Documentation of advance directives among home health and hospice patients: United States, 2007

The Impact of Socioeconomic Status on the Presence of Advance Care Planning Documents in Patients With Acute Cervical Spinal Cord Injury

INTRODUCTION

Patients presenting with spinal cord injury (SCI) often times have notable deficits or polytrauma and may require urgent decision making for early management. However, their presentation may affect decision-making ability. Although advance care planning (ACP) may help guide spine surgeons as to patient preferences, the rate at which they are available and disparities in ACP completion are still not understood. The objective of this study was to evaluate disparities in the completion of ACP among patients with acute SCI.

METHODS

All patients presenting with cervical SCI to the emergency department at an urban, tertiary level I trauma center from 2010 to 2021 were identified from a prospective database of all consults evaluated by the spine service. Each patient's medical record was reviewed to assess for the presence of ACP documents such as living will, power of attorney, or advance directive. Community-level socioeconomic status was assessed using the Distressed Communities Index. Bivariable and multivariable analyses were performed.

RESULTS

We identified 424 patients: 104 (24.5%) of whom had ACP. Patients with ACP were older (64.8 versus 56.5 years, P = 0.001), more likely White (78.8% versus 71.9%, P = 0.057), and present with ASIA Impairment Scale grade A SCI (21.2% versus 12.8%, P = 0.054), although the latter two did not reach statistical significance. On multivariable logistic regression, patients residing in at-risk communities were significantly less likely to have ACP documents compared with those in prosperous communities (odds ratio [OR]: 0.29, P = 0.03). Although patients living in distressed communities were less likely to complete ACP compared with those in prosperous communities (OR 0.50, P = 0.066), this did not meet statistical significance. Female patients were also less likely to have ACP (OR: 0.43, P = 0.005).

CONCLUSION

Female patients and those from at-risk communities are markedly less likely to complete ACP. Attention to possible disparities during admission and ACP discussions may help ensure that patients of all backgrounds have treatment goals documented.

Comparing two advance care planning conversation activities to motivate advance directive completion in underserved communities across the USA: The Project Talk Trial study protocol for a cluster, randomized controlled trial

BACKGROUND

Advance care planning (ACP) is a process involving conversations between patients, loved ones, and healthcare providers that consider patient preferences for the types of medical therapies received at the end of life. Underserved populations, including Black, Hispanic, rural, and low-income communities are less likely to engage in ACP than other communities, a health inequity that results in lower-quality care and reduced hospice utilization. The purpose of this trial is to compare efficacy of two interventions intended to motivate ACP (particularly advance directive completion) for those living in underserved communities.

METHODS

This 3-armed cluster, randomized controlled mixed methods design is being conducted in 75 community venues in underserved communities across the USA. The goal of the trial is to compare the efficacy of two interventions at motivating ACP. Arm 1 uses an end-of-life conversation game (Hello); Arm 2 uses a nationally utilized workshop format for ACP conversations (The Conversation Project); and Arm 3 uses an attention control game (TableTopics). Events are held in partnership with 75 local community-based host organizations and will involve 1500 participants (n=20 per event). The primary outcome is completion of a visually verified advance directive at 6 months post-event. Primary analyses compare efficacy of each intervention to each other and the control arm. Secondary mixed methods outcomes include (a) other ACP behaviors and engagement; (b) communication quality; (c) impact of sociocultural environment on ACP (via qualitative interviews); and (d) implementation and sustainability. Subgroup analyses examine outcomes for Black, Hispanic, and rural groups in particular.

DISCUSSION

This trial will add to the evidence base behind various conversational ACP interventions, examine potential mechanisms of action for such interventions, and provide qualitative data to better understand the sociocultural environment of how community-based ACP interventions are experienced by underserved populations. Results will also provide important data for future researchers to learn whether visual verification of advance directives is necessary or whether reliance on self-reported outcomes is of comparable value. Data from this study will inform ways to effectively motivate underserved communities to participate in advance care planning.

TRIAL REGISTRATION

ClinicalTrials.gov NCT04612738. Registered on October 12, 2020. All information from the WHO Trial Registration Data Set can be found within the protocol.

