Abstract
Antiretroviral therapy has dramatically increased the lifespan of people living with HIV (PLWH), but advance care planning (ACP) and hospice services are underutilized in this population. The purpose of this study was to understand barriers and facilitators to ACP among this group. PLWH (n = 25) were recruited from an HIV Clinic at a Veterans Affairs (VA) Medical Center in Atlanta, GA to represent a range of sociodemographic characteristics and experiences. Semi-structured interviews were analyzed using thematic analysis. More than half of participants (64%) indicated not engaging in ACP. We identified four key barriers to ACP: (1) a self-image among PLWH as “survivors” (and a reluctance to think about ACP); (2) a history of mistrust and mistreatment; (3) weak social ties and a desire to avoid disclosure of HIV status; and (4) a value for self-reliance. Findings have important implications for interventions to overcome these barriers.
• This paper identifies lack of advance care planning among older persons living with HIV (PLWH) as a substantial challenge; more than half (64%) of the current sample did not engage in ACP. • It provides insights for providers and other stakeholders (e.g., involved family and friends) regarding barriers to advance care planning (ACP) among PLWH. • Findings have important implications for interventions to address barriers to ACP among PLWH, including education for clinicians as well as PLWH. • Potential interventions to address barriers we identified are discussed.
• Anti-bias interventions targeted at improving clinician understanding of the experiences of PLWH may improve relationship rapport and openness to ACP conversations. • Given weak social ties and strained family relations, clinicians should prioritize finding if there is a preferred surrogate decision maker over completing full advance directives for PLWH. • Interventions to increase ACP completion could emphasize self-reliance and independence by prioritizing one’s personal preferences over potential desires of other legal surrogates.What this paper adds
Applications of study findings
Introduction
Owing to the development of effective antiretroviral therapy (ART), older adults aged 50 and over now account for >40% of people living with HIV (PLWH) who live in the United States (Sullivan et al., 2020). As life-expectancy increases, causes of death have shifted from HIV-related opportunistic infections to medical conditions associated with aging, such as cardiovascular disease, cancer, and neurodegenerative disorders (Heron, 2019).
Advance care planning (ACP) is important for assuring older adults’ end-of-life preferences are integrated into their care. The American Geriatrics Society issued a statement supporting the inclusion of ACP conversations into clinical practice with older adults, calling it a “critical tool” and noting that ACP discussions have been shown to result in care that reflects personal preferences and improves quality of life, also known as “goal-concordant care” or “patient-centered care” (AGS, 2017; Silveira et al., 2010). In addition, effective ACP increases satisfaction with the healthcare system, reduces stress, anxiety, and depression for older adults, family caregivers, and relatives (Detering et al., 2010). Interventions such as the Serious Illness Conversation Guide have provided detailed instructions for outpatient clinicians on how to begin these conversations in the clinic and have been shown effective in multiple studies (Bernacki et al., 2019). Unfortunately, none of these interventions have been validated for discussions with PLWH.
Over the past three decades, end-of-life care and planning has shifting dramatically for PLWH. In the early days of HIV, an HIV diagnosis was rarely made without an AIDS presentation and ACP was mainly institution of hospice services, which were often in short supply. Now, as PLWH are aging with (and eventually dying from) the same age-related medical conditions that affect the general population, it has become critical that discussions about ACP take place in the clinic as a part of routine health maintenance. Unfortunately, research has shown that aging PLWH are less likely to engage in important discussions about ACP than the general population (de Caprariis et al., 2013; Sangarlangkarn et al., 2016).
One review reported that the majority of PLWH had variable exposure to ACP—less than half had end-of-life communication and even fewer were advised about Advance Directives (ADs) (Sangarlangkarn et al., 2016). Physicians felt that they had limited time and inadequate preparation to have discussions with PLWH surrounding ACP (Sangarlangkarn et al., 2016). Additionally, the number of completed ADs does not increase with increasing age or number of years since HIV diagnosis (de Caprariis et al., 2013).
