Viral Suppression Is Associated with HIV Treatment Self-Efficacy in a Cohort of Women in Washington, DC

Abstract

The goal of HIV treatment is viral suppression as it is linked with improved health outcomes and decreased risk of viral transmission. We assessed the sociodemographic, behavioral, and patient–provider interaction associations with viral suppression with an administered survey to HIV-seropositive women in the metropolitan Washington, DC, site of the Women's Interagency HIV Study (WIHS) between 2017 and 2018. Logistic and mixed models were used to explore related factors between HIV viral suppression groups and HIV treatment self-efficacy, respectively. Higher HIV treatment self-efficacy and disclosure concerns were positively associated with viral suppression, while illicit drug use had a negative association. In mixed models, more health care provider trust was associated with higher HIV treatment self-efficacy, while depressive symptoms were associated with lower HIV treatment self-efficacy. Depression, illicit substance use, and HIV treatment self-efficacy are potentially modifiable factors that can influence viral suppression. Implementation studies are needed to determine whether interventions to manage depression or self-efficacy and improve trust in health care providers will influence treatment outcomes.

Introduction

The goal of treatment with combination antiretroviral therapy (cART) in persons living with HIV (PLWH) is viral suppression, which is associated with improved health outcomes and decreased viral transmission risk.¹ Our group examined longitudinal viral suppression patterns using group-based trajectory patterns among women in the Women's Interagency HIV Study (WIHS), a multisite cohort of women living with HIV (WLWH) and women at high risk for HIV acquisition. We identified three trajectory patterns in the cohort-wide analysis: high, intermediate, and low probability of viral suppression. The probability of viremia decreased over the two decades of the study, but a subset of women in the high probability group maintained viremic status.² Women enrolled in the District of Columbia (DC) site demonstrated similar trajectory patterns. In this group, less than one-third of the women achieved and sustained viral suppression.³

Understanding factors associated with viral suppression is important to help develop targeted interventions tailored to WLWH, a group that makes up 23% of PLWH in the United States.⁴ This knowledge may help craft actionable plans to achieve the US Department of Health and Human Services goals to end the HIV epidemic.⁵ It is with this background that we aimed to identify factors that contributed to HIV treatment success among WLWH in DC.

To contextualize health seeking behaviors, viral suppression, and HIV treatment successes, we utilized the framework of social cognitive theory as it has been applied to health promotion and disease prevention.⁶ The framework of health promotion by social cognitive theory promotes four core determinants of health promotion, including knowledge, perceived self-efficacy, outcome expectations, and perceived facilitators/impediments.⁶

From prior studies, we know that treatment self-efficacy influences the behavior of patients and correlates with treatment outcomes in HIV and other disease states.⁷ However, the factors influencing HIV treatment self-efficacy among PLWH are not well described in the literature, and there is mixed evidence on the association of knowledge/health literacy and outcomes among PLWH.⁸ Gender, race/ethnicity, socioeconomic status, and/or other demographics are known important factors in the ability to remain engaged in care and achieve viral suppression.^9,10 The experience of stigma and patient–provider relationship, which may be perceived facilitators/impediments to success, also affects treatment adherence, engagement in care, and likely influence rates of viral suppression.^7,11,12

Historically these factors have been associated with medication adherence and engagement in care cross sectionally, and have been associated with longitudinal self-reported medication adherence and explored within populations enrolled in longitudinal cohort studies.^13,14 Our study assessed the interaction of stigma, treatment self-efficacy, patient–provider relationship, and patient-related factors with longitudinal virologic outcomes, a direct measure of treatment adherence among WLWH.

