Associations Between HIV-Related Stigma,Racial Discrimination,Gender Discrimination,and Depression Among HIV-Positive African,Caribbean,and Black Women in Ontario,Canada

Abstract

African, Caribbean, and Black (ACB) women are greatly overrepresented in new HIV infections in comparison with Canada's general population. Social and structural factors such as HIV-related stigma, gender discrimination, and racial discrimination converge to increase vulnerability to HIV infection among ACB women by reducing access to HIV prevention services. Stigma and discrimination also present barriers to treatment, care, and support and may contribute to mental health problems. We administered a cross-sectional survey to HIV-positive ACB women (n=173) across Ontario in order to examine the relationships between HIV-related stigma, gender discrimination, racial discrimination, and depression. One-third of participants reported moderate/severe depression scores using the Beck Depression Inventory Fast-Screen guidelines. Hierarchical block regression, moderation, and mediation analyses were conducted to measure associations between independent (HIV-related stigma, gender discrimination, racial discrimination), moderator/mediator (social support, resilient coping), and dependent (depression) variables. Findings included: (1) HIV-related stigma was associated with increased depression; (2) resilient coping was associated with reduced depression but did not moderate the influence of HIV-related stigma on depression; and (3) the effects of HIV-related stigma on depression were partially mediated through resilient coping. HIV-related stigma, gender discrimination, and racial discrimination were significantly correlated with one another and with depression, highlighting the salience of examining multiple intersecting forms of stigma. Generalizability of findings may be limited due to nonrandom sampling. Findings emphasize the importance of multi-component interventions, including building resilient coping skills, mental health promotion and assessment, and stigma reduction programs.

Introduction

A frican, Caribbean, and Black (ACB) women are 7-fold overrepresented in new HIV infections in comparison with their white counterparts in Canada.¹ This overrepresentation of ACB women in Canada's HIV epidemic is in part due to social and structural factors such as HIV-related stigma, gender discrimination, and racial discrimination, which converge to increase vulnerability to HIV infection by reducing access to HIV prevention services and HIV testing.^2
–4 Stigma and discrimination also present barriers to treatment, care, and support for people living with HIV (PLHIV).^2,3,5,6

There is an urgent need to better understand the associations between interlocking forms of stigma and discrimination and their impact on mental health.^7
–9 A wide evidence base suggests that stress resulting from stigma and discrimination contributes to mental health disparities.^7,10
–12 Depression is particularly salient to examine among PLHIV. PLHIV are disproportionately affected by higher rates of depression—the highest contributor to disability adjusted life years among noncommunicable diseases^13,14—in comparison with non-PLHIV.² Depression has been associated with both HIV infection risks and with disease progression.^15,16 The array of stressors experienced by PLHIV, including stigma and discrimination, illness, disclosure concerns, side effects from antiretroviral treatment, and relationship issues contribute to depression risks.^2,17

HIV-related stigma refers to processes of devaluing, labeling, and stereotyping manifested in the loss of status, unfair and unjust treatment, and social isolation of people living, and associated with HIV.⁶ Racism and racial discrimination refer to inequitable and oppressive systems founded on ethno-racial differences, including beliefs, attitudes, exclusion, harassment, and institutional policies and practices.^8,18 Sexism and gender discrimination refer to oppressive and inequitable systems based on gender bias in attitudes, treatment, values, harassment, violence, and institutional policies and practices.^18,19 Higher rates of depression have been associated with HIV-related stigma,^2,12 racial discrimination,^9,11,20 and gender discrimination.^21,22 Less understood, however, are the relationships between these multiple forms of stigma and discrimination and how they may intersect to influence mental health outcomes.^5
–7,9

Social inequities based on race and gender may be exacerbated by HIV-related stigma.²³ Yet the associations between socio-demographic factors—such as race and gender—and HIV-related stigma are underexplored.²⁴ For example, a meta-analysis with PLHIV in North America found that HIV-related stigma was associated with poor mental health outcomes, yet only two of 24 studies examined associations between HIV-related stigma and race/ethnicity and three examined associations between HIV-related stigma and gender.¹² Wingood et al. (2007) found HIV-related stigma was associated with depression among African American women living with HIV (WLWH) but not among white WLHW, suggesting the adverse effects of HIV-related stigma may be exacerbated for women of color.²⁵ Studies with HIV-negative/status unknown African American women reported that when both racism and sexism were measured only sexism predicted distress;²² another study that measured both racial and gender discrimination indicated only gender discrimination predicted depression.²⁶ We found no studies with PLHIV that examined associations between racial discrimination, gender discrimination, HIV-related stigma, and depression concomitantly. This area therefore warrants further exploration.

