Use of Advance Directives among Older U.S. Adults by Dementia Status: 2012

Abstract

Dear Editor:

Individuals with dementia face unique challenges in medical decision making due to the difficulty in understanding complex medical information and articulating preferences for care. Advance directives (ADs), in which the patient appoints a surrogate decision maker known as durable power of attorney (DPA) and/or records their wishes on medical treatment in advance in a living will (LW), have been shown to be associated with receiving preference-concordant care, particularly in the dementia population.^1,2 This study examines trends in ADs use among older U.S. adults by dementia status from 2012 to 2016.

We used data from the Health and Retirement Study (HRS), a longitudinal study that interviews a nationally representative sample of ∼20,000 U.S. adults aged 50 years and above or their proxies every 2 years. Our study population consists of adult participants aged 65 years in the 2012, 2014, or 2016 HRS surveys. Outcome measures included presence of two forms of ADs: DPA and written LW, available in HRS starting in 2012. Probable dementia was determined using a validated algorithm based on demographics and cognitive assessment information in the HRS interviews.³ This classification algorithm has been shown to maximize accuracy of dementia classification compared with other competing methods.⁴ We estimated logistic regressions for each outcome measure, adjusting for age, gender, race/ethnicity, and year of survey, stratified by probable dementia and nondementia respondents.

After adjusting for age, gender, and race/ethnicity, the use of both forms of ADs increased from 2012 to 2016, for both respondents with and without probable dementia (Table 1). For these with probable dementia, the change from 2012 to 2014 for each form of AD (DPA: odds ratio [OR] 1.47 [95% confidence interval (CI), 1.22–1.78]; LW: OR 1.48 [95% CI, 1.23–1.78]) and having any ADs (OR 1.57 [95% CI: 1.28–1.92] were statistically significant from 2012 to 2014, and was significant from 2014 to 2016. Among those without dementia, both the change in ADs use from 2012 to 2014 and the change from 2014 to 2016 were significant for all outcome variables, with somewhat lower OR from 2014 to 2016 (DPA: OR 1.19 [95% CI: 1.12–1.27]; LW: OR 1.14 [95% CI: 1.07–1.22]; any ADs: OR 1.14 [95% CI: 1.07–1.22]) compared with from 2012 to 2014 (DPA: OR 1.33 [95% CI: 1.25–1.42]; LW: OR 1.25 [95% CI: 1.18–1.33]; any ADs: OR 1.24 [95% CI: 1.16–1.32). In 2016, the probability of having either form of ADs and having any ADs was >50%, for both individuals with and without dementia, although the probability of LW was lower among those with probable dementia (53.2% [95% CI: 46.5–57.0]) than those without dementia (58.5% [95% CI: 57.3–59.7]).

Table 1.

Use of Advance Directives, 2012–2016

	Adjusted probability, % (95% CI) ^a			Change from 2012 to 2014, OR (95% CI) ^b	Change from 2014 to 2016, OR (95% CI) ^c
	2012	2014	2016
	n = 10,402	(n = 10,098;	(n = 9685;
	No. of probable dementia (%) = 1136 (10.9)	No. of probable dementia (%) = 1081 (10.7))	No. of probable dementia (%) = 971 (10.0))
Has DPA
Probable dementia	55.2 (51.5–59.0)	63.5 (59.9–67.1)	65.3 (61.7–69.0)	1.47 (1.22–1.78)	1.09 (0.89–1.34)
No dementia	50.9 (49.7–52.1)	57.7 (56.5–58.9)	61.6 (60.4–62.8)	1.33 (1.25–1.42)	1.19 (1.12–1.27)
Has LW
Probable dementia	41.6 (38.0–45.3)	50.3 (46.5–54.0)	53.2 (49.4–57.0)	1.48 (1.23–1.78)	1.14 (0.94–1.39)
No dementia	50.4 (49.2–51.6)	55.5 (54.3–56.7)	58.5 (57.3–59.7)	1.25 (1.18–1.33)	1.14 (1.07–1.22)
Any ADs
Probable dementia	58.9 (55.2–62.6)	67.9 (64.4–71.3)	70.2 (66.8–73.7)	1.57 (1.28–1.92)	1.14 (0.92–1.41)
No dementia	60.2 (59.0–61.4)	64.9 (63.8–66.1)	67.7 (66.6–68.8)	1.24 (1.16–1.32)	1.14 (1.07–1.22)

Each row presents results from a separate logistic regression, with a given form of ADs (or any ADs) as the dependent variable, and age, gender, race/ethnicity, and year as independent variables. Adjusted probabilities were calculated as predicted probabilities of the outcome variable averaged across respondents in the study population median age category (75–84), for each year and by dementia status.

ORs reported were inverses of the ORs for year 2012, with year 2014 as the omitted category.

ORs reported were ORs for year 2016, with year 2014 as the omitted category.

ADs, advance directives; CI, confidence interval; DPA, durable power of attorney; LW, living will; OR, odds ratio.

We found an overall increase in use of ADs in U.S. older adults with and without probable dementia from 2012 to 2016. Medicare began reimbursing for advance care planning in 2016, and comprehensive care planning service in 2017 for cognitively impaired patients. It is important to examine whether these policies resulted in further increases of AD use among older adults, especially for LW among those with dementia, which we plan on studying in future research.

Footnotes

Acknowledgments

This work was supported by funding from the Weill Cornell Medicine JumpStart Research Career Development Program (JL).

References

Luth

, Prigerson

: Associations between race and dementia status and the quality of end-of-life care. J Palliat Med, 2018; 21:970–977.

Nicholas

, Bynum

, Iwashyna

, et al.: Advance directives and nursing home stays associated with less aggressive end-of-life care for patients with severe dementia. Health Affairs, 2014; 33:667–674.

, Tchetgen

EJT

, Osypuk

, et al.: Combining direct and proxy assessments to reduce attrition bias in a longitudinal study. Alzheimer Dis Assoc Disord, 2013; 27:207.

Gianattasio

, Wu

, Glymour

, Power

: Comparison of methods for algorithmic classification of dementia status in the Health and Retirement Study. Epidemiology, 2019; 30:291.

Use of Advance Directives among Older U.S. Adults by Dementia Status: 2012–2016

Abstract

Footnotes

Acknowledgments

References