Intergenerational Decision Making: The Role of Family Relationships in Medical Decision Making

A symposium held at the 42nd annual Society for Medical Decision Making conference on October 26, 2020, focused on intergenerational decision making. The symposium covered existing research and clinical experiences using formal presentations and moderated discussion and was attended by 43 people. Presentations focused on the roles of pediatric patients in decision making, caregiver decision making for a child with complex medical needs, caregiver involvement in advanced care planning, and the inclusion of spillover effects in economic evaluations. The moderated discussion, summarized in this article, highlighted existing resources and gaps in intergenerational decision making in four areas: decision aids, economic evaluation, participant perspectives, and measures. Intergenerational decision making is an understudied and poorly understood aspect of medical decision making that requires particular attention as our society ages and technological advances provide new innovations for life-sustaining measures across all stages of the lifespan.

Low Skepticism and Positive Attitudes About Advance Care Planning Among African Americans: a National, Mixed Methods Cohort Study

BACKGROUND

African Americans have low engagement in advance care planning (ACP). This has been attributed to healthcare distrust and skepticism about ACP. A better understanding of these attitudes is needed to address health disparities related to end-of-life care.

OBJECTIVE

To explore the ACP-related values and beliefs of diverse African American communities across the USA and then the perceived value of an inexpensive end-of-life conversational game.

DESIGN

Prospective, convergent, mixed methods cohort study involving fifteen underserved, African American communities across the USA.

PARTICIPANTS

Of the 428 who attended events at purposively sampled sites, 90% consented to the research; 37% participated in one of 15 focus groups (n = 141).

INTERVENTION

An end-of-life conversation game, played in groups of 4-6.

MAIN MEASURES

The validated, 7-item ACP values and beliefs questionnaire (scaled 7 = least skeptical, 49 = most skeptical) was administered pre-game. Post-game focus groups explored perceptions about ACP and the intervention.

KEY RESULTS

Participants had positive attitudes (low skepticism) about ACP with a median score of 12.00 (7.00, 20.00). Values and beliefs did not significantly differ by geographical region; however, rural areas were observed to be slightly more skeptical than urban areas (median score 14.00 vs. 11.00, p = 0.002). Themes from focus groups converged with survey data showing participants valued the ACP process and consider further engagement in ACP to be worthwhile. Subthemes emphasized the need for and value of ACP.

CONCLUSIONS

Skepticism about ACP may contribute to low rates of ACP engagement in underserved African American communities. The positive attitudes uncovered in our study either negate previous findings or suggest reduced skepticism.

TRIAL REGISTRATION

This study has been registered at clinicaltrials.gov ( NCT03456921 ).

"It's Case by Case, and It's a Struggle": A Qualitative Study of Hospice Practices, Perspectives, and Ethical Dilemmas When Caring for Hospice Enrollees with Full-Code Status or Intensive Treatment Preferences

Objective: Characterize hospice staff practices and perspectives on discussing end-of-life care preferences with patients/families, including those desiring intensive treatment and/or full code. Background: Patients in the United States can elect hospice while remaining full code or seeking intensive interventions, for example, blood transfusions, or chemotherapy. These preferences conflict with professional norms, hospice philosophy, and Medicare hospice payment policies. Little is known about how hospice staff manage patient/family preferences for full-code status and intensive treatments. Methods: We recruited employees of four nonprofit US hospices with varying clinical and hospice experience for semi-structured, in-depth interviews. Open-ended questions explored participants' practices and perceptions of discussing end-of-life care preferences in hospice, with specific probes about intensive treatment or remaining full code. Interdisciplinary researchers coded and analyzed data using the constant comparative method. Results: Participants included 25% executive leaders, 14% quality improvement administrative staff, 61% clinicians (23 nurses, 21 social workers, 7 physicians, and 2 chaplains). Participants reported challenges in engaging patients/families about end-of-life care preferences. Preferences for intensive treatment or full-code status presented an ethical dilemma for some participants. Participants described strategies to navigate such preferences, including educating about treatment options, and expressed diverse reactions, including accepting or attempting to shift enrollee preferences. Discussion: This study illuminates a rarely studied aspect of hospice care: how hospice staff engage with enrollees choosing full code and/or intensive treatments. Such patient preferences can produce ethical dilemmas for hospice staff. Enhanced communication training and guidelines, updated organizational and federal policies, and ethics consult services may mitigate these dilemmas.

Association of Participation in an End-of-Life Conversation Game With Advance Care Planning Behavior and Perspectives Among African American Individuals

Importance

Less than 25% of African American individuals have completed advance directives and are thus vulnerable to poor end-of-life care. Low-cost interventions are needed to increase engagement in advance care planning (ACP).