Methods

Participants

Study participants included WLWH from the metropolitan Washington, DC site of the WIHS, a prospective observational cohort study of WLWH. The study methodology has been previously described; women were enrolled in four waves in 1994–1995, 2000–2001, 2011–2012, and 2013–2015 with subsequent enrollment periods to recruit for losses and ensure the cohort reflected the population of WLWH in the United States. All enrolled participants complete semiannual study visits.^15–17

Data collection and survey design

At study visits, participants completed structured interviews, and data were collected on sociodemographic status, health care utilization, self-reported medical comorbidities, illicit substance use as well as tobacco and alcohol use. Testing included an assessment of depressive symptoms using the Center for Epidemiological Studies Depression (CES-D) Scale.¹⁸ Symptoms of post-traumatic stress disorder (PTSD) were measured using the PTSD Checklist (PCL).¹⁹ In addition, study participants underwent laboratory testing to measure HIV RNA (viral load) at each semiannual visit. Sociodemographic data, assessment of illicit drug, and/or alcohol use were obtained from the WIHS survey.

An additional questionnaire was developed and administered from 2017 to 2018 to all consenting women at the DC WIHS site to further assess potential patient-related (self-efficacy, perceived HIV-related stigma, social support, trust and comfort with health care providers, preference regarding health care decision-making, and HIV knowledge) and health care provider factors (demographics) associated with viral suppression. All study procedures were approved by the Institutional Review Board.

Outcomes

The primary outcome was HIV viral load and participants were grouped into two categories, virally suppressed and not virally suppressed. Participants were considered suppressed if the viral load was ≤200 copies/mL during each semiannual study visit between 2015 and 2018, for which there was viral load data. Participants with viral load >200 at any time point tested were considered not suppressed. The second outcome was HIV treatment self-efficacy score measured using the HIV Treatment Adherence Self-Efficacy Scale (HIV-ASES).²⁰

Variables

Data from the WIHS core study included race, age, housing stability, illicit drug use (marijuana or hash, crack, cocaine, heroin, illicit methadone, methamphetamines, amphetamines, narcotics, hallucinogens, and other injected or noninjected drugs), alcohol intake (dichotomized as 0–7 or >7 drinks per week), presence of depressive symptoms (CES-D) ≥16, and smoking (never, current, former smoker). Covariates from the additional administered survey included the following: health care provider trust as measured by the Primary Care Assessment Survey (PCAS), number of years seeing current provider, concordance between ideal and reported relationship with health provider, HIV treatment self-efficacy as measured by the HIV-ASES, stigma as measured by the HIV Stigma Scale, and social support as measured by the Inventory of Socially Supportive Behavior (ISSB).^20–26

The HIV-ASES is a 12-item questionnaire designed to assess patient self-efficacy for adherence to medical treatment plans related to HIV. In the questionnaire, participants are asked to rank their confidence using a 10-point scale in their ability to perform various treatment-related tasks or behaviors. The scale ranges from cannot do at all to completely certain can do. In prior studies of PLWH, this scale demonstrated both reliability (ρs > 0.90) and validity with internal consistency and a 3-month test-retest reliability of rs >0.70. Factor analyses supported subscales using adherence integration and adherence perseverance, which consisted of 9 and 3 items, respectively.²⁰

The HIV Stigma Scale is a 40-item instrument used to measure perceived stigma by PLWH. Participants were asked to describe their agreement with each statement in the questionnaire using a 4-point scale. Responses ranged from strong disagreement to strong agreement. The HIV Stigma Scale measured four factors: personalized stigma (loading of 18 items), disclosure concerns (loading of 12 items), negative self-imaging (loading of 9 items), and concern with public attitudes about PLWH (loading of 12 items). Prior studies showed evidence of internal consistency reliability of the total scale and subscales with an overall α = 0.96.²⁴

ISSB is a 40-item self-report measure used to assess how often individuals received various forms of assistance during the preceding months. Participants are asked to rate the frequency of each item on a 5-point scale from 1 (not at all) to 5 (every day). The 5-point ratings of each item were summed to form a total score to assess social support. In previous studies, the internal consistency reliability has been consistently >0.9.^25,27,28

