maximum benefits of this care. To improve quality of life for people dying of cancer, it is therefore important to understand factors that influence decisions about whether and when to receive hospice care. 

thumbnail image

Figure  . Recent report indicates Medicare Managed Care plan patients more likely to receive hospice care.

Download figure to PowerPoint

A recent report in JAMA (2003;289:2238–2245) demonstrates that patients dying of cancer who are enrolled in Medicare Managed Care plans (MMC) are more likely to receive hospice care than are those with fee-for-service Medicare insurance (MFFS).

Ellen P. McCarthy, PhD, MPH, and colleagues from Harvard Medical School and the Beth Israel Deaconess Medical Center investigated this association in the Surveillance, Epidemiology, and End Results (SEER)-Medicare linked database. This linked database represents a collaboration of the National Cancer Institute and Center for Medicare and Medicaid Services, and includes patients' clinical and demographic information as well as provider characteristics and utilization data. Using the SEER-Medicare database, the researchers examined records of 260,090 Medicare beneficiaries who died of lung, colorectal, prostate, female breast, bladder, pancreatic, gastric, or liver cancer between 1988 and 1998.

During this period, 32.4% of MMC patients and 19.8% of MFFS patients received hospice care. The duration of hospice care was a week longer for MMC patients than MFFS; median lengths of stay were 32 days and 25 days, respectively. These associations remained statistically significant in multivariate analyses that included patient demographic data, stage at diagnosis, and geographic location.

One important question to ask is whether the additional hospice utilization among MMC patients represents appropriate or inappropriate referrals. Although the size and methodology of this study precludes review of medical records, the SEER-Medicare database does provide some data on which to address this question. For example, the association between hospice care and insurance type was strongest among patients with Stage IV cancer.

The duration of hospice care is another variable that sheds some light on appropriateness of referrals. According to the authors, periods of 2 to 3 months before death have been recommended as appropriate periods of hospice care. Much shorter or much longer durations are less likely to provide optimal care. More MFFS than MMC patients were referred to hospice within 1 week of death (22.6% and 18.6%, respectively); in many cases this probably did not provide enough hospice care. On the other hand, MMC patients were also more likely than MFFS patients to start hospice care at least 180 days before their death (7.8% versus 6.1%), suggesting referral too early in at least some of these cases.

Not surprisingly, the authors suggest that financial incentives are a major reason for the differences in hospice referral patterns; once an MMC plan refers a patient to hospice it no longer incurs costs for remaining end-of-life care. However, the authors suggest that other factors may also contribute to the results. Selection of MMC versus MFFS insurance by patients is not random. Patients with a preference for more intensive care might be likely to prefer MFFS insurance and might be less likely to choose hospice care.

ACS President Mary A. Simmonds, MD, agrees that far too few individuals take advantage of hospice care. She notes that those who do have reported better management of their symptoms and more satisfaction with their care compared with patients cared for in the traditional medical system, who undergo more diagnostic studies and procedures and receive therapy that may be futile.

“Hospice care, therefore, happens to be more cost-effective,” explains Simmonds. “The study published in JAMA is the first large, comprehensive, population-based examination of the utilization of hospice services. It demonstrates that there is variation in utilization and that there are a number of factors, including financial incentives, which influence the initiation of hospice care. It will be important to understand these factors so that providing hospice care can be incentivized and so that education can be directed to encourage utilization of hospice services.”