Objectives

To investigate whether an end-of-life conversation game motivates African American attendees to engage in ACP and to assess whether the game is well received and endorsed.

Exposures

Attendance at an end-of-life conversation game (Hello) played in groups of 4 to 6 participants for 60 minutes.

Design, Setting, and Participants

Prospective, mixed-methods cohort study conducted from 2018 to 2019 with a 3- to 11-month follow-up interview. Game events were held in 53 community venues across the US; 15 were purposively sampled for onsite research procedures. Of 428 attendees at purposively sampled sites, 386 (90%) consented to research procedures (6 attendees were removed from analysis for protocol deviation). Of 367 attendees who provided accurate contact information, 232 (63%) were contacted, and 220 were included in follow-up analyses.

Main Outcomes and Measures

The primary outcome was advance directive completion rates after the intervention. Secondary outcomes included rates of other ACP behaviors, ACP engagement, conversation satisfaction and realism, and participants' Net Promoter Score (a measure of endorsement). Follow-up telephone interviews explored the game experience and relevant ACP behaviors of attendees.

Results

Of 380 individuals who participated (mean [SD] age, 62.2 [13.8] years; 304 were female [80%], and 348 were [92%] African American), none withdrew because of an adverse event. After the intervention, 91 of 220 attendees (41%) completed a new advance directive; 176 of 220 attendees (80%) discussed end-of-life wishes with loved ones, and 214 of 219 attendees (98%) completed at least 1 ACP behavior. There was a moderate increase in the self-efficacy domain on the ACP Engagement Survey (mean [SD] change from before to after the game, 0.54 [0.98]; P < .001). The mean (SD) conversation satisfaction score was 6.21 (0.93) (range, 1-7, with 7 being highest satisfaction), and the overall Net Promoter Score was 57.89 (range, -100 to 100, with 100 being highest endorsement). Interviews revealed 5 themes about the game: (1) it was a useful forum for ACP; (2) it provided new information and perspective; (3) it was emotionally beneficial; (4) it increased appreciation for ACP; and (5) it empowered and motivated participants to perform ACP. Mixed-methods integration showed convergence across data sets.

Conclusions and Relevance

Among a nationwide sample of African American individuals, the end-of-life conversation game appeared to be well received and was associated with high rates of ACP behavior. This low-cost and scalable tool may help reduce health disparities associated with end-of-life care.

Adaptation and Preliminary Validation of the Advance Care Planning Engagement Survey for Surrogate Decision Makers

CONTEXT

Although measures that assess patient engagement in the advance care planning (ACP) process exist, there are no validated tools to assess surrogate decision makers' (SDMs') role in ACP.

OBJECTIVES

The objective of this study was to adapt and begin to validate a patient-oriented questionnaire for use with SDMs of patients with chronic illness.

METHODS

Questions from the 55-item patient-oriented ACP engagement survey were adapted for SDMs and assessed for face validity. The resultant 47-item questionnaire was administered to 65 SDMs of patients with chronic illness. Responses were assessed and items were flagged for removal based on item redundancy, nonresponses, and ceiling effects. A preliminary exploratory factor analysis was performed, internal consistency was assessed, and domains were constructed based on findings.

RESULTS

The 47-item questionnaire was administered to 65 participants (mean age 51.8; 81% female; 96% Caucasian). Seventeen items were removed owing to redundancy (r > 0.80), and 13 items lacking face validity were removed. In a preliminary exploratory factor analysis of the resultant 17-item questionnaire, a three-factor solution was deemed most statistically and conceptually sound. Items were organized into domains: 1) serving as an SDM (seven items); 2) contemplation (four items); 3) readiness (six items). Internal consistency for each domain was high (Cronbach alpha 0.90-0.91).

CONCLUSION

The 17-item ACP engagement survey for SDMs (ACP-17-SDM) is a conceptually sound and reliable questionnaire adaptation of the original ACP engagement survey. This questionnaire may be used by researchers in parallel with the patient-oriented ACP engagement survey to more fully understand the impact of ACP interventions on SDMs. Larger studies are needed to more closely examine construct validity.