The PCAS is a questionnaire originally designed to inform physician and physician practice performance monitoring and quality improvement. It consists of 51 questions and measures 11 areas of performance. We utilized the 8 items related to trust. Prior studies note the internal consistency reliability of the trust section as α = 0.86.²⁶

Statistical analysis

Participants who completed a full study visit between 10/01/2017 and 09/30/2018 and the administered survey were included in this analysis. Participants with fewer than two viral load measurements between 2015 and 2018 were excluded. The baseline visit was defined as the visit at the time of the questionnaire. One-hundred and sixty-five participants were included in the final analysis. Descriptive statistics were used to describe demographic characteristics of study participants and chi-square or Fisher's exact tests used to test for significance for categorical variables and one-way analysis of variance was used for continuous variables (Table 1). Cronbach's alpha was calculated to check the internal reliability for the HIV-ASES and HIV Stigma Subscales.

Table 1.

Baseline Demographic Characteristics of Study Participants by Viral Suppression Group

n (%)	Overall (n = 165)	Viral suppression group		p-Value
		Suppressed n (%) 119 (72.1%)	Not suppressed n (%) 46 (27.9%)
Age
Mean (SD)	52.9 (7.4)	53.5 (7.1)	51.6 (8.1)	0.1513
Race
African American	125 (75.8)	89 (74.8)	36 (78.3)
White	25 (15.2)	18 (15.1)	7 (15.2)	0.9454
Hispanic	8 (4.9)	6 (5.0)	2 (4.4)
Others	7 (4.2)	6 (5.0)	1 (2.2)
HIV therapy
No therapy	13 (7.9)	3 (2.5)	10 (21.7)	<0.0001
cART	152 (92.1)	116 (97.5)	36 (78.3)
CD4 T lymphocyte count 100 cells/μL
Mean (SD)	712.1 (351.8)	762.6 (318.8)	575.4 (401.4)	0.0023
HIV viral load copies/mL
Median (IQR)	20 (20–26.5)	20 (20–20)	144 (20–1130)	<0.0001
Years in school, mean (SD)	11.3 (2.0)	11.1 (2.2)	11.6 (1.1)	0.1552
Depressive symptoms (CES-D ≥ 16)a	43 (26.1)	31 (26.1)	12 (26.1)	0.9962
Report abuse
Sexual	74 (44.8)	56 (47)	18 (39.1)	0.6937
Physical	95 (57.6)	69 (58)	26 (56.5)	0.6402
Domestic violence	66 (40.0)	53 (44.5)	13 (28.3)	0.0981
Mean post-traumatic check list score (SD)	27.3 (18.8)	27.3 (18.8)	29.5 (14.4)	0.2793
Reported drug use	27 (16.4)	15 (12.6)	12 (26.1)	0.0358
Alcohol use
0–7 drinks/week	159 (96.4)	115 (96.6)	44 (95.7)	0.6714
>7 drinks/week	6 (3.6)	4 (3.4)	2 (4.4)
Smoking status
Never smoker	58 (35.2)	42 (35.3)	16 (34.8)
Current smoker	42 (25.5)	29 (24.4)	13 (28.3)	0.8624
Former smoker	65 (39.4)	48 (40.3)	17 (36.9)
Stable housing	148 (89.7)	107 (89.9)	41 (89.1)	0.8817
Currently insured	164 (99.4)	118 (99.2)	46 (100)	0.9932

Baseline demographic data collected at time of survey completion.

^aDepressive symptoms were assessed using the CES-D Scale.

cART, combination antiretroviral therapy; CES-D, Center for Epidemiological Studies Depression; IQR, interquartile range; SD, standard deviation.

