Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Apr 1.
Published in final edited form as: J Behav Med. 2022 Jan 13;45(2):285–296. doi: 10.1007/s10865-021-00277-z

Perceived Discrimination, Adherence to Antiretroviral Therapy, and HIV Care Engagement Among HIV-Positive Black Adults: The Mediating Role of Medical Mistrust

Lu Dong 1, Laura M Bogart 1, Matt G Mutchler 2, Sean J Lawrence 2, David J Klein 1, Mahlet Gizaw 1, Glenn J Wagner 1
PMCID: PMC8957584  NIHMSID: NIHMS1771906  PMID: 35028783

Abstract

Perceived discrimination and medical mistrust are contributors to HIV inequities. The current study examined whether medical mistrust mediated the associations between perceived discrimination and adherence to antiretroviral therapy (ART) as well as care engagement in a sample of 304 Black adults living with HIV. Perceived discrimination and medical mistrust were measured using validated scales; ART adherence was electronically monitored for a month; care engagement was determined by medical record data. Results support significant total indirect effects from perceived discrimination (due to HIV-serostatus, race, sexual orientation) to ART adherence through three types of medical mistrust (towards healthcare organizations, one’s physician, and HIV-specific mistrust). The total indirect effects were also significant for care engagement and were largely driven by mistrust towards one’s own physician. Findings suggest interventions at the provider or healthcare organization levels should address medical mistrust to improve the health and well-being of Black Americans living with HIV.

Keywords: perceived discrimination, intersectional stigma, medical mistrust, ART adherence, care engagement

Introduction

Black Americans are disproportionately affected by the HIV epidemic in the United States. In 2018, Black people accounted for 42% of all new HIV diagnoses in the United States; specifically, Black gay and bisexual men had the highest percentage (26%) of new HIV diagnoses compared to other groups (Centers for Disease Control and Prevention, 2020). Black or African Americans also made up the largest percentage (41%) of people living with HIV at the end of 2018 (Centers for Disease Control and Prevention, 2020). Compared to other racial/ethnic groups, they are less likely to receive and stay engaged in HIV care, have pre-exposure prophylaxis (PrEP) for HIV prevention, adhere to antiretroviral treatment, and be virally suppressed (Centers for Disease Control and Prevention, 2020; Lyons, Dailey, Yu, & Johnson, 2021).

A major contributor to the significant HIV-related inequities affecting Black Americans is the experience of discrimination associated with intersectional stigma, which refers to the intersection of devalued social categories (e.g., minority race, sexual minority orientation, HIV-serostatus) at the individual level as a reflection of interlocking systems of privilege and oppression at the structural level (e.g., racism, sexism, heterosexism) (Bowleg, 2012; Turan et al., 2019). The negative impact of discrimination on health has been well-documented (Williams, Lawrence, & Davis, 2019). Specifically, the stress associated with discrimination contributes to the health disparities of racial/ethnic and sexual minorities (Ren, Amick, & Williams, 1999; Thomas, Bardwell, Ancoli-Israel, & Dimsdale, 2006; Williams, Yu, Jackson, & Anderson, 1997), including those living with HIV (Braksmajer, Simmons, Aidala, & McMahon, 2018; Earnshaw, Bogart, Dovidio, & Williams, 2015; Rintamaki, Davis, Skripkauskas, Bennett, & Wolf, 2006). For example, in a sample of 181 HIV-positive Black participants in Los Angeles who experienced greater racial discrimination were more likely to have lower CD4 cell counts, higher HIV viral loads, and any emergency department visits in the past six months (Bogart, Landrine, Galvan, Wagner, & Klein, 2013). Greater emergency department use is indicative of poor health, possibly due to delayed help-seeking and decreased primary care engagement (Bogart et al., 2013).

Medical mistrust is also considered a key contributor to health disparities that Black Americans experience in general and specific to HIV-related outcomes (Bogart, Takada, & Cunningham, 2021; Institute of Medicine Committee on, Eliminating, & Ethnic Disparities in Health, 2003). Medical mistrust refers to the distrust of the healthcare system, providers, and treatments (Hall, Dugan, Zheng, & Mishra, 2001; Omodei & McLennan, 2000). It is considered an active and understandable response to direct or vicarious experiences of discrimination and oppression (Benkert, Cuevas, Thompson, Dove-Meadows, & Knuckles, 2019; Bogart, Takada, et al., 2021). Empirical studies support that perceived racial discrimination both within and outside healthcare is related to medical mistrust among Black Americans (Durant, Legedza, Marcantonio, Freeman, & Landon, 2011; Hammond, 2010; Williamson, Smith, & Bigman, 2019). There is abundant literature on the negative impact of medical mistrust on health behaviors and outcomes and its role in racial health inequities (Institute of Medicine Committee on Understanding and Eliminating Racial and Ethnic Disparities in Health Care, 2003; Office of Disease Prevention and Health Promotion, 2020; Williamson & Bigman, 2018), including the role of medical mistrust in poor HIV outcomes among racial/ethnic minority people living with HIV (for a review, see Bogart, Takada, et al., 2021). For example, greater physician mistrust was associated with lower readiness for antiretroviral therapy (ART) in a sample of 95 HIV-positive African American and Latinx patients from hospital-based HIV clinics in New York City (Gwadz et al., 2014). In a sample of HIV-positive Black men, general medical mistrust was prospectively associated with lower ART adherence (Dale, Bogart, Wagner, Galvan, & Klein, 2016). In addition, general medical mistrust was associated with lower self-reported engagement in routine health care such as annual physical examination in Black sexual minority men (Eaton et al., 2015). In addition, race-based medical provider mistrust was associated with lower ART adherence in racial/ethnic minority adults (Kalichman et al., 2016), including in Black adults living with HIV (Kalichman, Eaton, Kalichman, & Cherry, 2017).