Others evaluated the experiences of PLWH (prior to the universal utilization of ART) as they were asked to complete ADs and examined what they felt to be the purpose of ACP completion—noting themes of autonomy, facing death, and strengthening relationships (Martin et al., 1999). Another recent study evaluated the impact of ACP on family/surrogate understanding of end-of-life wishes for PLWH, however, no studies have analyzed a broad sampling of aging PLWH themselves (Lyon et al., 2020). The lack of information about the unique experiences of aging PLWH and their ability to participate in ACP conversations may help to explain the low prevalence of completed ADs in this population. In the current study, we sought to explore what factors must be navigated by the older patient in order to effectively participate in conversations surrounding ACPs, advocate for themselves, and receive care that reflects their personal preferences and improves their quality of life.
Method
Study Design and Population
Data for the current study are from a larger qualitative study designed to investigate social determinants of health (SDH) that shape Veterans’ experiences aging with HIV, including their health-care decision-making, and is informed by the influential Dahlgren and Whitehead (1991) social ecological model. SDH are non-medical social, economic, and environmental factors that influence the health and well-being of people and communities and are major drivers of health disparities and inequities over the life course (Marmot & Wilkerson, 2006). Whereas the larger study focuses on social and structural factors that shape health and health outcomes among PLWH more broadly, the current study focuses in depth on barriers and facilitators to ACP which has important implications for these patients’ quality of life at end of life. Study participants include 25 US Military Veterans participating in the Veterans Aging Cohort Study (VACS) who were recruited from an HIV clinic at the Atlanta Veterans Affairs (VA) Medical Center using maximum variation sampling VACS is an observational, prospective longitudinal study of PLWH and age, race/ethnicity, sex, and site-matched patients without HIV receiving care at VA medical centers across the United States (Justice et al., 2006). The study was funded by the Emory Center for AIDS Research (CFAR) and jointly approved by Emory University’s Institutional Review Board and the Atlanta VA Institutional Review Board (#68789). All participants completed informed consent before participating in the interviews.
Data Collection
Primary data collection involved the completion of 25 semi-structured in-depth interviews and a social network map administered by two of the co-authors (S.N.H. and M.M.P.) (Antonucci, 1986). Interviews were conducted in empty adjacent clinic space. The qualitative questions addressed a range of issues relevant to the research aims of the larger study. In addition to health care decision-making, topics included life history and significant life events; co-habitation status; living arrangements, social and community support; daily routines; health care; use of long-term care; health perceptions and behaviors: and expectations regarding the future. In-depth interviews ranged in length from 47 to 198 minutes (mean = 98 minutes) and were digitally recorded and professionally transcribed verbatim, and participants were numbered to maintain anonymity. We used VACS survey and clinical data, including scores on the VACS 2.0 Index, to describe the sample and provide context for the qualitative data. The VACS 2.0 Index is an update to the VACS 1.0 Index and predicts all-cause as well as cause-specific mortality and other health outcomes in PLWH (McGinnis et al., 2022). It creates a score by summing age, body mass index, HIV-related indictors (CD4 count and HIV-1 RNA), and general markers of organ system injury, including hepatitis C infection. McGinnis and her colleagues note that the index can help identify PLWH who can engage in fewer clinical encounters and potentially transition from HIV care to primary care as well as identify those at higher mortality risk (e.g., those with whom ACP discussions may need to be initiated). Their data show that each 5-point increase on the VACS 2.0 Index is associated with an approximate 6.5% increase in predicted 5-year mortality. For a score of 50, predicted mortality at 1 year is approximately 3%, 8% at 3 years, and 12% at 5 years.
Data Analysis
We used a thematic analysis approach to analyze these data (Braun & Clarke, 2022). The first author led the analysis which involved a step-by-step process of reading interview text line-by-line and applying initial codes to phrases, words, and patterns (i.e., themes) salient to the research aims. These initial codes were then refined, sorted, and synthesized in an iterative manner to establish and finalize major overarching themes through multiple readings of the text. The analysis was conducted in ongoing consultation with the last author, a senior qualitative researcher, and final themes were confirmed based on input from the full author team. We used a qualitative software program, NVIVO 12, to manage and facilitate the analysis.
Results
Characteristics of Participants According to Their ACP Engagement
Characteristics of Participants According to ACP Engagement (N = 25).
Note. Percentages may not sum to 100 due to rounding.
aPhysical function was measured using the VACS Physical Function Scale (Oursler, et al., 2011).
bDepression was measured using the PHQ-9 index (Kroenke & Spitzer, 2002).
cTypical scores range from 0–100; higher scores indicate greater disease burden.
dData is missing on one participant in the No ACP category.
