Prevalence, effectiveness, and predictors of planning the place of death among older persons followed in community-based long term care

BACKGROUND: Little is known about whether patients plan for the site of their death and whether such planning is effective. OBJECTIVE: To determine the prevalence, effectiveness, and predictors of planning the place of death among older homebound persons followed in a community-based, physician-led house call program. DESIGN: Retrospective chart review. SETTING: A geographically defined catchment area in southeast Baltimore, Maryland. PATIENTS: One hundred twenty-five patients who died between July 1995 and November 1998 who were followed in a physician-led house call program. MAIN OUTCOME MEASURES: Presence of a plan to die in a specific place and concordance between planned and actual place of death. RESULTS: Eighty patients (64%) made a plan to die in a specific place, and these plans were executed successfully in 73 cases (91%). The median time between formulating a plan to die in a specific place and death was 36 days. In logistic regression analysis, making a plan to die in a specific place was positively associated with an advance directive of Do Not Resuscitate (DNR) (odds ratio (OR) 11.7, confidence interval (CI) 3.7, 32.5) and negatively associated with the lack of an identifiable main medical problem other than being homebound (OR 0.17; CI, 0.02-0.88). CONCLUSIONS: Among a group of frail older persons living in the community, planning to die in a particular place was common and implemented successfully most of the time. Providing physician care at home may facilitate improved end-of-life care for older persons.