Understanding factors that can ameliorate the impact of stigma on depression can inform interventions to promote mental health and reduce stigma. Coping has been described as both a moderator^21,27,28 and mediator²⁹ in the relationship between racism and mental health. Coping styles were found to mediate the influence of HIV-related stigma on depression and distress among PLHIV in the U.S.³⁰ Less research has examined coping and its relationship with sexism and mental health.^21,31

Social support may also play a role in reducing the health impacts of stigma. Results from cross-sectional studies with PLHIV in North America indicate an inverse relationship between HIV-related stigma and social support;^12,32 one study indicated social support mediated the impact of HIV-related stigma on emotional distress among PLHIV.³⁰ Social support has been reported to moderate³³ and mediate³⁴ the relationship between racism and mental health; social support and its association with sexism and mental health has been less studied.³⁵

We aimed to address three important gaps in the literature in this study. First, most studies have examined the constructs of racial discrimination, gender discrimination, and HIV-related stigma separately^7,12,20,22—precluding an understanding of the distinct experiences of ACB women who may experience these phenomena concurrently. Second, most HIV research with ACB women in Canada has been qualitative,^3
–5 so there is a lack of knowledge regarding levels and correlates of depression among this population. Third, the roles of social support and resilient coping as moderators or mediators of HIV-related stigma, sexism, and racism are not well understood.

The objective of this study was to contribute to understanding about the associations between the independent variables (racial discrimination, gender discrimination, HIV-related stigma), moderator/mediators (resilient coping, social support), and dependent variable (depression) among HIV-positive ACB women in Ontario, Canada.

Methods

Study design and population

We conducted a multi-method community-based research (CBR) study with Women's Health in Women's Hands (WLWH) across Ontario. Phase I was a qualitative study of research needs and priorities among WLWH; this phase involved 15 focus groups with diverse HIV-positive women living with HIV (n=104) in five cities across Ontario.⁵ Findings informed the conceptualization of an intersectional stigma conceptual model that highlighted the intersection of multiple forms of stigma—including racism, sexism, HIV-related stigma—and indicated social support and coping are integral strategies WLWH employ to address stigma.⁵ Intersectional approaches explore interdependent relationships between social identities and inequities that produce distinct experiences of oppression and opportunity.^36,37 This article reports results from Phase II that was informed by, and built on, Phase I findings.

The study hypotheses included: (1) higher levels of HIV-related stigma, racial discrimination, and gender discrimination would be associated with higher levels of depression; (2) higher levels of social support and resilient coping would be associated with lower levels of depression; (3) social support and resilient coping would interact with HIV-related stigma, racial discrimination, and gender discrimination to reduce the strength of the relationship between stigma/discrimination and depression (moderation); and (4) HIV-related stigma, racial discrimination, and gender discrimination would be associated with reduced social support and resilient coping, which would contribute to increased depression (mediation).

Data collection

A structured cross-sectional survey was implemented with HIV-positive ACB women in three regions of Ontario between June 2010 and January 2011. These regions were selected as they include AIDS service organizations (ASO) that provide supportive services to ACB women. Surveys were administered with peer research assistants (PRAs) that lived and worked in the regions where the surveys were conducted. We defined a peer as someone who shared a common lived experience with participants³⁸ and employed PRAs (n=9) who shared at least one lived experience with participants (e.g., female gender, African/Caribbean ethnicity, HIV-positive serostatus). The purpose of utilizing PRAs was to minimize power differentials between researchers and participants.³⁸ We conducted a 2-day training with PRAs on topics including: research methods, ethics, survey implementation, career development information, and provided ongoing support via bi-weekly teleconferences.

Recruitment was undertaken at several ASO, community health centers, hospitals, and other community-based organizations in each region that offer services to HIV-positive ACB women. Research Ethics Board approval was attained from Women's College Hospital at the University of Toronto. PRA recruited a purposive sample of HIV-positive ACB women. The PRA recruited participants via service provider networks and by word of mouth to promote study participation. The recruitment strategy aimed to include a total of 200 participants with representation from each region. The recommended sample size for linear multiple regression (fixed model, R² increase) with 12 correlates (moderation analyses: 3 independent variables, 2 moderator variables, 7 two-way interaction terms) as calculated using G*Power 3.1 (effect size 0.15, p<0.05, power: 0.95) is 184. Inclusion criteria for survey participants were adults aged 18 years and over, capable of providing informed consent, who self-identified as (1) a woman, (2) HIV-positive, and (3) Black, African, and/or Caribbean race/ethnicity.

