Structural Inequities, Syndemics, and Resilience: The Critical Role of Social Support in Overcoming Barriers and Empowering Engagement in HIV Care for Young Black Sexual-Minority Men in the US South

Abstract

Young Black sexual minority men (YBSMM) living in the US South are among those most disproportionately impacted by HIV in the USA. This health inequity is, in part, due to lower rates of sustained engagement in the HIV care continuum, resulting in a lower prevalence of viral suppression and higher overall community-level viral load. Social, structural, and economic inequities have previously been linked with poorer HIV care engagement among YBSMM. HIV-related social support, individual-level resilience, and healthcare empowerment have been shown to be independently associated with improved HIV care engagement. The current study sought to assess the relative contribution of individual, structural, and economic factors on engagement in HIV care and to elucidate the potentially mediating role of healthcare empowerment. Data from 224 YBSMM with HIV in the US South indicated that greater levels of socioeconomic distress, intimate partner violence, and depressive symptoms were associated with lower levels of engagement in HIV care, while greater levels of individual-level resilience and healthcare empowerment were associated with higher levels of HIV care engagement. Importantly, healthcare empowerment mediated the association between resilience and engagement in HIV care and the association between social support and engagement in HIV care. Findings emphasize the critical role that HIV-related social support plays in fostering resilience and overcoming syndemic factors to promote empowerment and engagement in HIV care for YBSMM in the USA.

Introduction

Compared to other persons affected by HIV in the USA, young (ages 18–29) Black sexual minority men (YBSMM) are currently the most disproportionately impacted by HIV, particularly in the US South [1]. YBSMM have a 40 to 60% lower likelihood of engagement at each step of the HIV care continuum than other sexual minority men, resulting in markedly less viral suppression [2]. YBSMM are more likely to fall out of HIV care than older Black sexual minority men [3, 4]. An estimated 41% of YBSMM living with HIV fall out of HIV care after they have initially been linked to care [2, 5, 6]. Moreover, HIV transmissibility is heightened by less engagement in the HIV care continuum and lower viral suppression contributing to higher HIV incidence among YBSMM [7–9].

Social, structural, and economic inequities have previously been linked with poorer HIV care engagement among YBSMM [10–13]. Specifically, socioeconomic distress has been shown to detrimentally impact HIV care continuum engagement among YBSMM, particularly among YBSMM in the US South [11–14]. Social and structural stressors encountered by YBSMM with HIV are numerous and well-documented and include systemic barriers that can prevent them from receiving the care and support they need. In particular, navigating healthcare systems that may be dismissive or negligent can be a major challenge for this population [15, 16].

Research has also suggested that intimate partner violence, a correlate of socioeconomic distress and other structural inequities, is associated with a number of poorer HIV care engagement indicators [17]. These include missed HIV care visits [17], lapses or breaks in care [17], reduced likelihood of ART use [18], reduced adherence to ART [18], and lower likelihood of viral suppression [17–19]. Moreover, due to discrimination and victimization and from the internalization of intersectional stigma [20–25], YBSMM may also experience higher levels of depression [25, 26] than other groups of young sexual minority men. Depression is a correlate of socioeconomic distress and intimate partner violence among sexual minority men [27–30], and systematic reviews and meta-analyses have found that persons with depression were also less likely to be retained in HIV care [31, 32].

Syndemic theory provides an empirically guided framework to depict the cumulative effects of structural, social, and economic stressors on YBSMM, including factors that increase the risk for intimate partner violence, depression, and poorer HIV care engagement [33–35]. Syndemic theory suggests that these co-occurring psychosocial health conditions such as substance use, mental health problems, and violence are mutually reinforcing among marginalized populations [33–35]. This theory provides a framework to assess the effects of socioeconomic distress, intimate partner violence, and depression on HIV care engagement among YBSMM.

In addition to the multiple risk factors that negatively impact HIV outcomes, there are also key protective factors that may result in improved engagement in the HIV care continuum. For example, many persons living with HIV also manifest specific resiliencies that are highly relevant to HIV care engagement (e.g., self-management) [36–42]. Receiving more social support has been shown to be associated with greater treatment adherence [43–45] and lower viral load and more viral suppression among persons living with HIV [46–48]. More HIV-related social support has been shown to be associated with improved HIV care engagement among YBSMM [49].