A logistic model was used to describe the relationship between the dichotomous outcome variable HIV viral suppression groups and behavioral and patient–provider-related covariates. The multivariate analysis included variables with p-value <0.1 in the univariate analysis. Two factors of HIV treatment self-efficacy were run in the multivariate model separately to avoid collinearity. We conducted an additional analysis to determine factors related to the total score of HIV treatment self-efficacy using mixed modeling. Variables with p-value <0.1 in the univariate were included in the multivariate model. SAS version 9.3 was used for all statistical analyses.

Internal consistency reliability

The two subscales of HIV-ASES and four subscales of HIV Stigma demonstrated excellent internal consistency reliability. For HIV-ASES, the overall standardized Cronbach's α = 0.95, the total scale α = 0.86, adherence integration α = 0.93, and adherence perseverance α = 0.99. For HIV stigma, the overall Standardized Cronbach's α = 0.88, total scale α = 0.79, personalized stigma had Cronbach's α = 0.86, disclosure concerns α = 0.86, self-imaging Cronbach's α = 0.90, and the concern with public attitudes about PLWH α = 0.85.

Results

Participant demographics and other descriptive characteristics

Demographic characteristics of the study cohort are described in Table 1. A total of 165 of 179 HIV seropositive women completed the survey and 72% of the participants were in the suppressed group. The majority of survey participants were African American at 76% (n = 125) with a mean age of 53 years [standard deviation (SD) 7.4]. At the baseline visit, the majority of participants, 92% (n = 152), were on cART. Participants who were virologically suppressed had lower illicit drug use at 13% (n = 15) versus 26.1% (n = 12), p = 0.0358, among those who were not suppressed. Twenty-six percent of participants (n = 43) had depressive symptoms above the recommended clinical cutoff score of 16. Fifty-eight percent (n = 95) of participants reported a lifetime history of physical abuse, 45% (n = 74) sexual abuse, and 40% (n = 66) domestic violence. Sixteen percent (n = 27) reported current illicit drug use, and 4% (n = 6) reported consuming >7 drinks per week. Ninety percent (n = 148) of participants had stable housing and 99% (n = 164) had health care insurance.

Across both groups, participants reported seeing their current HIV care provider for a median of 9 years [interquartile range (IQR) 3–16] and 91% (n = 150) reported that they would continue seeing their current care provider. Participants had a median of 3 (IQR 2–5) HIV health care providers over the course of their HIV diagnosis, with no significant difference between women with viral suppression versus those not suppressed, p = 0.2644. Seventy-six percent (n = 126) of participants reported that their relationship with their provider around health care decision-making was well matched with their ideal and no differences were noted by viral suppression group, virologically suppressed 77% (n = 92) versus not virologically suppressed 74% (n = 34), p = 0.7426.

Total stigma scores were 91 [standard deviation (SD) 19] and 86 (SD 20) in the virologically suppressed group and the not suppressed group, respectively, p = 0.1615. Mean health literacy scores (range, 4 = never need help to 20 = always need help) were similar between the virologically suppressed groups and not suppressed groups with a mean score of 6.4 (SD 2.9) versus 6.5 (SD 3.1), respectively, p = 0.7377. Reported levels of social support were similar between groups, 0 = 0.2644.

Factors associated with viral suppression

Factors associated with viral suppression are outlined in Table 2. In univariate analyses illicit drug use was negatively associated with viral suppression odds ratio (OR) 0.41 [95% confidence interval (CI) 0.17–0.96, p = 0.0395]. Disclosure concerns were associated with increased odds of viral suppression OR 1.07 (95% CI 0.99–1.15, p = 0.09). Higher HIV treatment self-efficacy, both integration OR 1.02 (95% CI 1.004–1.05, p = 0.0188) and perseverance OR 1.06 (95% CI 1.001–1.12, p = 0.045), were associated with increased odds of viral suppression.

Table 2.