A survey was developed to collect information on socio-demographic variables (e.g., age, income, education), HIV-related stigma, racial discrimination, gender discrimination, social support, resilient coping, and depression. Pilot testing and attaining feedback throughout the survey development process from PRA was conducted to enhance the survey's content validity and accessible survey language. PRA verbally administered surveys; surveys were approximately 90 min in duration and participants chose whether to complete the survey in a private room at a community agency or at their home. Participants provided informed consent before the survey was administered and no identifying information was collected. Participants received a $50 honorarium to compensate for time. Following survey completion the PRA offered participants information regarding support and health services in their region.

Measures

The ‘Stigma Scale Revised’ was used to measure HIV-related stigma.³⁹ This 10-item scale was adapted from Berger's stigma scale and includes 4 subscales: personalized stigma, disclosure, negative self-image, and public attitudes.³⁹ Overall Cronbach's α was 0.86; personalized stigma subscale α=0.89, disclosure α=0.76, negative self-image α=0.87, public attitudes α=0.76.

Racism was measured using the 9-item modified version of the ‘Everyday Discrimination Scale’.⁴⁰ Clark et al. (2004) modified original scale instructions from ‘In your day-to-day life how often have any of the following things happened to you?’ to ‘In your day-to-day life how often have any of the following things happened to you because of your race’.⁴⁰ The 9 items include: treated with less courtesy, treated with less respect, receive poorer service, people act as if you are not as smart, people act as if they are afraid of you, people act as if you are dishonest, people act as if they are better, called names, threatened or harassed. Cronbach's α was 0.91.

Gender discrimination was measured by modifying the Clark et al.'s Everyday Discrimination Scale⁴⁰ and replacing ‘In your day-to-day life how often have any of the following things happened to you because of your race with ‘In your day-to-day life how often have any of the following things happened to you because of your gender. That is, because you are a woman.’ Cronbach's α was 0.94.

The 19-item MOS Social Support survey was used to measure social support; subscales include emotional, informational, tangible, affectionate, and positive social interaction.⁴¹ Overall Cronbach's α was 0.95; emotional support subscale α=0.92, informational support α=0.79, tangible support α=0.86, affectionate support α=0.61, positive social interaction α=0.93.

Resilient coping, processes of positive adaptations to high stress, was measured using the Brief Resilient Coping Scale.⁴² This scale assesses dispositional (e.g., self-confidence, optimism) and situational (e.g., active problem solving) dimensions of coping.⁴² Cronbach's α was 0.87.

Depression was measured using the Beck Depression Inventory Fast-Screen (BDI-FS), a tool that provides a quick assessment of affective and cognitive components of depression.⁴³ Cronbach's α was 0.85.

Data analysis

We conducted descriptive analyses to calculate frequencies, means, and standard deviations (SD) among variables. Scale items were summed to calculate total scores for sexual/gender discrimination and depression; subscale and total scores were calculated for HIV-related stigma and social support. We also transformed scale scores to a 0–100 scale for comparison. Authors provided scoring guidelines for the Brief Resilient Coping Scale⁴² and the BDI-FS⁴³ and for these variables the items were summed and scored accordingly. Analyses were conducted using IBM SPSS 20. We conducted Pearson's bivariate correlational analyses to explore associations between variables.

Hierarchical (blockwise entry) linear regression was based on the intersectionality framework that posits HIV-related stigma, racial discrimination and gender discrimination are important correlates of depression, and social support and resilient coping are potential moderators.⁵ In the first regression analyses, subtypes of HIV-related stigma (personalized stigma, disclosure, negative self-image, public attitudes), racial discriminate, and gender discrimination were entered as correlates in block 1; in block 2, social support subtypes (emotional, tangible, affectionate, positive social interaction) and resilient coping were entered as correlates.