Resilience, the process of adapting and adjusting to overcome social and structural hardships, has been shown to be associated with HIV care engagement among YBSMM [50, 51]. There has been significant interest in developing interventions that help to foster or bolster resilience to help individuals facing significant social and structural barriers to engage in health-promoting behaviors [52]. A danger in focusing on fostering resilience is that it may inadvertently place the responsibility for “overcoming” hardships on the very individuals who are experiencing socially and structurally produced hardships. Increasing social support, specifically HIV-related social support, may be one promising way of increasing resilience among YBSMM [53, 54], sharing the need with others to support and foster resilience in the face of HIV.

Prior work has linked healthcare empowerment (i.e., the state of being engaged, informed, collaborative, and committed to one’s healthcare) [55] with greater engagement in the HIV care continuum [56]. Being empowered about one’s healthcare is affected by a confluence of individual, social, structural, and economic factors such as age, race, socioeconomic distress, depressive symptoms, prior experiences with providers, and prior experiences of stigma and discrimination in healthcare settings [55, 56]. However, the relative contribution of these factors, as well as forms of HIV-related social support and resilience on the formation of healthcare empowerment and the associated impact on HIV care continuum engagement among YBSMM in the US South, remains to be assessed. The current analysis sought to assess the relative contribution of these individual, structural, and economic factors on engagement in HIV care and to elucidate the potentially mediating role of healthcare empowerment. We hypothesized that each of these syndemic factors would be associated with engagement in HIV care and that healthcare empowerment would mediate the associations, suggesting they may each be key intervention targets for increasing HIV care engagement (see Fig. 1).

Fig. 1.

Model of healthcare empowerment as a mechanism for young Black sexual minority men (YBSMM) engaging in HIV care at Time 3 (N = 224). Note: In this model, individual and contextual factors and healthcare empowerment are conceptualized as directly affecting the extent to which YBSMM engage in HIV care. Additionally, the model conceptualized healthcare empowerment as a mechanism that helps to explain why individual and contextual factors affect healthcare outcomes. Although individual and contextual factors are still important, healthcare empowerment may be a key point of intervention

Method

Participants

The current cross-sectional analyses were conducted with data collected at the third timepoint in an ongoing longitudinal community-based cohort study, the Texas Young Black Men’s Health Survey, of YBSMM living with HIV who were recruited from the community via long-chain peer referral [57] from either the Dallas or Houston metropolitan areas of Texas. Time 3 data were used as this was the timepoint where the study measures of interest were collected. Potential participants who were interested in joining the study were directed to contact the study coordinator for verification, eligibility screening, verbal consent, and scheduling of a baseline assessment at the study site. Eligibility criteria at baseline included the following: (1) currently lived in the greater Dallas/Houston Texas area, (2) was 18–29 years old, (3) identified as African American or Black, (4) reported having sex with a man in the last year, and (5) confirmed HIV-positive status via an OraQuick Rapid HIV test for HIV. The study sites were either at the University of Texas Southwestern in Dallas or by a team from the University of Texas, Houston, at a local community-based organization in Houston.

The recruitment and study methods are reported in greater detail elsewhere [58]. Briefly, at baseline, of the 368 participants who met the above inclusion criteria, 331 (89.9%) consented to participate in the full longitudinal cohort. They then were enrolled and completed the baseline (Time 1) survey from September 2015 through November 2016. Of the 331 respondents enrolled at baseline, 224 (67.7%) completed the online survey at Time 3 where the current study measures were collected from October 2019 through March 2020.

The median Time 3 survey length was 48.9 min. Participants were compensated $75 for completing the Time 3 survey assessment. All study procedures were approved by the institutional review boards at the University of California at San Francisco and the institutions of the data collection subcontractors in each geographic area in Texas.

Measures

Participants reported the following demographic characteristics at baseline (Time 1): age in years, highest level of education completed, personal income in the last year, and employment status. They also completed measures of the following study variables at Time 3.

Socioeconomic Distress

Three survey items assess the frequency of the following events in the past 12 months: (1) running out of money for basic necessities, (2) borrowing money from a friend or a relative to get by financially, and (3) having sex with someone in exchange for money, drugs, or housing [59–62]. Response options ranged from 1 (“Many times”) to 4 (“Never”), and items were reversed coded such that higher values reflected greater socioeconomic distress. When the items were summed, the average score was 7.28 (standard deviation (SD) = 2.10) and ranged from 3 to 12. Cronbach’s alpha at Time 3 was 0.65.