When specifically asked during our interviews whether they had completed an AD, 16 of our 25 participants (64%) had not completed one. Of the nine who had completed some type of AD, two reported having arranged their finances for their future funeral, and two stated that they would use 911 emergency services as their end-of-life plan. Seven participants completed advance directive paperwork and documented a health care surrogate. As indicated in Table 1, the No ACP group included a higher percentage of African Americans and more lived alone and had histories of homelessness and incarceration compared with the ACP group. No other notable differences between the two groups were identified (including no difference in VACS Index scores).
Qualitative Findings
Participants often commented similarly, “that’s a long way away” in reference to their thoughts about advance care planning for the end-of-life—despite, in many cases, having relatively high VACS Index scores (see Table 1) suggesting that such planning might be pertinent (McGinnis et al., 2022). We identified four overlapping themes that represented barriers to ACP: (1) Self-image as “survivors” (and a reluctance to think about ACP); (2) A history of mistrust and mistreatment; (3) Weak social ties and a desire to avoid disclosure of HIV status; and (4) A value for self-reliance (i.e., not needing or wanting support from other people). These intersecting relationships and how they shape ACP completion are illustrated in the diagram in Figure 1. Diagram of relationships between ACP completion and emergent participant themes.
Self-Image as “Survivors”
Participants embodied a self-image as “survivors,” in which some participants related their longevity, compared with peers with HIV who developed AIDS and died many years before, with a sense of “being healthy.” Some participants felt so secure in this identity as “survivors” that they were unwilling to engage in ACP conversations about the end of their lives, feeling that death was far away—a perception we conceptualize as a form of negative resilience where participants turn a blind eye to the future, a strategy that can have negative consequences.
Rather than responding directly to why he had not engaged in ACP with his medical team, one participant responded with his advice to other PLWH that he counsels, “[I] tell them to just be positive and strong and it’s not as bad as it looks and sounds. I mean it’s a lifetime illness, but people have—dialysis—it could be like that” (PT13). This participant’s opinion was shared by others who felt that they were responsible for their survivorship, related to good exercise and eating habits, as well as taking their medications, which they were committed to continue to do. One described a detailed exercise routine that was keeping him healthy; another noted that while he looks healthy now, “looks can be deceptive. I have cirrhosis of the liver and hepatitis C and I have a lot of things that I deal with that are chronic and you know that are semi-irreversible so—but thank goodness I have worked very hard to not die” (PT9). Many participants shared the belief that they would live many more years—as one participant stated, “I keep telling everybody I am going to outlive my aunt and she lived to be 104” (PT20). Overall, we found that participants who had this sense of being invincible were less inclined to engage in concrete discussions about the inevitable end of their lives.
A History of Mistrust and Mistreatment
Layered with a background of survivorship for participants was an additional theme of mistrust and mistreatment, which colored participants’ experiences in the health care system and other aspects of their lives. Participants discussed barriers they faced in connecting with family, friends, coworkers, and most of all, health care providers—making it difficult to form the trusting relationships necessary to engage in sensitive discussions about their end-of-life desires.
A key barrier to connection related to feeling judged by others, both in the military and in civilian life, and participants frequently described experiences which occurred years previously, in the pre-ART era. Prejudice experienced in the military was often particularly damaging and long-lasting, involving inappropriate disclosure of HIV status or forced “outing” by colleagues at work with damaging results, including harassment such as covering a personal locker in condoms, as participant 18 shared. Another participant described, “They kicked me out [of the military] for being gay and said you will never be eligible for benefits—get out of here fag” (PT24). This judgment extended outside the military—to medical providers, community members, or even their families. “People stay away from me’, cause they think they [might get] sick” (PT 15). Multiple participants referred to physicians and dentists who avoided touching them or canceled their appointments for fear that they would contract HIV. One participant shared an example of gross mistreatment he experiences first-hand at the end of a friend’s life: I have watched so many of them die. [My friend] at the Baptist hospital—my buddy called me up and he says can I come in—his partner was there, and I say, “Yeah I can come.” The nurses wouldn’t clean him at that time, so we bathed him and put clean clothes on him. Three hours later he passed away and I told [my buddy] I can’t handle this anymore; I reached my breaking point. (PT 12)
It is important to note, however, that there were some reports of trusting relationships, particularly in relation to VA HIV care providers. Although many participants described mistreatment they experienced because of their HIV status during their time serving in the military, many others also expressed thankfulness for the excellent care they reportedly received at the Atlanta VA HIV Clinic. In commenting about this care, one participant stated, “I have nothin’ but praises for the place here. This [clinic] has been a godsend” (PT7). Many valued the clinic and its providers for the continuity of care they received (i.e., several participants reported seeing the same HIV providers over several years) as well as the feeling of being supported and cared for personally—that their calls were returned, their visits were prompt and professional, and that their care was handled with respect and without stigma. To highlight this relationship, two participants included their VA HIV care providers on social network maps as an example of a strong, trusting relationship.