A few studies have explored whether medical mistrust explains the association between perceived discrimination and poor health outcomes. For example, in a sample of HIV-positive Latinx men, general medical mistrust was found to mediate the associations between two types of perceived discrimination, related to being Latino and being HIV-positive, and self-reported adherence to ART (Galvan, Bogart, Klein, Wagner, & Chen, 2017). This potential mediation path has also been minimally explored in other health conditions. One study reported that racial discrimination was associated with lower self-reported medication adherence in a sample of Black adults with hypertension, and trust in physicians mediated the association (Cuffee et al., 2013). Another study found that general medical mistrust mediated the association between enacted stigma (sexual orientation and race-based) related to health care and engagement with annual physical examination in a sample of Black sexual minority men (Eaton et al., 2015). No prior study has examined the association between perceived discrimination and prospectively and objectively measured medication adherence, either HIV-specific or not, and few studies have extended the association to include other outcomes such as care engagement.

The current study aimed to examine whether medical mistrust mediates the association between perceived discrimination and adherence to ART as well as HIV care engagement in a sample of Black men and women living with HIV. We hypothesized that there would be significant total indirect effects between high perceived discrimination and poor ART adherence (or care engagement) via greater medical mistrust, in the forms of three specific indirect effects: mistrust towards healthcare organizations, physicians, and HIV-specific mistrust beliefs (also known as HIV conspiracy beliefs). HIV-specific medical mistrust (i.e., medical mistrust about the origin, prevention, and treatment of HIV) was added because the sample included all HIV-positive participants and, as with mistrust towards healthcare organizations and physicians, it is robustly associated with health outcomes and behaviors related to HIV prevention, testing, and ART nonadherence in Black individuals (Bogart, Takada, et al., 2021).

The present study was conducted in Los Angeles County, California. As of 2018, there were 58,000 individuals living with HIV/AIDS in Los Angeles County, out of which approximately 20% were Black individuals (Division of HIV and STD Programs Department of Public Health County of Los Angeles, 2020). From 2006 to 2018, Black men and women had the highest rates of HIV diagnoses as well as lowest treatment coverage compared to all other racial/ethnic groups (Division of HIV and STD Programs Department of Public Health County of Los Angeles, 2020). HIV prevalence rates from 2004 to 2017 among Black sexual minority men (36% in 2017) in Los Angeles County were consistently and significantly higher than White (15% in 2017) and Latinx (18%) sexual minority men (Sey & Ma, 2017).

Methods

Participants and Procedures

Data collection for the present analysis was conducted from January 23, 2018 to July 2, 2020 at AIDS Project Los Angeles (APLA) Health, the largest community-based HIV service organization in Los Angeles County. The Community Advisory Board (CAB) at APLA Health met several times a year during the study period to provide input on study design, recruitment, and interpretation of results. The CAB members were clients and staff from APLA Health and other local organizations primarily serving Black people living with HIV.

The current study used baseline data (N = 304) from an ongoing randomized controlled trial of a culturally congruent behavioral intervention to improve ART adherence, retention in care, and viral suppression in Black adults living with HIV. Inclusion criteria for the study were: (1) Black/African American racial/ethnic identity, (2) HIV-positive serostatus, (3) 18 years of age or above, (4) prescribed ART in the past 12 months, (5) self-reported adherence problems (i.e., missed at least one ART dose in the past month) and/or detectable viral load; and (6) willing to use an electronic adherence monitoring device. Institutional review board approval was obtained from the institution’s Human Subjects Protection Committee. All participants gave informed consent and received $30 for the baseline assessment and $30 for a one-month follow-up visit. During the follow-up visit, study staff downloaded electronically monitored adherence data (as described below) and conducted blood draws via venipuncture to assess HIV viral load. Study staff were certified in phlebotomy and conducted venipuncture at the primary care clinic connected to APLA Health.

Measures

Socio-demographic characteristics.

A self-report questionnaire was used to obtain information about participant's age (calculated from date of birth), gender identification (cis or transgender male or cis or transgender female, gender queer/gender nonconforming, or different identity), sexual orientation, education level, housing status in the past year, current employment status, and number of years since HIV diagnosis.

Medication Events Monitoring System (MEMS).

Adherence to ART medication was measured continuously from the baseline assessment time-point to 1-month (30 days) after baseline using MEMS (AARDEX, Inc., Zurich, Switzerland). The electronic drug exposure monitoring cap records the time and date when the medication bottle is opened, and data are automatically saved in the software to produce reports of daily medication-taking patterns. At the 1-month follow-up visit, MEMS data for each participant were downloaded. Participants also completed a questionnaire about how often they 1) opened the bottle without removing a dose, 2) took a dose from another source other than the MEMS bottle, and 3) removed more than one dose at a time from the bottle over the past month. In the current sample, 21% reported using these strategies. Their MEMS data were adjusted accordingly subtracting MEMS dose counts from when they opened the bottle without removing a dose, and adding counts when they took a dose from another source other than the MEMS bottle or removed more than one dose at a time from the bottle based on questionnaire data (Bangsberg, Hecht, Charlebois, Chesney, & Moss, 2001). A continuous adherence measure (percentage of prescribed doses taken in the past month) was derived for the 30-day period between baseline and 1-month follow-up. We also report the percentage with 80% or more adherence. The continuous MEMS adherence measure was used in the analysis. The analytic sample with complete MEMS adherence data includes 230 participants.