Intimate Partner Violence

Participants’ experiences of intimate partner violence by a male partner in the past 12 months were assessed using six items adapted from the Intimate Partner Violence among Gay and Bisexual Men short-form [IPV-GBM; 63]. Based on the frequencies they reported, their responses were coded 0 (never happened or did not happen in the past 12 months) to 6 (happened more than 20 times in the past 12 months), with summed higher scores reflecting the greater frequency of experiencing intimate partner violence by a male partner. Cronbach’s alpha at Time 3 was 0.86.

Depressive Symptoms

We used the 20-item Center for Epidemiologic Studies Depression Scale (CES-D; Radloff, 1977). These items assessed how often respondents felt or behaved in specific ways indicative of depressive symptoms during the past week. Sample items included “I felt depressed” and “I felt lonely.” For the present study, we focused on the following subscales: Negative Affect Symptoms (seven items), Somatic Symptoms (four items), and Interpersonal Symptoms (two items). Response options ranged from 0 (“Rarely or none of the time [Less than 1 day]”) to 3 (“Most or all of the time [5–7 days]”). Items are summed, with higher scores indicating more severe depressive symptomology. Cronbach’s alpha values at Time 3 were 0.91 for the Negative Affect subscale, 0.83 for the Somatic Symptoms subscale, 0.85 for the Interpersonal Symptoms subscale, and 0.94 for the three subscales combined.

Resilience

We used a 14-item resilience scale [64] to assess “the capacity to withstand life stressors, and to thrive and make meaning from challenges” [65 Sample items included “I usually manage one way or another” and “I usually find something to laugh about.” Response options ranged from 1 (“Strongly disagree”) to 6 (“Strongly agree”), with summed higher scores indicating more resilience. Cronbach’s alpha for this measure at Time 3 was 0.95.

HIV-Related Social Support

We used seven items adapted from the Perceived Social Support from Friends and Family [66] to measure the extent to which participants perceived that they could rely on their “Black gay/bi male friends” for emotional, informational, material, and other support for living with HIV. Sample items included “My Black gay/bi male friend(s) help(s) me deal with the emotional burden of being HIV-positive” and “I can turn to my Black gay/bi male friend(s) if I need help getting to my health care provider.” Response options ranged from 1 (“Disagree strongly”) to 6 (“Agree strongly”), with higher summed scores indicating greater HIV-related social support. Cronbach’s alpha for this measure at Time 3 was 0.94.

Healthcare Empowerment

We used the four-item Informed, Committed, Collaborative, and Engaged subscale of the Healthcare Empowerment Inventory [67] to measure the extent to which YBSMM felt engaged and empowered in their healthcare. Sample items included “I prefer to get as much information as possible about treatment options” and “I am very active in my health care.” Response options ranged from 1 (“Strongly disagree”) to 5 (“Strongly agree”), with higher summed scores reflecting greater healthcare empowerment. Cronbach’s alpha for this measure at Time 3 was 0.79.

Engagement in HIV Care

We used the US government guidelines for HIV care, which delineate the following stages of engagement in the HIV continuum of care once a patient has been diagnosed with HIV: linkage to HIV primary care, treatment for HIV, retention in HIV care, and viral suppression [68]. Consistent with prior studies, participants’ responses to various questions in our survey served as binary (no/yes) indicators or proxies of engaging in HIV care. The coefficient omega for this measure at Time 3 was 0.82.

Specifically, engagement in HIV care was assessed as follows:

For whether participants had been engaged in HIV primary care, we asked, “In the past 12 months, how many times have you received primary care for HIV (not in an ER or during an in-patient hospital stay)?” Participants who responded that they received primary care only once, had no primary care in the past 12 months, or were never in primary care for HIV were scored 0. Participants who reported receiving primary care more than once were scored 1.

For whether participants were on ART, we asked, “Are you currently taking HIV medications?” Participants who responded no were scored 0, and those who responded yes were scored 1.

For whether participants were retained in HIV care (which we define as having had two viral load tests within the past 12 months), we asked, “In the past 12 months, how many times has your blood been taken for an HIV viral load test (a blood test that measures how much HIV is in your blood)?” Participants who responded with zero or one were scored 0, and those who responded with greater than one were scored 1.

