Log in to MyACC
Advance Care Planning Among Older Adults in Low SES Neighborhoods
Quick Takes
- Using electronic health record (EHR) and place-based (i.e., geocoded) data, investigators were able to identify socioeconomically disadvantaged communities of older adults who had both low neighborhood socioeconomic status (SES) and low advance care planning.
- This study highlights the need for targeted neighborhood-level advance care planning interventions, especially among older adults living in disadvantaged neighborhoods.
Study Questions:
What is the association between neighborhood SES and advance care planning, and are there communities with both low neighborhood SES and low rates of advance care planning?
Methods:
All available data on empaneled patients 65 years of age and older at the University of California San Francisco who resided in 9 Bay Area counties were analyzed. Empaneled patients had a primary care provider and at least 1 in-person or patient portal encounter during the previous 3 years. Patient addresses (n = 13,104) were geocoded (longitude and latitude). To ensure consistency in medical care and patient health care goals, only documented forms of advance care planning were used to measure the advance care planning outcome variable. The primary factor of neighborhood SES was an index of a combination of area-level measures, including income, education, poverty, employment, occupation, and housing or rent values, divided into quintiles scaled to the distribution of all US Census tracts in the Bay Area. Health care use co-variates were gleaned from patients’ EHRs and included primary care, outpatient specialty, emergency department, and inpatient encounters during the previous year.
Results:
Data were analyzed on patients who were, on average, age 75 ± 8 years; female (58.2%); and Black (6.8%), White (40.8%), Latinx (7.0%), Asian/Pacific Islander (28.9%), or other ethnic minority (5.7%) who preferred to speak a non-English language. Only 3,827 (29.2%) patients had a documented advance care plan. Patients lived in US Census tracts from the lowest neighborhood SES (quintile 1) to the highest neighborhood SES (quintile 5). Patients living in US Census tracts with lower neighborhood SES (quintiles 1-4) had lower odds of advance care planning documentation compared with those living in US Census tracts with the highest neighborhood SES (quintile 5); quintiles 3 and 4 were the only quintiles that were statistically significant (odds ratio 0.86; 95% confidence interval, 0.75-0.98 and odds ratio 0.86; 95% confidence interval, 0.76-0.98, respectively). After adjusting for health care use co-variates, lower quintiles of neighborhood SES were associated with progressively lower odds of advance care planning documentation (p < 0.001).
Conclusions:
Compared with patients living in the highest SES neighborhoods, those living in lower SES neighborhoods had lower odds of advance care planning documentation. In fact, patients living in the lowest neighborhood SES quintile had 29% lower odds of advance care planning documentation. After controlling for health care use, lower neighborhood SES was associated with lower odds of advance care planning documentation. This is an important finding because advance care planning is primarily introduced in clinical settings and to persons with acute or progressive illness, leaving many community-dwelling older adults with little or no opportunity for advance care planning documentation.
Perspective:
Advance care planning is important to ensure that older adults are receiving care consistent with their goals, but advance care planning rates are low (20-30%) among disadvantaged socioeconomic populations compared with the general population of older adults. After adjusting for health care use, lower neighborhood SES was associated with lower advanced care planning documentation in the EHR. In the absence of health care use and other system-level models of advance care plan engagement, insufficiencies in advance care planning will most likely persist. Further evaluation using novel geocoding techniques are needed, as well as testing of community-based targeted interventions to ensure that disadvantaged older adults living in low-income and under-resourced neighborhoods have opportunities for advance care planning.
Clinical Topics: Geriatric Cardiology, Prevention
Keywords: Advance Care Planning, Documentation, Goals, Social Class, Aged, Vulnerable Populations, Poverty Areas, Electronic Health Records, Geographic Mapping, Censuses
< Back to Listings