HIV viral load was obtained via venipuncture for 196 participants, and self-reported viral load was obtained for 282 participants. MEMS adherence from baseline to 1-month follow-up was significantly associated with the binary variable of undetectable viral load from blood samples at baseline (OR = 6.70, p = 0.009, n = 187), providing support for the validity of the MEMS adherence measure. MEMS adherence from baseline to 1-month follow-up was not significantly associated with self-reported undetectable viral load at baseline (OR = 3.28, p = 0.056, n = 212).

Multiple Discrimination Scale.

Perceived discrimination was measured using experiences with ten different discrimination events in the past year due to being HIV-positive, being Black, and being gay (Bogart et al., 2013). Sample items include: "In the past year, were you ignored, excluded, or avoided by people close to you because you are [HIV positive/Black/gay]?" "In the past year, were you denied a place to live or did you lose a place to live because you are [HIV positive/Black/gay]?", and "In the past year, were you physically assaulted or beaten up because you are [HIV positive/Black/gay]?" For each subscale, a mean score was created, averaging across all ten items. Reliability coefficients (α) for the three subscales in the current study were 0.87 (HIV-positive), 0.91 (Black), and 0.92 (gay).

Medical Mistrust.

Medical mistrust was measured using three subscales to capture mistrust towards health care organization, physician, and HIV-specific mistrust beliefs. The mistrust toward health care organizations subscale (LaVeist, Isaac, & Williams, 2009) includes seven items such as "patients have sometimes been deceived or misled by health care organizations," "health care organizations have sometimes done harmful experiments on patients without their knowledge," and "mistakes are common in health care organizations." Items were rated on a 4-point scale: strongly disagree, disagree, agree, and strongly agree. Reliability coefficient (Cronbach’s α) for this subscale in the current study was 0.81.

The mistrust toward physician subscale includes (Thom, Ribisl, Stewart, Luke, & Physicians, 1999) 11 items such as "I trust my doctor so much I always try to follow his/her advice," "I sometimes distrust my doctor's opinions and would like a second one," and "I feel my doctor does not do everything he/she should about my medical care." Items were also rated on a 5-point scale: totally disagree, disagree, neutral, agree, and strongly agree. Reliability coefficient (α) for this subscale in the current study was 0.89.

The HIV conspiracy beliefs subscale (Bogart & Thorburn, 2005) measures HIV-specific mistrust beliefs and includes nine items such as "HIV is a man-made virus," "People who take antiretroviral medications for HIV are human guinea pigs for the government," and "The medication used to treat HIV causes people to get AIDS." Items were rated on a 5-point scale: strongly disagree, slightly disagree, not sure or have no opinion, slightly agree, and strongly agree. Reliability coefficient (α) for this subscale in the current study was 0.90. For each of these three mistrust subscales, we created a mean score across all items, and items were reverse coded as needed so that a higher score indicates greater mistrust.

HIV Care Engagement.

Care engagement was determined by combining responses to two survey items, “In the last 6 months, how many appointments did you have with an HIV doctor?” and “In the last 6 months, how many appointments with an HIV doctor did you miss without rescheduling?” It is defined as having one or more HIV care visits and less than two missed appointments in the past six months (Rumptz et al., 2007).

Data Analysis

We used Mplus version 8 to conduct path analysis with parallel multiple mediators (Muthén & Muthén, 2017). Analyses were conducted using the full sample. Maximum likelihood (ML) estimation method was used with data assumed missing at random (MAR). While the sample included subgroups (e.g., heterosexual and female participants), we did not perform subgroup analyses due to small subgroup sample sizes. Figure 1 shows an illustration of the model. All variables were observed variables. There were two outcome variables (Y): MEMS-measured ART adherence was modeled as a continuous outcome variable, and care engagement as a binary outcome variable. There were three independent variables tested in separate models (X): perceived discrimination due to each of the three forms of stigma (from HIV-serostatus, race/Black, and sexual orientation). There were three mediators modeled as paralleled indirect paths (M1-3): three specific forms of medical mistrust (mistrust toward one's individual physician, mistrust toward healthcare organizations, and HIV-related mistrust beliefs). Note that the three mediators were allowed to covary, and they exhibit significant but weak correlations (rs = 0.24-0.34, ps < .0001) We tested whether perceived discrimination due to each of the three forms of stigma was related to ART adherence or care engagement through three specific forms of medical mistrust. Specifically, perceived discrimination due to each form of stigma influenced all three forms of medical mistrust (a1, a2, a3), which in turn influenced ART adherence or care engagement (b1, b2, b3). Specific indirect effects through each parallel mediator were estimated as the corresponding product term (a1*b1, a2*b2, a3*b3). Total indirect effects refer to the sum of all specific indirect effects. Direct effects (c’) refer to the effect of each type of perceived discrimination on ART adherence or care engagement while accounting for the parallel medical mistrust mediators. Total effects refer to the sum of the indirect and direct effects. We reported estimates and 95% bootstrapping confidence intervals (obtained from 10,000 replications) of the specific and total indirect effects as well as direct effects and total effects. Because the current analyses focused on mediation or indirect effects (i.e., whether perceived discrimination indirectly affects adherence or care engagement through medical mistrust), we did not test the model with only the direct path from independent to dependent variables. Recent methodological research suggests that mediation effects can be present in the absence of a total effect, hence, it is not a necessary step to first establish a relationship between perceived discrimination and ART adherence or care engagement as part of the mediation test (Hayes, 2017; Kenny & Judd, 2014).

Figure 1.

Figure 1.

Open in a new tab

Conceptual Diagram for the Parallel Mediation Model.