For whether participants had an undetectable viral load, we asked, “What was the result of your most recent viral load test?” Participants who responded “detectable” or “don’t know” were scored 0 for detectable. Those who responded “undetectable” were scored 1.

Preliminary Analyses

In our preliminary analyses, we conducted descriptive statistics in Stata 16 [69]. Specifically, we examined participant characteristics at the time of their enrollment at baseline, Time 1, based on means and standard deviations (SDs) for age and frequencies and percentages for education, income, and employment status. We also provide descriptive statistics of study variables in the form of means and SDs for continuous study variables (i.e., socioeconomic distress; intimate partner violence by a male partner in the past year; depressive symptomology, specifically, negative affect, somatic symptoms, interpersonal symptoms; resilience; HIV-related social support; healthcare empowerment) and frequencies and percentages for the categorical study variables reflecting engagement in HIV care (i.e., on ART, retained in HIV care, undetectable viral load).

Structural Equation Modeling

We performed structural equation modeling (SEM) using Mplus 7 [70] to identify significant direct and indirect effects of stressors (i.e., socioeconomic distress, intimate partner violence), depressive symptoms, and protective factors (individual resilience, HIV-related social support) on engagement in HIV care via the mediator, healthcare empowerment. Given the presence of ordinal indicators for the continuous latent variables, we estimated model parameters using mean–variance adjusted weighted least squares estimation (TYPE = WLSMV in Mplus) and Delta parameterization [70]. Good model fit was defined as meeting any two of the following three criteria: a root mean square standard error of approximation (RMSEA) value of 0.06, a comparative fit index (CFI) value of at least 0.95, and a standardized root mean square residual (SRMR) criterion of 0.08 or less [71, 72].

Prior to the regression models, we created measurement models, or latent variables, of each primary study variable (i.e., socioeconomic distress, intimate partner violence, depressive symptoms, resilience, HIV-related social support, healthcare empowerment) using confirmatory factor analysis to account for measurement error [73]. Upon determining that each latent variable fit the data, we then tested SEM models of engagement in HIV care, the outcome variable, independently regressed on socioeconomic distress, intimate partner violence, depressive symptoms, resilience, HIV-related social support, and healthcare empowerment in separate, bivariable models. Next, we tested a multivariable SEM of engagement in HIV care simultaneously regressed on all other latent variables. Finally, we tested the mediation model of the direct and indirect effects of the stressors (i.e., socioeconomic distress, intimate partner violence), depressive symptoms, and resiliencies (i.e., resilience, HIV-related social support) on engagement in HIV care via the mediator, healthcare empowerment.

For the confirmatory factor analyses, we reported standardized lambda (λ) and R² values for the factor loadings. For the SEM analyses, we reported unstandardized regression coefficients (b), their standard errors (SE), and p-values in the text. We reported standardized regression coefficients (b*) in the figures. For indirect and direct effects in the mediation model, we reported bs for “indirect” regression coefficients (b_indirect) and bs and b*s for “direct” regression coefficients (b_direct). We also reported R² values for all outcome variables. We reported these statistics in the text, the tables, or both.

Due to the limited sample size, the criterion for the alpha value at which the null hypothesis was rejected was set to 0.10. We handled missing data using multiple imputations. These analyses were cross-sectional, not longitudinal; as such, they were testing for correlations among the variables rather than temporal associations or causation. Only Time 3 data were included in the present study, as some measures were added at Time 3 only and, thus, were not available at other time points.

Results

Table 1 displays the participant characteristics at the time of enrollment at baseline and descriptive statistics for study variables at Time 3. As shown in Table 2, each measurement model fits the data based on their RMSEA, CFI, or SRMR values. Additionally, all constituent items were loaded onto their respective latent variables, and all latent variables accounted for up to 96% of the variance in their constituent items.

Table 1.