The purpose of the moderation analysis was to test if the moderators (social support, resilient coping) changed the strength or relationship between the independent and dependent variables. The independent and moderator variable total scores were mean centered and multiplied together to calculate the interaction terms. The sample size was insufficient to conduct moderation analyses using each subscale as a correlate. Therefore HIV-related subscales and social support subscales were combined to create two overall scores. In block 1, the HIV-related stigma overall score, racial discrimination, and gender discrimination were entered as correlates; in block 2, social support and resilient coping total scores were entered as correlates; in block 3, the two-way interaction terms were entered as correlates.

Mediation analysis was employed to assess if the independent variables (HIV-related, racial/gender discrimination) were associated with significant changes in the mediator variables (social support, resilient coping), which in turn would impact depression. We used Preacher and Hayes bootstrapping method, a nonparametric test that has increased power in comparison with the Sobel's test.⁴⁴ Preacher and Hayes provided SPSS macros to calculate total, direct and indirect effects for each mediator and independent variable, including significance tests using bootstrap procedures.⁴⁴ Bootstrapping is an appropriate method for this study as it may be used with small sample sizes and does not assume normal distribution of indirect effects, therefore reducing chance of type 2 error.⁴⁵ We report results for bootstrap tests, with a resample procedure of 1,000 bootstrap samples and 95% confidence interval (CI).

Results

Study population

Socio-demographic of participants (n=173) are described in Table 1. The mean participant age was 40.7 years (SD 8.8) and the median monthly income was $1400.00 (USD) (range 0–7916). Most participants (n=154; 89.0%) were born outside of Canada: 69.5% (n=107) in 25 different African countries, 29.9% (n=46) in 7 Caribbean countries, and 0.6% (n=1) in the United Kingdom.

Table 1.

Socio-Demographic Characteristics of Participants (n=173)

	Mean	Standard deviation
Age (years) (n=173)	40.7	8.8
Monthly income (US$)	3937.8	11589.3
Median monthly income: $1400.00 (Range 0–7916)
Highest level of education completed (n=154)
Less than high school	28.6%	44
High school	26.6%	41
College diploma	26.6%	41
University degree	16.2%	25
Relationship status (n=162)
Single	42.6%	69
Married	17.9%	29
Separated	14.2%	23
In a relationship/dating	9.9%	16
Widowed	8.0%	13
Common law/living with partner	4.9%	8
Divorced	2.5%	4
Children (n=156)
Currently caring for, or supporting, children	67.9%	106
HIV-positive serostatus of children being cared for	22.6%	24
Mean number of children being cared for: 2.3 (SD 1.4)
Employment status (n=172)
Part-time	16.9%	29
Full-time	16.3%	28
ODSP (disability)	51.2%	88
Social assistance (welfare)	11.1%	19
Canadian Pension Plan (retired)	4.7%	8

Percentages are calculated from non-missing responses.

Depression, HIV stigma, racial discrimination, gender discrimination, resilient coping, and social support

Overall and subscale score means across variables are presented in Table 2. Following the scoring guidelines authors provided for the depression and resilient coping scales,^42,43 over one-tenth (15.6%; n=27) of participants were severely depressed, 17.9% (n=31) moderately depressed, 30.6% (n=53) mildly depressed, and 35.8% (n=62) had no depression diagnoses. One-quarter (24.9%; n=43) reported low resilient coping, 46.8% (n=81) moderate resilient coping, and 28.3% (n=49) high resilient coping scores.

Table 2.

Summary of Overall and Subscale Scores Across Variables (n=173)

Variable	Mean	Standard deviation	Range	Score on 0–100 scale
Gender discrimination	26.6	8.6	8–48	46.5
Racial discrimination	29.6	7.9	8–48	54.0
HIV-related stigma (overall)	35.3	7.8	10–50	63.3
Personalized HIV-related stigma	10.6	3.8	3–15	63.3
HIV-related disclosure	8.6	2.2	2–10	82.5
HIV-related negative self-image	7.3	3.9	3–15	35.8
HIV-related public attitudes	8.4	2.1	2–10	80.0
Social support (overall)	56.9	17.9	19–95	49.9
Emotional support	12.1	4.8	4–20	50.6
Informational support	12.6	4.5	4–20	53.8
Tangible support	11.1	4.7	4–20	44.4
Affectionate support	9.6	4.4	3–15	55.0
Positive social interactions	11.3	4.8	4–20	45.6
Resilient coping	14.8	3.6	4–20	67.5
Depression	5.3	4.2	0–18	29.4

Overall scores (0–100 scale) included: HIV-related stigma (63.3), racial discrimination (54.0), gender discrimination (46.5), resilient coping score (67.5), and social support (overall score) (49.9). The highest HIV-related stigma subscale scores were HIV-related disclosure (82.5) and HIV-related public attitudes (80.0). The highest social support subscale scores were affectionate (55.0) and informational (53.8) support.