Weak Social Ties and a Desire to Avoid Disclosure of HIV Status
While some participants indicated they had support from friends and family, the majority struggled with limited social connectedness or a fear of disclosure of their HIV status limiting the depth of their relationships. These participants had difficulty identifying people with whom they could depend upon for social support or instrumental support and were unable to respond to how they would be supported by their community in the event of an injury or a decline in functional status. One participant stated, “I don’t have anybody that could help me” (PT17). Another said, “I just have to wait for the flat line. Can’t go expect[ing] somebody to help you. Somebody is not going to help you.” (PT23). Similar to those who identified as “survivors,” we found that participants who lacked social support were also less likely to have engaged in ACP conversations.
Even those with some family involvement did not universally feel these family members would provide care at the end of life, which led to one participant being wistful about his lack of concrete plans, “call 911 – that’s my plan, call 911. There’s nothing I can get ready, just that one emergency number I can call” (PT9). Another participant was dejected, stating “it would be nice to travel til’ you drop, but that’s not feasible, so I know I’m just going to end up in some bed in some nursing home” (PT24). One participant described a particularly dismal living situation with his sister, who he did not see as a potential resource at the end of his life. I am looking into going into a nursing home this month because I am tired of living in my sister’s house and paying her a fortune and not being able to get anything else from her except living in her house. If she saw me laying in the floor or stumbling by my bed and falling on the floor, she would not help me up. (PT5).
Related to participants’ perceptions regarding their inability to rely on close relationships at the end of life was the fear of HIV status disclosure, perhaps due to the stigma some participants reported experiencing because of their illness. One participant described that although he had appointed a health care power of attorney, he had not disclosed his HIV status to this person—when asked why, he stated, “I just I guess because I have been in denial all these years. You know—it is not a thing that you discuss with people.” (PT 15) Multiple other participants indicated that many members of their limited social support networks (including friends and family) were not aware of their HIV status—complicating the ability to have effective ACP conversations about their end-of-life wishes, or potentially, to receive caregiving support for a potentially HIV-related diagnosis.
A Value for Self-Reliance
Often stemming from experiencing stigma related to their HIV status, many participants described the importance of becoming self-reliant, which they related to a general desire for privacy and independence (“minding my own business”) and made it difficult to engage in discussions of the future physical decline and dependence that accompany end-of-life care. Some participants exhibited this value for self-reliance when discussing the importance of guiding their own HIV care and avoiding relying on help, one participant expressed the view of several. He stated, “I am very independent, you know that even if I need help I am not going to ask for it” (PT1). Another participant referred to the process of becoming educated about his disease as a form of self-empowerment, “[After] I got HIV, I went to school and learned about it; decided what medicine [and] should I take medicine or not take medicine; what would be the better risk for me?” (PT3). A third participant described how he took control of his medical care by directing his medical team, “I tell them things, I talk, and they listen and answer, give me feedback, and they’ll put the requests in.” (PT22) While some of these examples of taking control of one’s health and healthcare reflect positively, we found that this value for self-reliance was more often a barrier to ACP than a facilitator. Below, we discuss implications of these findings, including suggestions for ways that some barriers to ACP we identified, such as the value for self-reliance, could be channeled to promote such planning.