Time 3 participant characteristics and descriptive statistics at their baseline assessment (N = 224)

Variable	Descriptive statistics
Participant characteristics at baseline
	Mean (SD)	Range
Age in years at baseline	24.33 (2.66)	18–29
Highest level of education completed
		n (%)
Less than high school diploma or GED		1 (0.5)
High school diploma or GED		94 (42.0)
Some college		107 (47.8)
College degree		16 (7.2)
Any graduate studies		2 (0.9)
Personal income last year
Less than $10,000		52 (23.2)
$10,000–$19,999		44 (19.6)
$20,000–$39,999		79 (35.3)
$40,000–$59,999		29 (13.0)
$60,000 or more		2 (0.9)
Employment status
Employed full-time		124 (55.4)
Employed part-time		36 (16.1)
Unemployed		56 (25.0)
On disability		7 (3.1)
Descriptive statistics of primary study variables
	Mean (SD)	Range
Socioeconomic distress	7.28 (2.10)	3–12
Intimate partner violence by a male partner in the past year	3.28 (6.12)	0–36
Depressive symptomology
Negative affect symptoms	6.55 (35.42)	0–21
Somatic symptoms	7.33 (26.95)	0–21
Interpersonal symptoms	1.65 (3.52)	0–6
Resilience	72.9 (12.8)	19–84
HIV-related social support	31.1 (9.8)	7–42
Healthcare empowerment	17.23 (3.01)	4–20
Engagement in HIV care (%)
		n (%)
On antiretroviral therapy		184 (83.3)
Retained in HIV care		187 (85.4)
Undetectable viral load (virally suppressed)		173 (78.6)

Frequencies may not sum to 224 and percentages may not sum to 100 due to missing data or rounding

Table 2.

Confirmatory factor analyses of primary study variables: Socioeconomic distress, intimate partner violence by a male partner, depressive symptoms, resilience, HIV-related social support, and healthcare empowerment with engagement in HIV care at Time 3 (N = 224)

Variable	λstandardized (range)	R2 (range)	RMSEA	CFI	SRMR
Confirmatory factor analyses
Socioeconomic distress	0.42–0.98	0.17–0.96	< 0.001	1.00	< 0.001
Intimate partner violence by a male partner	0.74–0.91	0.54–0.83	0.11	.98	0.04
Depressive symptoms (by subscale)	0.75–0.97	0.56–0.94	< 0.001	1.00	< 0.001
Resilience	0.69–0.92	0.47–0.84	0.07	0.98	0.03
HIV-related social support	0.72–0.95	0.51–0.89	0.18	0.99	0.02
Healthcare empowerment	0.66–0.88	0.43–0.77	0.12	0.99	0.02
Engagement in HIV care	0.75–0.81	0.57–0.66	< 0.001	1.00	< 0.001

For depressive symptoms, the following subscales were used as indicators: negative affect, somatic symptoms, and interpersonal symptoms. Given the presence of original-scaled items, or indicators, we estimated model parameters using weighted least squares estimation (TYPE = WLSMV in Mplus) and Delta parameterization [70]. Each latent variable fit the data adequately based on the RMSEA, CFI, or SRMR values or was fully saturated

We then evaluated associations of each latent independent variable (i.e., socioeconomic, distress, intimate partner violence, depressive symptoms, resilience, HIV-related social support, healthcare empowerment) with the latent outcome variable of engaging in HIV care in separate models. Table 3 shows bivariate associations with regression coefficients between YBSMM who reported higher levels of socioeconomic distress (b = −0.25, SE = 0.10, b* = −0.21, p = 0.01, R² = 0.87), intimate partner violence (b = −0.23, SE = 0.09, b* = −0.30, p = 0.01, R² = 0.09), and depressive symptoms (b = −0.05, SE = 0.01, b* = −0.40, p < 0.001, R² = 0.16) reported less engagement in HIV care in the past 12 months than participants who indicated lower levels of socioeconomic distress, intimate partner violence, and depressive symptoms, respectively. In contrast, YBSMM who endorsed higher levels of resilience (b = 0.31, SE = 0.11, b* = 0.31, p = 0.01, R² = 0.10) and healthcare empowerment (b = 0.68, SE = 0.14, b* = 0.62, p < 0.001, R² = 0.38) reported greater engagement in HIV care in the past 12 months than YBSMM who endorsed lower levels of resilience and healthcare empowerment, respectively. HIV-related social support was not independently associated with engagement in HIV care (b = 0.11, SE = 0.11, b* = 0.13, p = 0.32, R² = 0.02).

Table 3.