Health-Related Decisions for Serious Illness Among Ethnically Diverse Older Adults

We investigated the influence of sociodemographic factors, acculturation, ethnicity, health status, and spirituality on older adults' health-related decisions when confronted with a choice between competing options. The sample included 451 participants: African Americans (15.74%), Afro-Caribbeans (25.5%), European Americans (36.36%), and Hispanic Americans (22.4%). Compared with others, European Americans and Hispanic Americans favored quality of life over a lengthy life. Sociodemographic factors, acculturation, ethnicity, health status, and spirituality accounted for variations of decisions. The variability of decisions calls for multiple care options to explore the value of different trade-offs in order to avoid predetermined clinical practice guidelines, especially in nursing.

Community Game Day: Using an End-of-Life Conversation Game to Encourage Advance Care Planning

CONTEXT

Advance care planning (ACP) is an important process that involves discussing and documenting one's values and preferences for medical care, particularly end-of-life treatments.

OBJECTIVES

This convergent, mixed-methods study assessed whether an end-of-life conversation card game is an acceptable and effective means for performing ACP for patients with chronic illness and/or their caregivers when deployed in a community setting.

METHODS

Twenty-two games (n = 93 participants) were held in community settings surrounding Hershey, PA in 2016. Participants were recruited using random sampling from patient databases and also convenience sampling (i.e., flyers). Quantitative questionnaires and qualitative focus group interviews were administered to assess the game experience and subsequent performance of ACP behaviors.

RESULTS

Both quantitative and qualitative data found that Community Game Day was a well-received, positive experience for participants and 75% of participants performed ACP within three months post-intervention.

CONCLUSIONS

These findings suggest that using a conversation game during community outreach is a useful approach for engaging patients and caregivers in ACP. The convergence of quantitative and qualitative data strongly supports the continued investigation of the game in randomized controlled trials.

Electing Full Code in Hospice: Patient Characteristics and Live Discharge Rates

BACKGROUND

It is unknown how many hospice enrollees elect to be full code and if this is associated with higher hospice live discharge rates.

OBJECTIVE

To measure the rates of hospice enrollees electing full code, the characteristics predicting full code status, and the association of full code status with various hospice live discharge patterns.

DESIGN

Retrospective cohort study of electronic medical record data.

SETTING/SUBJECTS

A total of 25,636 decedents enrolled in two Michigan hospices between 2009 and 2014.

MEASUREMENTS

Code status was defined as full code versus do-not-resuscitate (DNR) orders. Covariates include demographics, location (home, hospice facility, nursing home, and hospital), primary diagnosis, and length of stay. Hospice live discharge was defined as short (0-14 days), medium (15-179 days), and long (>179 days).

RESULTS

A total of 12.9% of hospice enrollees elected full code status. This was significantly (p < 0.05) predicted by male sex, younger age, nonwhite race, home setting of care, and cancer diagnosis. Those with full code status had 1.76 times the adjusted odds of hospice live discharge compared with those with DNR orders (95% confidence interval [CI] 1.44-2.16) and 2.47 times the odds of short live discharge (95% CI 1.69-3.62) with no significant difference in long live discharge. The association of full code orders with hospice live discharge was stronger for nonwhite enrollees, with a live discharge rate of 23.8% versus 11.6% for African Americans with full code versus DNR orders.

CONCLUSIONS

Those electing full code status on admission to hospice are at high risk of live hospice discharge after short enrollments, particularly nonwhite enrollees.

End of life care for glioblastoma patients at a large academic cancer center

Glioblastoma (GBM) is a universally fatal disease, complicated by significant cognitive and physical disabilities, inherent to the disease course. The purpose of this study was to retrospectively analyze end-of-life care for GBM patients at an academic center and compare utilization of these services to national quality of care guidelines, with the goal of identifying opportunities to improve end-of-life care. Single center retrospective cohort study of GBM patients at Johns Hopkins Hospital (JHH) between 2009 and 2014, using electronic medical records and hospice records. Comprehensive medical record review of 100 randomly selected patients with GBM, who were actively treated at JHH. Secondary analysis of all JHH GBM patients (n = 45) who received hospice care at Gilchrist Services, our largest provider, during this time period. Of 100 patients, 76 were referred to hospice. Despite the poor survival and changes in mental capacity associated with this disease, only 40% of individuals had documentation of code status and only 17% had any documentation of advance directives (ADs). None had documentation by a health care provider of a formal symptom, psychosocial, or spiritual assessment at greater than 50% of clinic visits. Only 17% used chemotherapy in their last month of life. 37% were hospitalized in the last month of life for an average of 9 days. Of the Gilchrist Services patients, the median length of stay in hospice was 21 days and 64% of these patients died in their residence with hospice services. Documentation of palliative care and end-of-life measures could improve quality of care for GBM patients, especially in the use of ADs, symptom, spiritual, and psychosocial assessments, with earlier use of hospice to prevent end-of-life hospitalizations.