Results

Table 1 presents the sociodemographic characteristics of the sample. The mean age of the sample was 48 years-old. The majority of the sample were men (81%), sexual minority individuals (74%), and not currently working part-time or full-time (84%). The mean time since HIV diagnosis was 17 years. Table 2 presents data related to the sample’s HIV care adherence and retention, as well as reported levels of discrimination and mistrust. Average MEMS adherence from baseline to 1-month follow-up was 74%, and 53% of the sample had at least 80% MEMS adherence during this period. A total of 78% were engaged in health care (i.e., had one or more visits and less than two missed appointments in the past six months).

Table 1.

Characteristics of the Sample of Black Adults Living with HIV (N = 304).

Variable Obs n (%) / M (SD, range)
Age 304 47.62 (12.46, 18-75)
Time since HIV diagnosis (years) 292 16.65 (9.72, 0.38-39.22)
Female 304 57 (18.75%)
Education 304
 Less than high school 44 (14.47%)
 High school diploma or GED 88 (28.95%)
 Some college, but no degree 99 (32.57%)
 College degree or above 73 (24.00%)
Working full-time or part-time 303 47 (15.51%)
Unstable housing in the past year 303 157 (51.82%)
Household income in past 12 months 303
 Currently have no income 46 (15.38%)
 $1-$10,000 104 (34.78%)
 $10,000-$20,000 116 (38.80%)
 $20,001-$30,000 19 (6.35%)
 $30,001-$40,000 7 (2.34%)
 Over $40,000 7 (2.34%)
Relationship status 303
 Married or living with significant other 33 (10.89%)
 Single/Divorced/Separated/Widowed 270 (89.11%)
Sexual orientation 304
 Straight or Heterosexual 80 (26.32%)
 Gay or homosexual man or same gender loving 171 (56.25%)
 Lesbian or homosexual woman 3 (0.99%)
 Bisexual 40 (13.16%)
 Not sure or in transition 3 (0.99%)
 Something else 7 (2.30%)
Open in a new tab

Table 2.

Descriptive Statistics of Multiple Discrimination, Medical Mistrust, MEMS-Measured ART Adherence, and Care Engagement.

Variables Obs M SD Range
Multiple Discrimination Scale (independent variable)
MDS scale, HIV-serostatus discrimination 302 1.48 2.33 0-10
MDS scale, Black racial discrimination 300 2.05 2.88 0-10
MDS scale, Gay sexual orientation discrimination 282 1.56 2.74 0-10
Medical Mistrust (mediators)
Mistrust of health care organizations 301 2.59 0.58 1-4
Mistrust of one’s physician 303 2.11 0.73 1-4.73
Mean agreement with HIV-related mistrust beliefs 300 2.75 0.96 1-5
Outcome variables
MEMS adherence from baseline to 1 month 230 74% 24% 0-100%
Obs N %
80% MEMS adherence from baseline to 1 month 302 121 53
Care engagement 302 235 78
Open in a new tab

Note. Care engagement is defined as having one or more visits and less than 2 missed appointments in the past 6 months.

As shown in Table 3, for the association between greater perceived discrimination due to HIV-serostatus at baseline and lower 1-month MEMS-measured ART adherence, mediation analysis results show significant total indirect effects (indirect effects −0.006, 95% bootstrapping CI [−0.012, −0.002]) through higher values for the three parallel mistrust mediator variables. Similarly, the total indirect effects were significant for perceived discrimination due to race/ethnicity (indirect effects −0.006, 95% bootstrapping CI [−0.012, −0.002]). For the association between greater perceived discrimination due to sexual orientation and lower adherence, there were only significant direct effects (−0.015, 95% CI [−0.028, −0.001]) but no significant specific nor total indirect effects through medical mistrust mediator variables. Figure 2 shows the path coefficients of two parallel mediation models with perceived discrimination due to HIV-serostatus (2a) and race/ethnicity (2b) as the respective independent variable.

Table 3.

Results from Parallel Mediation Analysis Predicting Electronically Monitored (MEMS) Adherence and HIV Care Engagement with Perceived Discrimination and Medical Mistrust.

Variables Specific and total indirect effects, direct effects,
and total effects		 Estimate S.E. 95% bootstrap CI
1-month MEMS ART adherence (continuous outcome)
Perceived discrimination due to HIV-serostatus Specific indirect effects through mistrust towards healthcare organizations −0.002 0.002 −0.007, 0.002
Specific indirect effects through mistrust towards one's physician −0.002 0.002 −0.006, 0.002
Specific indirect effects through HIV-related mistrust beliefs −0.002 0.002 −0.007, 0.001
Total indirect effects −0.006 0.002 −0.012, −0.002
Direct effects −0.017 0.009 −0.033, 0.002
Total effects −0.023 0.008 −0.038, −0.007
Perceived discrimination due to race (Black) Specific indirect effects through mistrust towards healthcare organization −0.002 0.002 −0.007, 0.002
Specific indirect effects through mistrust towards one's physician −0.001 0.002 −0.005, 0.002
Specific indirect effects through HIV-specific mistrust beliefs −0.003 0.002 −0.006, 0.000
Total indirect effects −0.006 0.003 −0.012, −0.002
Direct effects −0.009 0.007 −0.022, 0.005
Total effects −0.016 0.006 −0.028, −0.003
Perceived discrimination due to sexual orientation Specific indirect effects through mistrust towards healthcare organization −0.001 0.001 −0.004, 0.001
Specific indirect effects through mistrust towards one's physician −0.001 0.002 −0.005, 0.002
HIV-related mistrust beliefs −0.002 0.001 −0.005, 0.000
Total indirect effects −0.004 0.002 −0.008, 0.000
Direct effects −0.015 0.007 −0.028, −0.001
Total effects −0.019 0.007 −0.032, −0.004
Care engagement (binary outcome)
Perceived discrimination due to HIV-serostatus Specific indirect effects through mistrust towards healthcare organization 0.018 0.017 −0.010, 0.062
Specific indirect effects through mistrust towards one's physician −0.047 0.022 −0.110, −0.015
Specific indirect effects through HIV-related mistrust beliefs −0.030 0.018 −0.074, −0.002
Total indirect effects −0.060 0.026 −0.116, −0.014
Direct effects −0.087 0.060 −0.212, 0.023
Total effects −0.147 0.058 −0.271, −0.041
Perceived discrimination due to race (Black) Specific indirect effects through mistrust towards healthcare organization 0.019 0.019 −0.016, 0.059
Specific indirect effects through mistrust towards one's physician −0.047 0.020 −0.094, −0.014
Specific indirect effects through HIV-related mistrust beliefs −0.025 0.016 −0.060, 0.004
Total indirect effects −0.053 0.026 −0.108, −0.006
Direct effects −0.047 0.058 −0.156, 0.074
Total effects −0.099 0.051 −0.196, 0.005
Perceived discrimination due to sexual orientation Specific indirect effects through mistrust towards healthcare organization 0.004 0.008 −0.008, 0.027
Specific indirect effects through mistrust towards one's physician −0.043 0.020 −0.092, −0.012
Specific indirect effects through HIV-related mistrust beliefs −0.013 0.011 −0.046, 0.000
Total indirect effects −0.052 0.021 −0.101, −0.018
Direct effects −0.074 0.055 −0.179, 0.040
Total effects −0.126 0.054 −0.230, −0.019
Open in a new tab

