2009 HSR&D National Meeting Abstract
1063 — Do Living Wills Work in Practice?
Silveira MJ (AA HSRD COE), Kim S
(University of Michigan), Langa K
(AA HSRD COE)
1. To describe the frequency of lost decision making capacity and end of life decision making; 2. To describe the prevalence and content of living wills; 3. To determine the predictors of preferences; 4. To determine how often the outcomes of end of life decision making reflect the preferences stated in living wills.
We used data from the Health and Retirement Study (HRS), a biennial longitudinal survey of a nationally representative cohort of elderly adults regarding their medical, social, and financial situation. Our subjects included those who died between 2000-2006 and were 60 or older at time of death. Our variables came from answers given by proxies to exit interviews regarding the decedent’s circumstances at death. We used descriptive statistics to report the prevalence and content of living wills, prevalence of decision making at the end of life, and prevalence of decision making capacity. We used multivariable logistic regression to explore the association between these outcomes and subjects’ sociodemographic and clinical characteristics. We examined the association between preferences for treatment as stated in the living will and outcomes of decision making using Mcnemar’s test. We accounted for the complex sampling design of the HRS in all analyses.
Our study population included 3,746 elderly adults. Their completion of living wills increased throughout the study period from 30% in 2000 to 49% in 2006. Living wills were used to request comfort care (1,403, 91%) and limit aggressive treatment (1,462, 89%) more often than request all care possible (563, 4%). African Americans were more likely to request all care possible (OR 16.4, P = 0.000) and less likely to limit life-sustaining treatment (OR 0.20, P = 0.000) than their white counterparts. Of 3,746 decedents, 1,536 (41%) required decision-making about treatment prior to death. Among those subjects, 999 (65%) required others to make those decisions for them. Subjects who required decision making and had authored living wills received all care possible 5% more often (P = 0.000) and limited treatment 11% less often (P = 0.000) than they requested in their living wills.
Increasingly, elderly adults are writing living wills. Living wills are often put to use before death, as many elderly adults lose decision making capacity before death while requiring decision making about treatment. Living wills are usually used to request limited or no aggressive treatment; these preferences are honored most of the time.
Elderly adults are well advised to compose living wills.