Factors Associated with Viral Suppression Among Survey Participants

Factor	Univariate			Multivariate
	OR	95% CI	p-Value	OR	95% CI	95% CI
Race
Others	2.33	0.24–23.04	0.4723
Hispanic	1.17	0.19–7.22	0.8990
African American	0.96	0.37–2.50	0.4683
White (reference)	1	1
Age (years)	1.04	0.99–1.09	0.1521
Stable housing
Yes	1.09	0.36–3.28	0.8817
No (reference)	1
Years in school	0.85	0.67–1.07	0.1654
Depressive symptoms
Yes (CES-D ≥ 16)a	0.99	0.46–2.17	0.9962
No (CES-D <16) (reference)	1
Reported drug use
Yes	0.41	0.17–0.96	0.0395	0.34	0.13–0.92	0.0343
No (reference)	1			1
Alcohol use:
0–7 drinks/week (reference)	1	1
>7 drinks/week	1.31	0.23–7.39	0.7621
Smoking:
Never smoker (reference)	1	1
Current smoker	0.85	0.36–2.03	0.6105
Former smoker	1.08	0.48–2.39	0.6676
Health care provider trustb	0.95	0.77–1.19	0.6694
Years with current health care
Provider	1.01	0.97–1.05	0.7040
Self-efficacy: integrationc	1.02	1.004–1.05	0.0188	1.04	1.01–1.07	0.00328
Self-efficacy: perseverance	1.06	1.001–1.12	0.045	1.09	1.02–1.16	0.0117
Stigma
Personalized stigma	1.02	0.98–1.06	0.2667
Disclosure concerns	1.07	0.99–1.15	0.09	1.1	1.01–1.20	0.0277
Negative self-image	1.07	0.89–1.30	0.4793
Concern with public attitudes about people with HIV	1.03	0.97–1.09	0.3748
Social support-ISSB total scored	0.995	0.99–1.004	0.2634

^aDepressive symptoms were assessed using the CES-D Scale.

^bHealth care provider trust was measured on a scare of 1–10 (1 = not at all; 10 = completely).

^cTwo factors of self-efficacy were run in the multivariate model separately to avoid collinearity.

^dThe ISSB questionnaire was used to assess social support.

CES-D, Center for Epidemiological Studies Depression; CI, confidence interval; ISSB, Inventory of Socially Supportive Behavior; OR, odds ratio.

In multivariate analysis, illicit drug use OR 0.34 (95% CI 0.13–0.92, p = 0.0343) was negatively associated with being in the virologically suppressed group. Both subscales of HIV treatment self-efficacy, integration OR 1.04 (95% CI 1.01–1.07, p = 0.00328), and perseverance OR 1.09 (95% CI 1.02–1.16, p = 0.0117) were associated with increased odds of viral suppression as were disclosure concerns OR 1.10 (95% CI 1.01–1.20, p = 0.0277).

HIV treatment self-efficacy

Factors associated with HIV treatment self-efficacy scores are described in Table 3. Women with depressive symptoms as measured by the CES-D had a 14 point difference in total HIV treatment self-efficacy score (−14, 95% CI −21.4 to −7.1, p < 0.001). Greater trust in health care providers was associated with a higher total HIV treatment self-efficacy score (3.3, 95% CI 1.4–5.2, p = 0.0008). In multivariate analyses, health care provider trust (2.8, 95% CI 0.9–4.7, p = 0.0039) and symptoms of depression (−12.1, 95% CI −19.3 to −4.9, p = 0.001) were associated with increased and decreased HIV treatment self-efficacy, respectively. The Baron and Kenny approach was used to check mediation effects of HIV treatment self-efficacy, trust, and viral suppression, and HIV treatment self-efficacy was not a mediator between trust and viral suppression (p = 0.6694, |c| = 0.0479, |c′| = 0.1505).

Table 3.