Associations between variables

Bivariate correlations between independent, moderator, and dependent variables are presented in Table 3. Racial discrimination was positively correlated with gender discrimination, HIV-related stigma subscales, and depression; racial discrimination was negatively correlated with each social support subscale. Gender discrimination was positively correlated with two HIV-related stigma subscales (personalized stigma, negative self-image) and depression, and negatively correlated social support subscales and coping. Income (not shown) was not significantly correlated with independent, moderator, or dependent variables so was not included in regression analyses.

Table 3.

Bivariate Correlations Among Study Variables (n=173)

Variables	1	2	3	4	5	6	7	8	9	10	11	12
1. Racial discrimination
2. Gender discrimination	0.72^**
3. HIV-related stigma: personalized	0.31^**	0.21^*
4. HIV-related stigma: disclosure	0.31^**	0.03	0.57^**
5. HIV-related stigma: negative self-image	0.18^*	0.18^*	0.311^**	0.25^**
6. HIV-related stigma: public attitudes	0.22^**	0.10	0.47^**	0.61^**	0.35^**
7. Resilient coping	−0.13	−0.17^*	−0.29^**	−0.03	−0.31^**	−0.05
8. Social support: informational	−0.36^**	−0.29^**	−0.31^**	−0.13	−0.18^*	−0.13	0.43^**
9. Social support: emotional	−0.29^**	−0.23^**	−0.42^**	−0.24^**	−0.23^**	−0.23^**	0.41^**	0.82^**
10. Social support: tangible	−0.25^**	−0.25^**	−0.36^**	−0.14	−0.19^*	−0.15	0.26^**	0.69^**	0.65^**
11. Social support: affectionate	−0.25^**	−0.24^**	−0.26^**	−0.07	−0.19^*	−0.12	0.40^**	0.64^**	0.69^**	0.59^**
12. SS-PSI	−0.35^**	−0.30^**	−0.42^**	−0.16	−0.27^**	−0.24^**	0.30^**	0.70^**	0.78^**	0.67^**	0.77^**
13. Depression	0.28^**	0.25^**	0.25^**	0.15	0.45^**	0.22^**	−0.42^**	−0.29^**	−0.30^**	−0.19^*	−0.25^**	−0.28^**

p<0.05; ^** p<0.01.

We conducted independent sample t-tests to assess associations between independent, moderator/mediator, and dependent variables and categorical socio-demographic variables (not shown). Participants reporting being ‘single’ had significantly higher depression scores than those reporting being in a relationship; t(144)=1.99, p<0.05. Participants that did not have children they were caring for had significantly higher overall social support scores than participants who were caring for children; t(119)=2.18, p<0.05, and higher informational support, t(140)=2.27, p<0.05, affectionate support; t(139)=2.69, p<0.01 and positive social interaction, t(138)=2.00, p<0.05. Receiving disability benefits was significantly associated with lower tangible social support, t(161)=−2.08, p<0.05.

Subtypes of stigma and discrimination as correlates of depression

Regression analyses (Table 4) revealed that racial discrimination, gender discrimination, HIV-related stigma subscales (personalized, disclosure, negative self-image, and public attitudes subscales), and socio-demographic variables (relationship status, children, disability benefits) accounted for a significant amount of variability in overall depression scores, adjusted R²=0.21, F(9, 140)=5.38, p<0.001. HIV-related stigma–negative self-image subscale was a significant correlate. In block 2, social support (subscales: emotional support, informational support, tangible support, affectionate support, positive social interaction) and resilient coping accounted for a significant proportion of depression variance after controlling for the effects of the stigma, discrimination, and socio-demographic variables, R² change=0.08, F(6, 134)=2.55, p<0.05. Resilient coping was a significant correlate.

Table 4.