Pediatric Primary Care Involvement in End-of-Life Care for Children

OBJECTIVES

To examine the relationship between pediatric primary care involvement and hospice and home health care use at end of life.

METHODS

California Medicaid data were used to estimate the relationship between pediatric primary care involvement and use of hospice and home health care using generalized estimating equations.

RESULTS

Of the 2037 children who died between 2007 and 2010, 11% used hospice and 23% used home health. Among all children, primary care was not related to hospice use and was associated with home health use, usual source of care (OR = 1.83, P < .05), comprehensive care (OR = 1.60, P < .05), and continuous care (low: OR = 1.49, P < .05; moderate: OR = 2.57, P < .05; high: OR = 2.12, P < .05). Primary care for children aged 15 to 20 years was related to hospice use, usual source of care (OR = 4.06, P < .05) and continuous care (low: OR = 4.92, P < .05; moderate OR = 4.09, P < .05; high OR = 3.92, P < .05). Primary care for children under 5 years was associated with home health use, usual source of care (OR = 2.59, P < .05), comprehensive care (OR = 2.49, P < .05), and continuous care (low: OR = 2.22, P < .05; moderate: OR = 3.64, P < .05; high: OR = 3.62, P < .05). For children aged 6 to 14 years, this association was seen with continuous care (moderate: OR = 2.38, P < .05; high: OR = 2.13, P < .05). Home health for children aged 15 to 20 years was related to continuous care (moderate: OR = 2.32, P < .05).

CONCLUSION

Primary care involvement affected hospice use among older age-groups and home health use among younger age-groups. These findings underscore the need for clinical knowledge about end-of-life care for children of all ages among primary care providers.

Analysis of Advance Directive Documentation to Support Palliative Care Activities in Nursing Homes

As part of an intervention to improve health care in nursing homes with the goal of reducing potentially avoidable hospital admissions, 1,877 resident records were reviewed for advance directive (AD) documentation. At the initial phases of the intervention, 50 percent of the records contained an AD. Of the ADs in the resident records, 55 percent designated a durable power of attorney for health care, most often a child (62 percent), other relative (14 percent), or spouse (13 percent). Financial power of attorney documents were sometimes found within the AD, even though these documents focused on financial decision making rather than health care decision making. Code status was the most prevalent health preference documented in the record at 97 percent of the records reviewed. The intervention used these initial findings and the philosophical framework of respect for autonomy to develop education programs and services on advance care planning. The role of the social worker within an interdisciplinary team is discussed.

Older Adults of Underrepresented Populations and Their End-of-Life Preferences: An Integrative Review

This integrative review was conducted to examine the evidence for understanding diversity in end-of-life preferences among older adults of underrepresented groups. Findings from 21 studies were critically examined, grouped, and compared across studies, populations, and settings. Five major themes emerged: advance directives, hospice and palliative care, communication, knowledge and information, and home and family. Despite multidisciplinary attention, content and methodological limitations narrowed understanding of what matters most to these groups when making decisions at end of life. Rigorous longitudinal studies with more ethnically diverse samples are needed to detect modifiable factors related to disparities at the end of life.

Home care for children with multiple complex chronic conditions at the end of life: The choice of hospice versus home health

Families desire to bring their children home at end of life, and this creates a variety of unique care needs at home. This study analyzed the child and family factors associated with hospice versus home health care use in the last year of life among children with multiple complex chronic conditions. Using the Andersen Behavioral Healthcare Utilization Model, the predisposing, enabling, and need factors of the child and family were shown to be significant predictors of hospice and home health care use. Hospice and home health care have advantages, and families may wish to use the service that best fits their needs.

Use of a Computer Program for Advance Care Planning with African American Participants

ACKNOWLEDGMENTS

The authors wish to acknowledge the support and assistance of Dr. William Lawrence for his contribution to the M.A.UT model used in the decision aid, Making Your Wishes Known: Planning Your Medical Future (MYWK), Dr. Cheryl Dellasega for her leadership in focus group activities, Charles Sabatino for his review of legal aspects of MYWK, Dr. Robert Pearlman and his collaborative team for use of the advance care planning booklet "Your Life, Your Choices," Megan Whitehead for assistance in grant preparation and project organization, and the Instructional Media Development Center at the University of Wisconsin as well as JPL Integrated Communications for production and programming of MYWK.