Structural equation models of the associations of socioeconomic distress, intimate partner violence by a male partner, depressive symptoms, resilience, and HIV-related social support with engaging in HIV care via healthcare empowerment with engagement in HIV care at Time 3 (N = 224)

Variable	b	SE	b*	P	R 2	RMSEA	CFI	SRMR
Two-factor SEM models of engaging in HIV care
Socioeconomic distress	−0.25	0.10	−0.29	0.01	0.09	0.07	0.99	0.06
Intimate partner violence	−0.23	0.09	−0.30	0.01	0.09	0.06	0.99	0.06
Depressive symptoms	−0.05	0.01	−0.40	<0.001	0.16	0.01	0.98	0.04
Resilience	0.31	0.11	0.31	0.01	0.10	0.05	0.99	0.05
HIV-related social support	0.11	0.11	0.13	0.32	0.02	0.09	0.99	0.06
Healthcare empowerment	0.69	0.14	0.62	<0.001	0.38	0.05	0.99	0.05
Multi-factor SEM of engaging in HIV care without mediation					0.42	0.04	0.98	0.07
Socioeconomic distress	−0.10	0.10	−0.12	0.30
Intimate partner violence	0.08	0.16	0.08	0.63
Depressive symptoms	−0.03	0.02	−0.25	0.08
Resilience	−0.10	0.13	−0.10	0.45
HIV-related social support	−0.01	0.11	−0.01	0.94
Healthcare empowerment	0.56	0.18	0.57	<0.001
Multi-factor SEM of engaging in HIV care with mediation via healthcare empowerment					0.43	0.04	0.98	0.07
Socioeconomic distress
Effect on mediator	−0.04	0.08	−0.05	0.56
Direct effect on outcome	−0.10	0.10	−0.13	0.28
Indirect effect on outcome	−0.03	0.05	N/A	0.58
Intimate partner violence
Effect on mediator	−0.16	0.12	−0.18	0.16
Direct effect on outcome	0.08	0.16	0.09	0.60
Indirect effect on outcome	−0.09	0.07	N/A	0.21
Depressive symptoms
Effect on mediator	−0.01	0.02	−0.05	0.68
Direct effect on outcome	−0.03	0.02	−0.26	0.07
Indirect effect on outcome	−0.003	0.01	N/A	0.69
Resilience
Effect on mediator	0.34	0.11	0.34	0.001
Direct effect on outcome	−0.11	0.14	−0.11	0.40
Indirect effect on outcome	0.20	0.08	N/A	0.02
HIV-related social support
Effect on mediator	0.15	0.07	0.19	0.03
Direct effect on outcome	0.01	0.11	0.01	0.96
Indirect effect on outcome	0.09	0.05	N/A	0.08
Healthcare empowerment (mediator)	0.57	0.18	0.58	<0.001

Given the sample size and the exploratory nature of these analyses, the criterion for hypothesis testing is α = 0.10 instead of 0.05. b = unstandardized regression coefficient. SE, standard error for unstandardized coefficient. b*, standardized regression coefficient. Rows in bold font indicate significant effects, including direct and indirect effects via the mediating role of healthcare empowerment on engaging in HIV care. We estimated model parameters with these ordinal-scaled items using mean–variance adjusted weighted least squares estimation (TYPE = WLSMV in Mplus) and Delta parameterization [70]. Both structural equation models fit the data adequately based on the RMSEA, CFI, or SRMR values

In this model, healthcare empowerment mediated the association between resilience and engagement in HIV care and the association between HIV-related social support and engagement in HIV care. Specifically, there was an indirect effect of resilience on engagement in HIV care via healthcare empowerment, such that YBSMM who reported higher levels of resilience endorsed higher levels of engagement in HIV care; this association was accounted for by higher reported levels of healthcare empowerment (b_indirect = 0.20, SE = 0.08, p = 0.02). There was also a direct effect of HIV-related social support on resilience and an indirect effect of HIV-related social support on engagement in HIV care via healthcare empowerment. YBSMM who reported higher levels of HIV-related social support and higher levels of resilience also reported greater engagement in HIV care; this association was accounted for by greater reported levels of healthcare empowerment (b_indirect = 0.09, SE = 0.05, p = 0.08). However, neither resilience (b_direct = 0.34, SE = 0.11, p = 0.001) nor HIV-related social support (b_direct = 0.15, SE = 0.07, p = 0.03) were directly associated with engagement in HIV care once healthcare empowerment was added to the model.

