THE BROWN ATLAS OF DYING
in the UNITED STATES: 1989-2001

With a grant from the Robert Wood Johnson Foundation, the Center for Gerontology and Health Care Research at the Brown Medical School has initiated a five-year research effort to describe dying in the United States. In the first year, the research team focused their efforts on describing the variation and changes in the site of death in the United States. An important conclusion from our research is that where you die varies substantially based on where you live. "Geography is destiny." We found that dying at home varied from 18.4% to 35.8% from one region of the country to another. However, a 1997 Gallup Poll found that majority of Americans would want to die at home if they were terminally ill. This preference did not vary by region of the country. Thus, while our results cannot determine what the optimal rate of home death is, we do know that it varies in ways that peoples' preferences do not.

The proportion of people dying in a nursing home, home, or hospital in any state or local area probably raises more questions than this Atlas is able to answer. For example, the fact that one in three persons in the state of Rhode Island die in a nursing home may not be a bad thing, if patient preferences are being honored and they receive the appropriate palliative care in that setting. The call to action of these data is to review the distribution in each state and to ask some hard questions about the quality of care.

• Given that an increasing number of persons are dying in nursing homes, are those institutions able to provide holistic and compassionate end of life care? Consider asking your state health department, what does the Minimum Data Set show about pain management in your state's nursing homes?

• If only a small percentage of people are dying at home, is the availability of hospice and home health services less than might be the case in other states?

In 2001 there was more than 2 fold variation in those dying at home, even after adjustment for age and gender. Between 1989 and 2001, there was a striking increase in the percent of persons who died in a nursing home. We found that the proportion of deaths in nursing home increased from 19.2% to 23.2% (a 21% increase) with the proportion of persons dying in a nursing home varying from 11% (in Hawaii) to 35.3% (in Minnesota).

The Brown Atlas of Site of Death examines variation based on both states and counties that have more than 100,000 residents. Further research will examine the relationship of changes in nursing home bed capacity, state Medicaid reimbursement, physician supply, and hospital bed supply with the changes in dying in a nursing home observed between the years1989, 1997, and 2001.

A technical appendix outlines the methodology used to generate the maps and the adjustment for age and gender.