Discussion
The aim of this study was to investigate what factors must be navigated by older PLWH and their providers and other stakeholders (e.g., involved family and friends) to effectively participate in conversations surrounding ACPs, advocate for themselves, and receive care that reflects their personal preferences and improves their quality of life. Barriers we identified included a self-image as survivors (which we describe as a marker of “negative resiliency”), history of mistrust and mistreatment, weak social ties, and a desire to avoid disclosure of HIV status, and a value for self-reliance.
It has previously been shown that older adults are unlikely to initiate ACP conversations, though they identify that these conversations are important (Hopkins et al., 2020; Detering et al., 2010). Our results suggest that while some of our participants identify as survivors, which may have overall health benefits attributed to high degrees of hopefulness (as has been seen in oncology populations), they continue to have low likelihood of ACP completion, perhaps related to not finding discussions to yet be clinically relevant personally (Corn et al., 2020; Hall et al., 2019). If this barrier can be overcome, ACP completion has also been shown to bring benefits including patient-described increase in sense of peace and decrease in worry, anxiety, and depression (Detering et al., 2010; Hall et al., 2019).
Prior studies have shown the impact of AIDS-related grief and coping with loss among PLWH to increase psychological distress, substance use disorder, and depressive symptoms, but none have linked this directly to engagement in end-of-life care and ACP, as we propose in this study (Sikkema et al., 2003). As shown in Figure 1, those participants who did not engage in ACP were more likely to describe experiences of grief and loss, which they identified as having overcome (as “survivors”) or struggled with through lack of social support and trust in their medical team, both of which served as potential barriers to further ACP participation and AD completion. In addition, this is among the first studies to link stigma of HIV diagnosis with barriers to ACP conversations.
If our goal is to overcome barriers that have been identified and provide effective ACP counseling in the outpatient setting, providers must be willing to self-educate, about the general traumas faced by patients with HIV (both historically and presently), the experiences of the patient in front of them, and about how to have these conversations—lack of education has been identified as a provider-level barrier to increasing and about how to have these conversations—lack of education has been identified as a provider-level barrier to increasing ACP discussions in the clinic setting (Fulmer, et al., 2018). Providers must learn which patients to target these conversations toward, and the VACS Index, which is a reliable predictor of all-cause as well as cause-specific mortality and other outcomes in PLWH, could be used as a trigger to prompt conversations about the end-of-life. Prior interviews with older adults have noted that patients are more open to participating in ACP discussions when they find them to be clinically relevant, which emphasizes the importance of sharing high mortality risks from the VACS Index or other tools with the patient directly, though this may be challenging (Hall, et al., 2019). Anti-bias interventions can help providers increase introspection, sensitivity, and cultural awareness to support improved care outcomes for PLWH and to addressing patient discomfort and stigma related to mistrust of the medical system (Ford, 2017). In addition, deeper awareness of the challenges that many PLWH experience emphasizes the importance of multi-modal care delivery, such as in patient centered medical homes or VA “Centers of Excellence” clinics, where PLWH may schedule appointments with physicians and mental health or social work staff within the same office, who share similar expertise caring for this population. Care models like this also have clinical benefits—previous work has shown that patients who receive HIV care at the VA have increased virologic suppression and retention in care (Mangal et al., 2014).
Our findings regarding isolation and limited social ties among some PLWH highlights the importance of screening for and addressing unmet social needs and points to the necessity of incorporating social programming, such as therapy groups, exercise/weight loss programs, and support groups into patients’ care plans. One such group, the Silver Linings Project, was started in Atlanta for Black, gay men living with HIV over the age of 50—specifically to address the limited social ties that many patients face (THRIVESS, 2023). Through an educational curriculum provided by the Silver Linings Project, PLWH learn about the physical impact of having HIV infection on aging and discuss traumas, loss/grief, and the experience of stigma in the world today—many of the themes which were raised by participants in this study and would argue for broader distribution of their work in an office-based setting. Groups such as this, along with community health workers and peer navigators, may be beneficial for increasing ACP participation, both through dissemination of information from a trusted source, as well as improving the social ties which limited our participants’ ability to envision support at the end of life. These relationships may also be fruitful places to highlight the potential benefits of engaging in ACP discussions, as noted above. Identifying other social risk factors associated with higher VACS Index scores (i.e., indicating an increased risk for morbidity and mortality), such as housing instability, also could help trigger these discussions (Reddon et al., 2022).