BACKGROUND

For various cultural and historical reasons, African Americans are less likely than Caucasians to engage in advance care planning (ACP) for healthcare decisions.

PURPOSE

This pilot study tested whether an interactive computer program could help overcome barriers to effective ACP among African Americans.

METHODS

African American adults were recruited from traditionally Black churches to complete an interactive computer program on ACP, pre-/post-questionnaires, and a follow-up phone interview.

RESULTS

Eighteen adults (mean age =53.2 years, 83% female) completed the program without any problems. Knowledge about ACP significantly increased following the computer intervention (44.9% → 61.3%, p=0.0004), as did individuals' sense of self-determination. Participants were highly satisfied with the ACP process (9.4; 1 = not at all satisfied, 10 = extremely satisfied), and reported that the computer-generated advance directive accurately reflected their wishes (6.4; 1 = not at all accurate, 7 = extremely accurate). Follow-up phone interviews found that >80% of participants reported having shared their advance directives with family members and spokespeople.

CONCLUSION

Preliminary evidence suggests that an interactive computer program can help African Americans engage in effective advance care planning, including creating an accurate advance directive document that will be shared with loved ones.

Advance directives among hospitalized patients with heart failure

OBJECTIVES

The purpose of this study was to assess the frequency and correlates of documented advance directives (ADs) among patients hospitalized for heart failure (HF).

BACKGROUND

Discussing ADs with patients with HF is critical for identifying treatment goals consistent with patients' values and preferences and for facilitating health care team communication.

METHODS

We retrospectively identified electronic medical records of adult patients admitted to 2 large tertiary care hospitals with either the primary or secondary discharge diagnosis of HF from September 2008 to August 2013 to assess the presence of ADs in electronic medical records. We performed analyses including HF as either the primary or secondary admission diagnosis and HF as the primary admission diagnosis only. Multivariable models were constructed to investigate independent predictors of documented ADs.

RESULTS

Data included 44,768 admissions from 24,291 individual patients over 5 years. Mean age of patients at admission was 64.8 ± 15.9 years; 47.9% of these patients were female, 51.8% were black. The median length of stay for all admissions was 5 (3 to 10) days; 12.7% of patients had documented ADs. Older age, female sex, white race, higher socioeconomic status, higher risk for adverse in-hospital outcomes, length of stay ≥5 days, hospice discharge, palliative care consultation, and a do-not-resuscitate order were all associated with a significantly higher chance of having documented ADs. A significant increase in ADs over time was noted, but more than 80% of patients did not have ADs in medical records at the end of the study period.

CONCLUSIONS

In a diverse population of hospitalized patients with HF, most did not have a documented AD in the medical records. Although several factors were associated with a higher probability, major opportunities exist for all subgroups of patients with HF to improve documentation of ADs.

The Relationship Between Organizational Characteristics and Advance Care Planning Practices

Organizational characteristics may impede the uniform adoption of advance care planning (ACP) best practices. We conducted telephone interviews with site directors of a Midwestern state’s Medicaid waiver program administered by the Area Agencies on Aging and surveyed the 433 care managers (registered nurses and social workers) employed within these 9 agencies. Care managers at 2 agencies reported more frequent ACP discussions and higher levels of confidence. Both sites had ACP training programs, follow-up protocols, and informational packets available for consumers that were not consistently available at the other agencies. The findings point to the need for consistent educational programs and policies on ACP and more in depth examination of the values, beliefs, and resources that account for organizational differences in ACP.

Completion of Advance Directives Among Low-Income Older Adults

This study examined the prevalence of completion of advance directives (ADs) and the effects of race/ethnicity on AD completion using a cross-sectional design. Low-income older adults (n = 256) who were residents of supportive housing facilities or members of a senior center were interviewed in person. About 20% of the participants had completed ADs. Knowledge and attitudes toward ADs, income, and previous experience in an intensive care unit significantly predicted the completion of ADs after controlling for other factors. Those with higher levels of knowledge, positive attitudes, or those with higher incomes were more likely to complete ADs than their counterparts. Findings suggest that as a proxy for multiple socioeconomic, social and cultural factors, race, and ethnicity need to be considered in planning end-of-life care.