Note. Significant specific and total indirect effects, direct effects, and total effects are presented in bold. MEMS = medication event monitoring system. Parallel mediators include: 1) mistrust towards healthcare organization, 2) mistrust towards physicians, and 3) HIV-related mistrust beliefs. 95% bootstrap CI = 95% bootstrap confidence interval (10,000 replications).

Figure 2. Parallel Mediation Model for MEMS-Measured ART Adherence.

Figure 2.

Open in a new tab

(2a). Perceived discrimination due to HIV-serostatus and ART adherence

(2b). Perceived discrimination due to race (Black) and ART adherence

Note. Significant paths are shown in bold. Values in parenthesis indicate 95% Bootstrap Confidence Interval.

We also tested the parallel mediation model using HIV care engagement as the outcome variable. Table 3 shows that the total indirect effects through the three mistrust mediator variables were significant for the relationship between greater perceived discrimination due to HIV-serostatus (indirect effects −0.060, 95% CI: [−0.1116, −0.014]), race/ethnicity (−0.053, 95% CI: [−0.108, −0.006]), or sexual orientation (−0.052, 95% CI: [−0.101, −0.018]) and lower care engagement. The specific indirect path through mistrust of one's physician was significant for the associations between perceived discrimination due to each of the three types of discrimination and care engagement. Specifically, higher levels of perceived discrimination due to each of the three types were associated with higher levels of medical mistrust toward one's individual physician, which in turn was associated with lower care engagement. In addition, HIV-specific mistrust beliefs also showed a significant indirect effect between greater perceived discrimination due to HIV-serostatus and poor care engagement. None of the direct effects were significant between perceived discrimination and care engagement. Figure 3 shows an illustration of the parallel mediation model with perceived discrimination due to HIV-serostatus as the independent variable.

Figure 3. Parallel Mediation Model for Care Engagement.

Figure 3.

Open in a new tab

Note. Significant paths are shown in bold. Values in parenthesis indicate 95% Bootstrap Confidence Interval.

Discussion

In a sample of HIV-positive Black adults, who were mostly sexual minority men, we found that medical mistrust explains the association between perceived discrimination and nonadherence to ART, as well as HIV care engagement. Previously, this mediation model had only been tested in a sample of HIV-positive Latino sexual minority men for the outcome of self-reported ART adherence (Galvan et al., 2017). The current study extends prior evidence by examining HIV care engagement as the outcome and using prospectively collected electronically monitored ART adherence, and is among the first to test this mediation model in a sample of HIV-positive Black Americans.

Our results also support strong associations between perceived discrimination due to different intersecting identities (HIV-serostatus, Black/African American race, and minority sexual orientation) and medical mistrust. The results are consistent with other studies suggesting that experiences of discrimination contribute to greater medical mistrust in Black Americans generally and those living with HIV (Hammond, 2010; Hausmann, Kwoh, Hannon, & Ibrahim, 2013; Williamson et al., 2019). Although not examined in the current study, it is important to acknowledge that structural discrimination, such as inequitable policies and practices at the societal level and in healthcare systems specifically, represents another important pathway besides interpersonal discrimination (the focus of the current study) to greater medical mistrust among Black Americans (Bogart, Takada, & Cunningham, 2019).

In prior studies, greater medical mistrust was associated with lower electronically monitored ART adherence in HIV-positive Black men (Bogart, Wagner, Galvan, & Banks, 2010; Dale et al., 2016) and self-reported ART adherence among adults living with HIV (Thrasher, Earp, Golin, & Zimmer, 2008). In the current study, we focused on testing medical mistrust as indirect paths (i.e., perceived discrimination influencing adherence or care engagement outcome through medical mistrust) and did not specifically replicate prior findings of medical mistrust predicting ART adherence tested in separate regression models. In our parallel mediation model, the individual paths from medical mistrust to ART adherence did not reach statistical significance. It is possible that the strong link between medical mistrust and ART adherence was attenuated when perceived discrimination and different types of medical mistrust were simultaneously accounted for in the same model. However, it is important to note that collectively, the total mediation effect from perceived discrimination through three types of medical mistrust was statistically significant for ART adherence.

