Advance Care Planning and Hospice Utilization for People with Dementia: A Report from the Health and Retirement Survey

Objective: People with memory related diseases (PWMRD) such as Alzheimer’s disease and dementia frequently receive sub-optimal end-of-life care (EOLC), often enduring invasive procedures such as tube feeding, resuscitation, and surgery within days of their death. While early advance care planning (ACP) has shown effectiveness in improving EOLC for those with memory related diseases (MRD), there are many barriers to early ACP including failure to recognize dementia as a terminal illness, physician education, comfort levels, and time. Research suggests that a palliative approach to advanced MRD is the best way to minimize suffering and superfluous medical interventions and promotes hospice care at end of life (EOL). To improve quality EOLC among PWMRD, it is important to understand the prevalence of ACP and hospice utilization within this population. This study aims to empirically assess hospice utilization and ACP for PWMRD compared to those without MRD, and to assess the impact of ACP on hospice utilization for PWMRD.

Methods: Data for this analysis came from the 2016 Health and Retirement Exit Survey, a section of the Health and Retirement Study (HRS). The HRS is a national longitudinal study collecting data on health, demographics, and support systems of older Americans. This survey interviewed proxies for 1,310 individuals who died during this survey wave. The dependent variables were having a written EOLC plan (Yes=1 and No=0), having discussed EOLC preferences (Yes=1 and No=0), and hospice utilization (Yes=1 and No=0). The independent variable was diagnosis of MRD (Yes=1 and No=0). We conduct logistic regression to examine the relationships of dependent variables with MRD, adjusting for control variables.

Results: 83% of PWMRD had a diagnosis of dementia or Alzheimer’s Disease. About half of the sample utilized hospice services (51%) and 47% of the sample had written EOLC plans. 59% of the sample reported discussion around EOLC preferences. People with MRD were both more likely to have utilized hospice-care (OR=1.37; p<.05) and to have a written EOLC plan (OR=1.19), though the relationship between having MRD and a written EOLC plan was not statistically significant. Among control variables, age was a confounding factor and older age significantly increased the odds of both utilizing hospice and having a written EOLC plan. Individuals with MRD were 22% less likely to have discussed their EOL wishes with their proxy than those without MRD. Having a written EOLC plan significantly increased the odds of hospice utilization (OR=1.37) but having a discussion around EOLC preferences increased the odds of hospice utilization at a higher rate (OR=1.59).

Conclusions and Implications: Individuals with MRD utilized hospice and had written EOLC plans more frequently than those without MRD. However, discussion of EOLC preferences was lower for people with MRD than those without MRD, implying that written EOLC plans may have been initiated by someone other than the patient. As discussion around EOLC preferences increased the odds of receiving hospice at a higher rate than having a written EOLC plan, these results support the importance of earlier conversations around EOLC preferences with those diagnosed with MRD.