Regression Analyses of Predictors of Depression Among HIV-Positive African Caribbean Women Participants (n=173)

Model	R	R ²	Adj. R ²	Std. error estimate	R ² change	F change	Df	p
1	0.47	0.26	0.21	3.52	0.26	5.38	9, 140	0.000^*
2	0.58	0.33	0.26	3.40	0.08	2.55	6, 134	0.023^*

		Unstandardized coefficients
Model		B	Std. error	Standardized coefficients beta	t	p
1	Constant	−0.22	1.73		−0.13	0.89
	Racial discrimination	0.08	0.05	0.17	1.57	0.12
	Gender discrimination	0.01	0.05	0.01	0.12	0.90
	HIV-related stigma—personalized	0.37	0.29	0.11	1.35	0.21
	HIV-related stigma—disclosure	−0.30	0.36	−0.08	−0.82	0.41
	HIV-related stigma—neg. self image	1.11	0.25	0.37	4.54	0.00^*
	HIV-related stigma—public attitudes	0.06	0.38	0.02	0.16	0.87
	Relationship status (single)	0.81	0.60	0.10	1.35	0.18
	Receiving disability benefits	−0.17	0.59	−0.02	−0.29	0.77
	Caring for children	−0.44	0.67	−0.05	−0.66	0.51
2	Constant	6.83	2.64		2.59	0.01
	Racial discrimination	0.08	0.05	0.16	1.54	0.13
	Gender discrimination	−0.01	0.05	−0.02	−0.20	0.84
	HIV-related stigma—personalized	0.05	0.31	0.02	0.16	0.88
	HIV-related stigma—disclosure	−0.18	0.36	−0.05	−0.51	0.61
	HIV-related stigma—neg. self image	0.86	0.25	0.29	3.44	0.00^*
	HIV-related stigma—public attitudes	0.11	0.37	0.03	0.31	0.76
	Relationship status (single)	0.59	0.59	0.08	1.00	0.32
	Receiving disability benefits	0.06	0.59	0.01	0.09	0.93
	Caring for children	−0.92	0.67	−0.11	−1.37	0.17
	Resilient coping	−0.25	0.10	−0.23	−2.61	0.01^*
	Social support—emotional support	−0.31	0.47	−0.09	−0.67	0.51
	Social support—informational support	−0.33	0.47	−0.09	−0.69	0.49
	Social support—tangible support	0.18	0.37	0.06	0.49	0.62
	Social support—affectionate support	0.10	0.31	0.04	0.32	0.75
	Social support—positive soc interaction	−0.11	0.47	−0.03	−0.24	0.81

p<0.05.

With regards to the depression–affective subscale, racial discrimination, gender discrimination, HIV-related stigma subscales accounted for a significant amount of variability in scores, adjusted R²=0.08, F(6, 149)=3.25, p<0.01 (not shown). HIV-related negative self-image was a significant correlate. Social support subscales and resilient coping did not account for a significant proportion of variance in depression–affective subscale scores in block 2, R² change=0.05, F(6, 143)=1.28, p=0.27.

Racial discrimination, gender discrimination, and HIV-related stigma subscales accounted for a significant amount of variability in depression–cognitive subscale scores, adjusted R²=0.20, F(6, 147)=7.37, p<0.0001 (not shown). Only HIV-related negative self-image was a significant correlate. Social support subscales and resilient coping accounted for a significant proportion of variance in depression–cognitive subscale scores in block 2, R² change=0.11, F(6, 141)=3.80, p=0.01. Resilient coping was significantly correlated with lower depression–cognitive subscale.

Social support and resilient coping as moderators and mediators

Regression analyses (not shown) indicated that, after entering the three correlates (racial discrimination, gender discrimination, HIV-related stigma—total score) and two moderators (resilient coping, social support—total score), the interaction terms did not account for significant variance in depression, R² change=0.03, F(7, 160)=0.76, p=0.62. No interaction terms were significant, indicating that social support and resilient coping did not moderate the impact of the correlates on depression.

In the mediation analyses results (not shown), the direct effect of the three correlates (HIV-related stigma, racial discrimination, gender discrimination) on depression (R²=0.06, p<0.01) remained significant after controlling for the mediators. This suggests that social support and resilient coping do not fully mediate the effect of the correlates on depression. Only resilient coping had a significant indirect effect, partially mediating the impact of HIV-related stigma on depression, effect=0.03, p<0.05.