For the outcome of care engagement, our findings support that the relationship between discrimination and care engagement was explained by different forms of medical mistrust collectively—and specifically driven by mistrust towards one’s physician. This finding is consistent with a prior report that general medical mistrust mediated the association between health care related stigma and time since last physical exam in a sample of Black sexual minority men (Eaton et al., 2015). Our finding suggests that addressing mistrust towards healthcare providers and increasing trust in the provider-patient relationship may help improve care engagement and thereby other clinical outcomes. A number of healthcare provider interventions have been conducted to increase patients' trust in their own provider through increasing providers' cultural competency and empathy for patients (Rolfe, Cash-Gibson, Car, Sheikh, & McKinstry, 2014; Thom, Tirado, Woon, & McBride, 2006; Tulsky et al., 2011). However, these interventions have not specifically addressed medical mistrust and have generally not shown effects on increasing trust, nor are they specifically tailored for HIV care. Interventions at the provider level as well as healthcare organization level are needed to reduce patients' experience of discrimination within healthcare settings and increase providers' ability to acknowledge and address medical mistrust in a sensitive manner, thereby improving patients' health-related outcomes such as medication adherence, care engagement, and clinical outcomes.

Several limitations should be noted. First, the study used a convenience sample of participants from an ongoing clinical trial. The sample was characterized by multiple social-economic disadvantages, such as unemployment, unstable housing, and low income and education attainment. Hence, the results of the current study may not be generalizable to all Black adults living with HIV in the U.S. Second, indirect effects reported in the current study were small. Although observed indirect effects are often small in social and behavioral science data for a variety of reasons, including using longitudinal design and specifying multiple mediators (Walters, 2018), it is important to consider additional explanatory variables (e.g., psychological and physiological stress responses) that are not examined in the current study and explore alternative ways to examine the meaningfulness of the indirect effects (Walters, 2018). In addition, while ART adherence was prospectively measured, care engagement was assessed at the same time as measures of perceived discrimination and medical mistrust. Therefore, for the models with care engagement as the outcome variable, results were based on cross-sectional data, which precludes any inferences regarding the direction of the associations. Future studies should replicate our findings using care engagement measured longitudinally. Finally, due to small sample sizes of female participants as well as heterosexual participants, we did not perform subgroup analyses or test for statistical interactions by gender identity or sexual orientation. Future studies should examine whether our results, based on a sample with mostly Black sexual minority men, will hold in samples with other intersectional identities.

In sum, the current study provides evidence supporting three forms of medical mistrust (towards healthcare organization, towards one's physician, around the origin and treatment of HIV), either collectively or individually, explained the relationship between perceived discrimination and two outcomes: ART adherence and care engagement. Our results held for each type of perceived discrimination examined, including perceived discrimination due to HIV-serostatus, race, sexual orientation.

Our findings suggest that medical mistrust might be an intervention target, particularly at the provider and healthcare organization levels, to improve the health and well-being of Black people living with HIV and reduce HIV-related inequities in this population. Interventions could be developed to train healthcare providers to acknowledge systematic inequity and structural racism as root causes of mistrust in a non-judgmental, non-confrontational, genuine, and empathetic manner. Such intervention strategies have been used by peer counselors to improve adherence at the patient level for people living with HIV (Bogart et al., 2017; Wagner et al., 2016). In particular, future research will need to develop and evaluate interventions to increase the trustworthiness of healthcare providers and healthcare organizations, for example, by investing in authentic and long-term community engagement efforts to build community partnerships (Bogart, Dong, et al., 2021).

