Employing the perspective and methods of the demography of aging, the investigators analyze the use and consequences of hospice terminal care in seriously ill elderly patients and their spouses. The benefits of hospice terminal care are well documented. Too little is known, however, about the factors that affect whether and when elderly patients are enrolled in hospice programs. Moreover, important questions about the effects of hospice on the mortality and morbidity of elderly patients and their families remain unanswered. These are serious deficits in knowledge because sub-optimal use of hospice might decrease the quality and increase the cost of terminal care in the U.S. The proposed research involves the creation of a multi-purpose data set containing clinical, demographic, socioeconomic, and market information about a population-based cohort of approximately 772,000 seriously ill elderly patients at risk for hospice referral; using a new technique, similar information will also be developed about their spouses (where possible). Data will be extracted and linked, on an individual level, from Medicare claims data (Standard Analytic File, Provider of Services File, MEDPAR Files, National Claims History Files, and Vital Status Files), Census data, and the Area Resource File. There will be three primary analyses. The first is the first-ever denominator-based study of hospice use.
Its aim i s to evaluate patterns of hospice use in a population-based sample of terminally ill patients that is developed at the time of diagnosis rather than at the time of hospice entry; using Cox regression models, it will evaluate whether and when in the course of serious illness hospice is used as well as the patient, provider, and market factors associated with the timing of use. The second is a retrospective cohort study of the effect of hospice on mortality.
Its aim i s to test the hypothesis that hospice use -- and the switch from curative to palliative care that it ordinarily implies -- is not life-shortening, and possibly even life-prolonging. This analysis will be conducted using several complementary approaches, including Cox regression, matched case/comparison methods, and instrumental variable methods. The third is a study of the effect of hospice use by patients on the morbidity, mortality, and health care consumption of their bereaved spouses.
Its aim i s to demonstrate that hospice care for dying patients may improve the health status of their bereaved spouses (attenuating the so-called """"""""widow/widower effect""""""""). Our work expands the field of the demography of aging in several ways, by: examining factors that might alter the life course at its very end, developing new methods to identify spousal pairs in Medicare claims data, extending social support theory by applying it to end-of-life care, and examining the elderly's access to hospice (a unique form of medical care). The work has policy implications as well: demonstrating that hospice does not meaningfully shorten life, or that it confers benefits to survivors, or both, might increase patient and physician interest in hospice, especially in specific clinical and market situations that we will be able to identify.
Showing the most recent 10 out of 11 publications