Factors Associated with HIV Treatment Self-Efficacy Score Among Survey Participants

Factor	Univariate			Multivariate
	Estimatea	95% CI	p-Value	Estimate	95% CI	p-Value
Race
Others	−9.9	−27.8 to 8	0.2784
Hispanic	−0.8	−18.7 to 17.1	0.9335
African American	−4.1	−13.4 to 5.2	0.3859
White (reference)	0
Age (years)	0.3	−0.1 to 0.8	0.1529
Stable housing
Yes	2.7	−7.7 to 13.2	0.6070
No (reference)	0
Years in school	0.3	−1.0 to 1.7	0.6356
Depressive symptoms
Yes (CES-D ≥ 16)b	−14.2	−21.4 to −7.1	<.0001	−12.1	−19.3 to −4.9	0.001
No (CES-D <16) (reference)	0
Reported drug use	−5.8	−14.6 to 2.9	0.1931
Alcohol use
0–7 drinks/week (reference)	0
>7 drinks/week	−5.3	−24.3 to 13.7	0.5860
Smoking
Never smoker (reference)	0
Current smoker	−1.9	−10.3 to 6.5	0.6628
Former smoker	−1.8	−9.4 to 5.8	0.6472
Health care provider trustc	3.3	1.4 to 5.2	0.0008	2.8	0.9 to 4.7	0.0039
Number of years seeing current Provider	0.3	−0.1 to 0.6	0.1938
Concordance between ideal and reported relationship with providers
Yes	6.6	−1.2 to 14.4	0.0951	4.9	−2.5 to 12.4	0.1927
No (reference)	0

^aEstimated change in total HIV treatment self-efficacy score.

^bDepressive symptoms were assessed using the CES-D Scale.

^cHealth care provider trust was measured on a scare of 1–10 (1 = not at all; 10 = completely).

CES-D, Center for Epidemiological Studies Depression; CI, confidence interval.

Discussion

In this study, we describe the factors related to viral suppression among a representative group of WLWH in the metropolitan Washington, DC area. Viral suppression groups did not have substantive differences demographically except for more drug use among women who were not virologically suppressed. Both groups had similar ages, insurance and housing status, health literacy and social support scores, and trauma/abuse history. However, higher HIV treatment self-efficacy and more HIV-serostatus disclosure concerns were associated with viral suppression. Illicit drug use was negatively associated with viral suppression. In the original framework of our study, we hypothesized based on social cognitive theory that HIV treatment self-efficacy influenced outcomes such as viral suppression. Thus, we also evaluated factors associated with HIV treatment self-efficacy to better understand the women with and without viral suppression and found that depressive symptoms were associated with decreased HIV treatment self-efficacy, while trust in a health care provider was associated with greater HIV treatment self-efficacy.

As previously noted, knowledge, perceived treatment self-efficacy, outcome expectations, and perceived facilitators/impediments influence health promotion in the framework of social cognitive theory.⁶ Human agency, the control of thoughts, motivations, and actions is exercised through self-efficacy, goal planning/forethought, anticipating outcomes. Self-efficacy can influence health directly and indirectly by its impact on other core determinants of health. Thus, self-efficacy is a key determinant of health promoting behaviors, motivations, and actions.²⁹ Health literacy did not differ by viral suppression status in our analysis, and we did not assess some of other determinants, including outcome expectations or a direct assessment of perceived facilitators/impediments. However, we suspect that interventions to improve treatment self-efficacy may also influence these other determinants of health and viral suppression outcomes.⁶

Interventions with cognitive behavioral therapy improved AIDS-related self-efficacy among WLWH, and education to improve knowledge regarding antiretrovirals among patients with limited literacy-improved treatment self-efficacy.³⁰ Furthermore, interventions to improve self-efficacy such as providing education, training, and encouragement have been used in other medical conditions.^31,32 In HIV-seronegative populations, positive psychotherapy has been used to improve depression and self-efficacy.³³ Thus, interventions including but not limited to psychotherapy and treatment of depression and/or substance use could similarly be used in WLWH to improve self-efficacy and treatment outcomes. This may be particularly applicable to our cohort as increased depressive symptoms were linked with lower HIV treatment self-efficacy.