References

  1. Bangsberg DR, Hecht FM, Charlebois ED, Chesney M, & Moss A (2001). Comparing objective measures of adherence to HIV antiretroviral therapy: electronic medication monitors and unannounced pill counts. AIDS and Behavior, 5(3), 275–281. [Google Scholar]
  2. Benkert R, Cuevas A, Thompson HS, Dove-Meadows E, & Knuckles D (2019). Ubiquitous Yet Unclear: A Systematic Review of Medical Mistrust. Behav Med, 45(2), 86–101. doi: 10.1080/08964289.2019.1588220 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bogart LM, Dong L, Gandhi P, Ryan S, Smith TL, Klein DJ, … Ojikutu BO (2021). What Contributes to COVID-19 Vaccine Hesitancy in Black Communities, and How Can It Be Addressed? : RAND Corporation. [Google Scholar]
  4. Bogart LM, Landrine H, Galvan FH, Wagner GJ, & Klein DJ (2013). Perceived discrimination and physical health among HIV-positive Black and Latino men who have sex with men. AIDS Behav,, 17(4), 1431–1441. Retrieved from 10.1007/s10461-012-0397-5.pdf [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bogart LM, Mutchler MG, McDavitt B, Klein DJ, Cunningham WE, Goggin KJ, … Wagner GJ (2017). A Randomized Controlled Trial of Rise, a Community-Based Culturally Congruent Adherence Intervention for Black Americans Living with HIV. Ann Behav Med, 51(6), 868–878. doi: 10.1007/s12160-017-9910-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bogart LM, Takada S, & Cunningham WE (2019). Medical Mistrust, Discrimination, and the Domestic HIV Epidemic. In Ojikutu BO & Stone VE (Eds.), HIV/AIDS in US Communities of Color. New York, NY: Springer. [Google Scholar]
  7. Bogart LM, Takada S, & Cunningham WE (2021). Medical Mistrust, Discrimination, and the Domestic HIV Epidemic. In Ojikutu BO & Stone VE (Eds.), HIV in US Communities of Color (pp. 207–231). Cham: Springer International Publishing. [Google Scholar]
  8. Bogart LM, & Thorburn S (2005). Are HIV/AIDS conspiracy beliefs a barrier to HIV prevention among African Americans? JAIDS Journal of Acquired Immune Deficiency Syndromes, 38(2), 213–218. [DOI] [PubMed] [Google Scholar]
  9. Bogart LM, Wagner G, Galvan FH, & Banks D (2010). Conspiracy beliefs about HIV are related to antiretroviral treatment nonadherence among African American men with HIV. Journal of acquired immune deficiency syndromes (1999), 53(5), 648. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Bowleg L (2012). The problem with the phrase women and minorities: intersectionality-an important theoretical framework for public health. Am J Public Health, 102(7), 1267–1273. doi: 10.2105/AJPH.2012.300750 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Braksmajer A, Simmons J, Aidala A, & McMahon JM (2018). Effects of discrimination on HIV-related symptoms in heterosexual men of color. American journal of men's health, 12(6), 1855–1863. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Centers for Disease Control and Prevention. (2020). HIV Surveillance Report, 2018 (Updated). Retrieved from https://www.cdc.gov/hiv/library/reports/hiv-surveillance.html [Google Scholar]
  13. Cuffee YL, Hargraves JL, Rosal M, Briesacher BA, Schoenthaler A, Person S, … Allison J (2013). Reported Racial Discrimination, Trust in Physicians, and Medication Adherence Among Inner-City African Americans With Hypertension. American Journal of Public Health, 103(11), E55–E62. doi: 10.2105/Ajph.2013.301554 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Dale SK, Bogart LM, Wagner GJ, Galvan FH, & Klein DJ (2016). Medical mistrust is related to lower longitudinal medication adherence among African-American males with HIV. J Health Psychol, 21(7), 1311–1321. doi: 10.1177/1359105314551950 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Division of HIV and STD Programs Department of Public Health County of Los Angeles. (2020). HIV Surveillance Annual Report, 2019. Retrieved from http://publichealth.lacounty.gov/dhsp/Reports.htm [Google Scholar]
  16. Durant RW, Legedza AT, Marcantonio ER, Freeman MB, & Landon BE (2011). Different types of distrust in clinical research among whites and African Americans. Journal of the National Medical Association, 103(2), 123–130. doi:Doi 10.1016/S0027-9684(15)30261-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Earnshaw VA, Bogart LM, Dovidio JF, & Williams DR (2015). Stigma and racial/ethnic HIV disparities: moving toward resilience. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Eaton LA, Driffin DD, Kegler C, Smith H, Conway-Washington C, White D, & Cherry C (2015). The role of stigma and medical mistrust in the routine health care engagement of black men who have sex with men. Am J Public Health, 105(2), e75–82. doi: 10.2105/ajph.2014.302322 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Galvan FH, Bogart LM, Klein DJ, Wagner GJ, & Chen YT (2017). Medical mistrust as a key mediator in the association between perceived discrimination and adherence to antiretroviral therapy among HIV-positive Latino men. J Behav Med, 40(5), 784–793. doi: 10.1007/s10865-017-9843-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Gwadz M, Applegate E, Cleland C, Leonard NR, Wolfe H, Salomon N, … Sanfillipo L (2014). HIV-infected individuals who delay, decline, or discontinue antiretroviral therapy: comparing clinic-and peer-recruited cohorts. Frontiers in public health, 2, 81. Retrieved from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4100062/pdf/fpubh-02-00081.pdf [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Hall MA, Dugan E, Zheng B, & Mishra AK (2001). Trust in physicians and medical institutions: what is it, can it be measured, and does it matter? Milbank Q, 79(4), 613–639, v. doi: 10.1111/1468-0009.00223 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Hammond WP (2010). Psychosocial correlates of medical mistrust among African American men. American Journal of Community Psychology, 45(1-2), 87–106. Retrieved from 10.1007/s10464-009-9280-6?download=true [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Hausmann LR, Kwoh CK, Hannon MJ, & Ibrahim SA (2013). Perceived racial discrimination in health care and race differences in physician trust. Race and Social Problems, 5(2), 113–120. [Google Scholar]
  24. Hayes AF (2017). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach: Guilford publications. [Google Scholar]
  25. Institute of Medicine Committee on, U., Eliminating, R., & Ethnic Disparities in Health, C. (2003). In Smedley BD, Stith AY, & Nelson AR (Eds.), Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care. Washington (DC): National Academies Press (US) Copyright 2002 by the National Academy of Sciences. All rights reserved. [PubMed] [Google Scholar]