A psychological intervention to increase medication adherence among PLWH focused on acknowledging unresolved grief in order to transcend loss and move toward adjustment, social reintegration, and treatment adherence—a process which may serve to be helpful within a clinic setting to increase involvement in ACP discussions (Brion, et al., 2013). For those participants who emphasized the importance of self-reliance and being independent, future interventions could target the importance of making one’s own end-of-life plans through effective ACP as a method of prioritizing one’s personal preferences over potential desires of other legal surrogates.
Although the metric of a successful ACP discussion has historically been a completed state-specific advance directive, in this population with an increased likelihood of fragmented or weak social ties, we suggest that emphasis shift toward designation of a surrogate decision-maker informed of the patient’s HIV status, as this may allow for more goal-concordant care with the individual patient, and could avoid some racial, cultural, and sociodemographic barriers to full AD completion (Fagerlin et al., 2001). Recent palliative care commentary has emphasized goal-concordant care as being the ultimate objective of effective ACP, which we suggest, in our unique population, may hinge on the appointment of a desired surrogate (Morrison, et al., 2021).
While we identified many barriers to ACP, several participants also referred to positive aspects of their healthcare experience that might be utilized to promote ACP including long term relationships with HIV care providers that were based on respect and compassion. We propose that prioritization of these types of relationships as a location for ACP discussions may result in more willing engagement by PLWH and AD completion for goal-concordant care. In doing so, it will be important to consider the need for more HIV providers to be trained in palliative care skills such as ACP and AD completion, as this may not be taught in infectious disease fellowship, where the majority of these providers receive their training. Current IDSA guidelines for HIV primary care and ACGME core competencies for infectious disease fellowship do not include specific mention of ACP but should be amended in future versions (Aberg et al., 2014).
Our study has a few clear limitations—given that our study was set in a VA Medical Center in Atlanta, GA, our participant population represents exclusively veterans from one state in U.S., who aren’t necessarily reflective of the general US population, though their interviews often highlighted experiences in civilian life. We intentionally recruited participants over the age of 50 and performed our qualitative analysis with the hope of their having considered end-of-life care, however, we acknowledge that their unique experiences living through the AIDS epidemic of the 1980s–1990s may impact the transferability of findings to younger PLWH of their beliefs and barriers to ACP. While our study focuses on patient-level factors that influence ACP completion, future directions for study could include a direct look at provider-level factors through interviews with primary care, geriatrics, and HIV care providers (Fulmer et al., 2018).
Conclusion
This study offers a unique contribution to the literature through direct insight into the beliefs and experiences of PLWH themselves. As a result, we have been able to identify key themes that highlight the barriers and facilitators to end-of-life care discussions and suggest clinic-based interventions which could address these challenges. Further exploration into each patient’s experience will improve our ability to execute their wishes at the end-of-life.
Footnotes
Acknowledgments
We would like to acknowledge the following individuals for their invaluable input and support of this project: Edwin Clayton Carruth, Hannah Cooper, Kathleen McGinnis, Ramona Rai, Christopher Rentsch, and Janet Tate. Thank you also to Patrick Brown for his assistance with recruitment. We also are indebted to the Veterans who gave so generously of their time and participated in the interviews.
Declaration of Conflicting Interests
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: V.C. Marconi has received investigator-initiated research grants (to Emory University) and consultation fees from Eli Lilly, Bayer, Gilead Sciences, Merck, and ViiV. The other authors declare no conflicts of interest.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Qualitative data reported here were collected as part of a larger project entitled, Social Determinants of Health in Veterans Living and Aging with HIV in Georgia (M.M. Perkins, PI) funded by the Emory National Institutes of Health Center for AIDS Research (P30AI050409). V.C. Marconi also has received funding support from the Emory Center for AIDS Research (P30AI050409) for work related to this manuscript. T.M. Johnson II receives support related to ACP work from the Health Resources and Services Administration (HRSA) of the U.S. Department of Health and Human Services (HHS) (U1QHP33070). Additional support for this research was provided by grants awarded to A. C. Justice from the U.S. Veterans Health Administration and the National Institute on Alcohol Abuse and Alcoholism at the National Institutes of Health (U10-AA013566; U24-AA020794; U01-AA020790; U24-AA022001). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health, HRSA, or the U.S. Veterans Health Administration.