Depression is a major comorbidity in communities of PLWH, and an estimated one-fourth of PLWH have a diagnosis of depression with rates of depression higher than among the general population.^34,35 Depression and illicit drug use are associated with decreased cART use and adherence among WLWH.^36–38 Increased mortality, frequent missed appointments, and nonviral suppression are also correlated with depression among PLWH.³⁹ Prior studies in the WIHS noted that adequate treatment of major depressive disorder and clinically significant depressive symptoms was only 46.2% and 37.9%, respectively in the cohort.⁴⁰ This suggests that WLWH are undertreated for depression and this could be a factor mediating nonviral suppression in certain subpopulations. Depression frequently occurs in the presence of substance abuse and patients with concurrent substance use disorders and depression often have more severe depressive symptoms.^41–43 In our analysis, substance use was associated with decreased odds of viral suppression and noted in 16% of our participants. Antidepressants and other interventions have been successful in treating depression among PLWH.^44–46 In addition, treatment of both substance use disorder and depression leads to improved health outcomes.^47–49 Strategies to treat depression or substance use could be applied and studied in PLWH who have comorbid depression and/or substance use disorders to improve HIV treatment self-efficacy and consequently treatment outcomes. While not assessed in this analysis, prior studies indicate a significant interaction between resilience, depressive symptoms, and viral suppression among WLWH and those with resiliency skills and low depressive symptoms were more likely to have viral suppression.⁵⁰ Black women make up the largest portion of WLWH in the United States, and among this group, resilience resources modified HIV stigma and depressive symptoms.⁵¹ Future studies may further explore strategies to improve resilience and depressive symptoms as this may influence virologic outcomes.

Our finding may also have implications for initiation of cART. Current guidelines recommend the initiation of cART immediately or as soon as possible after HIV diagnosis.^52,53 However, the guideline panel notes that in specific cases, initiation of therapy may be delayed due to a patient's willingness or unwillingness to start therapy and/or other clinical factors.⁵³ Our data support the potential utility of screening for and addressing conditions that are associated with treatment failure and concurrent interventions to address these conditions to improve treatment outcomes.

The identified relationship between provider trust and HIV treatment self-efficacy in our analysis is important as prior studies demonstrated that trust in health care providers was associated with improved health care treatment outcomes.^54,55 Caring and comfort, technical competency, and communication have been behaviors associated with trust in a health care provider.⁵⁶ However, evidence is mixed regarding interventions that can increase patient trust in physicians or other health care providers, and whether there are modifiable factors that can improve patient trust in a health care provider remains to be seen.⁵⁷ This is especially important among WLWH, a population that is disproportionately Black in the United States, who have historically had poor trust in the health care system and health care providers.^4,58–60 While patient trust may not be modifiable through direct interventions, understanding patients trust of providers, allows for optimizing support services to improve linkage to care.

Prior analyses found that perceived HIV-related stigma affected adherence to treatment.¹¹ In our cohort, personalized stigma, negative self-image, and concern regarding public attitudes about PLWH were not associated with viral suppression, but higher stigma scores related to HIV-serostatus disclosure concerns were positively associated with viral suppression. The association of HIV-serostatus disclosure concerns and increased odds of viral suppression has been described in other cohorts.⁶¹ In a qualitative study, participants reported attending medical visits so as not to appear ill and potentially disclose HIV-serostatus.⁶² In other studies, HIV disclosure status was negatively associated with medication adherence and women having difficulty taking medications openly at home had decreased probability of being on antiretroviral therapy.^63,64 This fear of HIV-serostatus disclosure through a visible decline in health may have contributed to our results. However, there is likely a complex relationship between HIV-serostatus disclosure concerns, related stigma, and treatment outcomes particularly among WLWH. Other studies suggest that there are high rates of disclosure to sex partners particularly among white men, Blacks were less likely to disclose their HIV serostatus.⁶⁵ As this cohort was predominantly Black and Hispanic women, this may have some effect; however, disclosure status or concerns with disclosure has not linked with viral suppression.