  26. Institute of Medicine Committee on Understanding and Eliminating Racial and Ethnic Disparities in Health Care. (2003). Unequal treatment: confronting racial and ethnic disparities in health care. Washington (DC): National Academies Press (US). [PubMed] [Google Scholar]
  27. Kalichman SC, Eaton L, Kalichman MO, & Cherry C (2017). Medication beliefs mediate the association between medical mistrust and antiretroviral adherence among African Americans living with HIV/AIDS. J Health Psychol, 22(3), 269–279. doi: 10.1177/1359105315600239 [DOI] [PubMed] [Google Scholar]
  28. Kalichman SC, Eaton L, Kalichman MO, Grebler T, Merely C, & Welles B (2016). Race-based medical mistrust, medication beliefs and HIV treatment adherence: test of a mediation model in people living with HIV/AIDS. J Behav Med, 39(6), 1056–1064. doi: 10.1007/s10865-016-9767-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kenny DA, & Judd CM (2014). Power anomalies in testing mediation. Psychological Science, 25(2), 334–339. [DOI] [PubMed] [Google Scholar]
  30. LaVeist TA, Isaac LA, & Williams KP (2009). Mistrust of health care organizations is associated with underutilization of health services. Health services research, 44(6), 2093–2105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Lyons SJ, Dailey AF, Yu C, & Johnson AS (2021). Care Outcomes Among Black or African American Persons with Diagnosed HIV in Rural, Urban, and Metropolitan Statistical Areas - 42 U.S. Jurisdictions, 2018. MMWR Morb Mortal Wkly Rep, 70(7), 229–235. doi: 10.15585/mmwr.mm7007a1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Muthén LK, & Muthén BO (2017). Mplus User’s Guide. 8th Ed.. Retrieved from Los Angeles, CA: [Google Scholar]
  33. Office of Disease Prevention and Health Promotion. (2020). Social Determinants of Health ∣ Healthy People 2020. Retrieved from https://www.healthypeople.gov/2020/topics-objectives/topic/social-determinants-of-health [Google Scholar]
  34. Omodei MM, & McLennan J (2000). Conceptualizing and Measuring Global Interpersonal Mistrust-Trust. The Journal of Social Psychology, 140(3), 279–294. doi: 10.1080/00224540009600471 [DOI] [PubMed] [Google Scholar]
  35. Ren XS, Amick BC, & Williams DR (1999). Racial/ethnic disparities in health: the interplay between discrimination and socioeconomic status. Ethnicity & disease, 9(2), 151–165. [PubMed] [Google Scholar]
  36. Rintamaki LS, Davis TC, Skripkauskas S, Bennett CL, & Wolf MS (2006). Social stigma concerns and HIV medication adherence. AIDS Patient Care & STDs, 20(5), 359–368. [DOI] [PubMed] [Google Scholar]
  37. Rolfe A, Cash-Gibson L, Car J, Sheikh A, & McKinstry B (2014). Interventions for improving patients' trust in doctors and groups of doctors. Cochrane Database Syst Rev,(3), CD004134. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Rumptz MH, Tobias C, Rajabiun S, Bradford J, Cabral H, Young R, & Cunningham WE (2007). Factors associated with engaging socially marginalized HIV-positive persons in primary care. AIDS Patient Care and STDs, 21(S1), S-30–S-39. [DOI] [PubMed] [Google Scholar]
  39. Sey EK, & Ma Y (2017). National HIV Behavioral Surveillance among Men Who Have Sex with Men(MSM) NHBS-MSM5, 2017. Retrieved from http://publichealth.lacounty.gov/dhsp/Reports/HIV/NHBS_MSM5_Report_7-22-19.pdf [Google Scholar]
  40. Thom DH, Ribisl KM, Stewart AL, Luke DA, & Physicians TSTS (1999). Further validation and reliability testing of the Trust in Physician Scale. Medical care, 510–517. [DOI] [PubMed] [Google Scholar]
  41. Thom DH, Tirado MD, Woon TL, & McBride MR (2006). Development and evaluation of a cultural competency training curriculum. BMC Med Educ,, 6, 38. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Thomas KS, Bardwell WA, Ancoli-Israel S, & Dimsdale JE (2006). The toll of ethnic discrimination on sleep architecture and fatigue. Health Psychology, 25(5), 635. [DOI] [PubMed] [Google Scholar]
  43. Thrasher AD, Earp JAL, Golin CE, & Zimmer CR (2008). Discrimination, distrust, and racial/ethnic disparities in antiretroviral therapy adherence among a national sample of HIV-infected patients. JAIDS Journal of Acquired Immune Deficiency Syndromes, 49(1), 84–93. [DOI] [PubMed] [Google Scholar]
  44. Tulsky JA, Arnold RM, Alexander SC, Olsen MK, Jeffreys AS, Rodriguez KL, … Pollak KI (2011). Enhancing communication between oncologists and patients with a computer-based training program: a randomized trial. Ann Intern Med,, 155(9), 593–601. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Turan JM, Elafros MA, Logie CH, Banik S, Turan B, Crockett KB, … Murray SM (2019). Challenges and opportunities in examining and addressing intersectional stigma and health. BMC Med, 17(1), 7. doi: 10.1186/s12916-018-1246-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Wagner GJ, Bogart LM, Mutchler MG, McDavitt B, Mutepfa KD, & Risley B (2016). Increasing antiretroviral adherence for HIV-positive African Americans (project Rise): A treatment education intervention protocol. JMIR Res Protoc,, 5(1), e45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Walters GD (2018). Why are mediation effects so small? International Journal of Social Research Methodology, 22(2), 219–232. doi: 10.1080/13645579.2018.1517232 [DOI] [Google Scholar]
  48. Williams DR, Lawrence JA, & Davis BA (2019). Racism and health: evidence and needed research. Annual review of public health, 40, 105–125. Retrieved from 10.1146/annurev-publhealth-040218-043750 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Williams DR, Yu Y, Jackson JS, & Anderson NB (1997). Racial differences in physical and mental health: Socio-economic status, stress and discrimination. Journal of health psychology, 2(3), 335–351. Retrieved from https://deepblue.lib.umich.edu/bitstream/handle/2027.42/67159/10.1177_135910539700200305.pdf?sequence=2 [DOI] [PubMed] [Google Scholar]
  50. Williamson LD, & Bigman CA (2018). A systematic review of medical mistrust measures. Patient Educ Couns, 101(10), 1786–1794. doi: 10.1016/j.pec.2018.05.007 [DOI] [PubMed] [Google Scholar]
  51. Williamson LD, Smith MA, & Bigman CA (2019). Does discrimination breed mistrust? Examining the role of mediated and non-mediated discrimination experiences in medical mistrust. J Health Commun,, 24(10), 791–799. Retrieved from 10.1080/10810730.2019.1669742?needAccess=true [DOI] [PubMed] [Google Scholar]

RESOURCES