Additionally, in the mediation model shown in Fig. 2, depressive symptoms were directly associated with engagement in HIV care such that YBSMM who endorsed higher levels of depressive symptoms (relative to lower levels) reported less engagement in HIV care (b_direct = −0.03, SE = 0.02, p = 0.07). However, depressive symptoms were not indirectly associated with engagement with HIV care via healthcare empowerment (b_indirect = −0.003, SE = 0.01, p = 0.69). In contrast, socioeconomic distress (b_direct = −0.10, SE = 0.10, p = 0.28; b_indirect = −0.03, SE = 0.05, p = 0.58) and intimate partner violence (b_direct = 0.08, SE = 0.16, p = 0.60; b_indirect = −0.09, SE = 0.07, p = 0.21) had no significant direct or indirect association with the outcome, engagement in HIV care, in the mediation model. Both resilience and HIV-related social support were associated with the mediator, healthcare empowerment, but socioeconomic distress, intimate partner violence, and depressive symptoms were not associated with healthcare empowerment. YBSMM who reported higher levels of resilience endorsed higher levels of healthcare empowerment than YBSMM who reported lower levels of resilience (b = 0.34, SE = 0.11, p = 0.001). Additionally, YBSMM who reported higher levels of HIV-related social support indicated higher levels of healthcare empowerment than YBSMM who endorsed lower levels of HIV-related social support (b = 0.15, SE = 0.07, p = 0.03). However, socioeconomic distress (b = −0.04, SE = 0.08, p = 0.56), intimate partner violence (b = −0.16, SE = 0.12, p = 0.16), and depressive symptoms (b = −0.01, SE = 0.02, p = 0.68) were not significantly associated with the mediator, healthcare empowerment.

Fig. 2.

The mediating role of healthcare empowerment in associations with engagement in HIV care among YBSMM (N = 224). Note: Regression coefficients and other parameters are standardized, and significant parameter estimates are in bold font. We estimated model parameters with these ordinal-scaled items using mean–variance adjusted weighted least squares estimation (TYPE = WLSMV in Mplus) and Delta parameterization [70]. ^ϯp < 0.10, *p < 0.05, **p < 0.01, ***p < 0.001

Discussion

This study sought to explore the significance of HIV-related social support and resilience in managing one’s HIV in the face of other social and structural barriers such as socioeconomic distress, intimate partner violence, and depressive symptoms. We assessed the mediating role of healthcare empowerment in the association between HIV-related social support and individual-level resilience in engagement in HIV care among YBSMM living in Dallas, TX, and Houston, TX. Consistent with syndemic theory [74–76], the findings suggest that social support and resilience play independent roles in fostering healthcare empowerment and that each of these three factors plays a crucial role in helping to overcome depressive symptoms and other syndemic factors impacting engagement in HIV care. Therefore, as hypothesized, each of these factors may be a key intervention target for increasing HIV care engagement.

The present study provides valuable insights into the mediating role of resilience and healthcare empowerment in the association between HIV-related social support and engagement in HIV care among YBSMM in the US South. The results suggest that if the goal is to increase healthcare empowerment and ultimately engagement in HIV care, one promising avenue to do so is by increasing HIV-related social support. HIV-related social support may include emotional, informational, material, and other forms of support, each of which has direct relevance to healthcare empowerment. This study suggests that healthcare empowerment, or feeling engaged and empowered in one’s healthcare, appears to be the mechanism of action that increases HIV care engagement among YBSMM in the South. HIV-related social support interventions, where peers help to empower YBSMM to get as involved in their healthcare as possible appear to promote resilience, overcome social and structural barriers, and increase engagement in their HIV care [53, 77].

Moreover, the study findings underscore the importance of fostering resilience in facilitating healthcare empowerment. Resilience, which refers to an individual’s ability to adapt to and bounce back from challenging circumstances, may facilitate the successful navigation of complex healthcare systems and factors that contribute to access barriers (e.g., limited cultural competence, and poor system coordination between medical and social aspects of care). Although we assessed global resilience, manifestations of resilience in the context of HIV care could include strengths such as self-advocacy, more effective communication with healthcare providers, and the ability to identify and pursue additional resources needed to manage HIV effectively (e.g., social support).