Our analysis was conducted in a DC-based cohort of WLWH who have elected to participate in a longitudinal research study. Thus, there is likely an element of selection bias among the study participants. Our cohort demographics do not cross the entire spectrum of PLWH in the United States and results may not be applicable to all groups living with the virus, including men, but our population is similar demographically in most respects to that of WLWH in the United States, and findings may provide some insight into this population.^66,67 Future studies may more fully evaluate other determinants of health promotion, including outcome expectations and perceived facilitators/impediments, as these factors can influence self-efficacy as well as incorporate additional exploration of self-efficacy related to other health disorders and agency.

In this cohort of women, higher HIV treatment self-efficacy scores and disclosure concerns were associated with viral suppression, while illicit drug use was associated with lower odds of viral suppression. Higher treatment self-efficacy scores in particular had an association with viral suppression, and HIV treatment self-efficacy was associated with greater trust in a health care provider and less depressive symptoms. While trust in health care providers may not be a modifiable factor as interventions have had mixed results, interventions are feasible to treat depression and substance use disorders and improve treatment self-efficacy. If implemented interventions to treat depression, substance disorders and/or improve treatment self-efficacy may improve viral suppression among certain populations of PLWH.

Author Contributions

A.B.S. drafted the article, contributed to survey design, and contributed to data analysis. S.K. conceptualized the study and survey design, contributed to data analysis, and critical review of the article and final approval of the version to be published. K.M. and L.G. contributed to survey design, contributed to data analysis, and critical review of the article. M.A.D., D.M, and L.A.-C. provided critical review of the article. C.W. managed the data, conducted the statistical analysis, and provided critical review of the article. K.L., K.B., A.K, A.D, and M.C. contributed to survey design.

Author Disclosure Statement

The authors have no conflicts of interests to disclose that are relevant to the content of this article.

Funding Information

This work was supported by the National Institute of Minority Health and Health Disparities at the National Institutes of Health under Grant 5U01A134994-24 to S.G.K.

This work was supported by the National Center for Advancing Transitional Sciences of the National Institutes of Health under award KL2TR001432 to K.G.M.

Data in the article were collected by the Women's Interagency HIV Study (WIHS), now the MACS/WIHS Combined Cohort Study (MWCCS). The contents of this publication are solely the responsibility of the authors and do not represent the official views of the National Institutes of Health (NIH). MWCCS (Principal Investigators): Metropolitan Washington CRS (Seble Kassaye and Daniel Merenstein), U01-HL146205. The MWCCS is funded primarily by the National Heart, Lung, and Blood Institute (NHLBI), with additional cofunding from the Eunice Kennedy Shriver National Institute Of Child Health & Human Development (NICHD), National Institute On Aging (NIA), National Institute Of Dental & Craniofacial Research (NIDCR), National Institute Of Allergy And Infectious Diseases (NIAID), National Institute Of Neurological Disorders And Stroke (NINDS), National Institute Of Mental Health (NIMH), National Institute On Drug Abuse (NIDA), National Institute Of Nursing Research (NINR), National Cancer Institute (NCI), National Institute on Alcohol Abuse and Alcoholism (NIAAA), National Institute on Deafness and Other Communication Disorders (NIDCD), National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), National Institute on Minority Health and Health Disparities (NIMHD), and in coordination and alignment with the research priorities of the National Institutes of Health, Office of AIDS Research (OAR). MWCCS data collection is also supported by UL1-TR000004 (UCSF CTSA), UL1-TR003098 (JHU ICTR), UL1-TR001881 (UCLA-CTSI), P30-AI-050409 (Atlanta CFAR), P30-AI-050410 (UNC CFAR), and P30-AI-027767 (UAB CFAR), and U01-HL146193 (DACC).