At the same time, it is important to recognize that the construct of resilience may be problematic when uncritically applied to marginalized populations, including YBSMM living with HIV, who face multiple and intersecting forms of oppression. A focus on resilience as an individual-level trait without attention to the factors that bolster or deplete resilience can reinforce a deficit-based framing that places blame on individuals for their health outcomes, rather than acknowledging the role that systemic oppression and structural inequalities (e.g., racism, homophobia, HIV stigma) play in shaping health disparities. At social and structural levels, a highly individualistic framing of resilience may also implicitly or explicitly justify the neglect of systemic inequalities and the provision of inadequate resources to marginalized communities. The emphasis on personal resilience can obscure the need for broader changes in policy and societal norms that are necessary to address health disparities. Additionally, these social and structural factors themselves can enhance or deplete individual-level resiliencies [54, 78]. Therefore, addressing them may have the dual positive impacts of mitigating proximal health disparities (e.g., reducing disparities exacerbated by HIV stigma) while also building resiliency as a protective factor relevant to a range of potential stressors.

Our study also incorporated factors known to have negative impacts on wellbeing and care continuum engagement. Syndemic conditions such as socioeconomic distress, intimate partner violence, and depressive symptoms are interrelated and have been shown to have a deleterious impact on HIV care engagement among YBSMM [74–76]. The current findings suggest that while HIV-related social support and resilience can help to overcome the deleterious impact of these syndemic conditions on HIV care engagement, depressive symptoms might be among the most important syndemic factors to consider when looking to address difficulties with HIV care engagement. Interventions seeking to increase care engagement among YBSMM will also need to target depressive symptoms head-on, as these symptoms have a direct effect on engagement in HIV care and also serve to undermine resilience and utilization of HIV-related social support [25, 26].

Socioeconomic distress, intimate partner violence, and depressive symptoms are syndemic factors that may limit access to healthcare and contribute to health disparities. Social support is especially important for YBSMM with HIV, who may face significant stigma, discrimination, economic hardship, violence, and depressive symptoms. The current findings highlight the crucial role of HIV-related social support in overcoming these challenges and in promoting health-seeking behavior and empowering YBSMM to engage with healthcare providers and adhere to their HIV treatment regimens.

Returning to a more specific focus on healthcare engagement and healthcare empowerment, it is important to recognize that the burden of navigating challenging healthcare systems should not fall disproportionately on young Black sexual minority men with HIV. The healthcare system, as it currently exists, is not always designed to meet the specific needs of this community, and it is important to acknowledge the ways in which systemic biases can contribute to disparities in care. For example, research has shown that healthcare providers may hold negative attitudes toward sexual minority individuals, which can lead to a lack of empathy and understanding in clinical settings [79, 80]. Additionally, racial biases may impact the quality of care that Black individuals receive, including persons living with HIV [81]. In tandem with necessary changes to the cultural humility in HIV care, interventions that specifically focus on increasing HIV-related social support, such as support groups, peer mentorship programs, and community-based initiatives may improve engagement in HIV care for YBSMM. These interventions should be culturally responsive and sensitive to the unique experiences of young Black sexual minority men with HIV and should aim to build on the strengths and resources of these individuals.

This study is not without limitations. With cross-sectional data, is not possible to assess the temporality of the associations. Relatedly, it is also not possible to determine the direction of the effects we identified. The model that we tested was empirically informed by the existing body of cross-sectional research in this area; however, other explanations are plausible. For example, while HIV-related social support may engender resilience (e.g., peer support is protective against risk), peer social support may also be an outcome of resilience (e.g., other resilient traits engender strong social relationships, and thus social support). Future research, especially longitudinal research, should seek to delineate the temporality and direction of associations between syndemic factors, HIV-related social support, resilience, healthcare empowerment, and HIV care engagement. Further, we used latent variables to account for measurement error, and these variables often require larger sample sizes than what was available in the present study. Finally, engagement in HIV care in the present study was based on self-report. Future studies should use medical records or biomarkers to assess HIV care engagement.

The findings of this study emphasize the critical role that HIV-related social support plays in fostering resilience and overcoming syndemic factors to promote engagement in HIV care among YBSMM in the US South. Interventions that aim to promote HIV-related social support and resilience can help to reduce health disparities and improve health outcomes among YBSMM with HIV. By recognizing the critical role of social support in fostering resilience and healthcare empowerment, healthcare providers and policymakers can develop individual and community interventions and strategies that support YBSMM in managing their HIV effectively, in tandem with improvements to the systems that deliver HIV care.