Discussion

Our finding that HIV-related stigma, gender discrimination, and racial discrimination were significantly correlated with one another and, with depression, support the intersectional stigma model's inclusion of multiple forms of stigma. Results highlight substantial levels of depression among HIV-positive ACB women in Ontario, with approximately one-third of participants reporting moderate/severe depression. A national Canadian mental health survey indicated that approximately 6% of women in Canada met the criteria for major depressive disorder in the last 12 months, and reported similar rates of mood disorders among people with chronic illness.⁴⁶ Our findings therefore suggest that the depression rate among this sample of HIV-positive ACB women is 5-fold higher than the rate of depression among Canada's general population of women and chronically ill persons. HIV-related stigma (overall score, negative self-image subscale) was associated with higher depression scores among HIV-positive ACB women. Associations between HIV-related stigma and depression corroborate research in North America with PLHIV that has highlighted associations between HIV-related stigma and deleterious mental health outcomes.¹²

Although racial discrimination and gender discrimination were correlated with higher rates of depression in bivariate analyses, these variables were not significant correlates of depression when included with HIV-related stigma in regression analyses. This suggests that HIV-related stigma may play a larger role in predicting depression among ACB women than gender/racial discrimination, supported by the higher overall HIV-related stigma scores. Meyer's thoughtful discussion of conceptual and measurement challenges regarding prejudice as stress highlights other potential explanations for this finding: there may be little within-group variability regarding racial/gender discrimination between African, Caribbean, and Black women, whereas there may be more variability regarding HIV-positive serostatus (e.g., beliefs/values regarding HIV).⁴⁷ It is also possible that psychological and/or socio-demographic variables may influence perception, experiences, and reporting of discrimination.⁴⁷ For instance, researchers have highlighted differences between African, Caribbean, and African American PLHIV in the U.S. with regard to partner elicitation, notification, and HIV-testing, suggesting that understanding socio-cultural differences is a fertile area for future research.⁴⁸

Additionally, while some researchers include discrimination in conceptualizations of stigma (e.g., ‘enacted stigma’),⁴⁹ others differentiate discrimination (a behavior) from stigma (an attitude).⁵⁰ Hence we may have missed important dimensions of social stress processes by only measuring discrimination—rather than stigma—associated with gender and race/ethnicity.⁷ Measuring these multiple forms of stigma and discrimination is furthermore complicated by the sexist and racist stereotypes that have permeated HIV discourse since the epidemic's beginning;²³ qualitative work highlights that race and gender continue to shape WLWH's experiences with HIV-related stigma.^5,51 Therefore, it may be difficult to disentangle the separate effects of racial and gender discrimination from HIV-related stigma; researchers have highlighted challenges in measuring intersectionality quantitatively.⁵² New theoretical models and methods may therefore be needed to better understand the intersections of HIV-related stigma, racial discrimination, and gender discrimination among HIV-positive ACB women.^7,52

Social support and resilient coping did not moderate the impact of HIV-related stigma, racial discrimination, or gender discrimination on depression. Resilient coping, however, did partially mediate the effect of HIV-related stigma on depression, corroborating previous research.³⁰ Our findings highlight high rates of depression among HIV-positive ACB women—primarily recruited from networks and organizations that provide psychosocial support—associated with the negative self-image HIV-related subscale. This suggests that social support may not be sufficient to reduce internalized forms of HIV-related stigma such as negative self-image. There has been limited assessment of social support and resilient coping as moderators/mediators of HIV-related stigma, racial/gender discrimination, and inconsistent findings.⁷ Our study provided no evidence to support the hypothesis that social support/resilient coping moderate the effects of stigma and limited evidence to support the hypothesis that resilient coping mediates the relationship between stigma and depression.

Our findings suggest HIV-related negative self-image (NSI) is a phenomenon associated with both affective and cognitive depression subscales. The HIV-related NSI subscale addresses internalized stigma, an individual's negative self-concept and internal sense of shame and blame associated with being HIV positive.⁶ One BDI-FS cognitive subscale item (self-dislike) may conceptually overlap with HIV-related NSI. However, HIV-related NSI is associated with both the affective and cognitive dimensions of depression, suggesting that although HIV-related NSI may be conceptually related to cognitive dimensions of depression, it remains a distinct social-cognitive phenomenon that also impacts affective dimensions of depression (e.g., emotions).

The study design has several limitations. First, the generalizability of findings is reduced by the small sample size, nonprobability sample, and cross-sectional design. The sample was recruited from organizations providing psychosocial support so may have oversampled PLHIV with depression. Second, the recommended sample size for 12 correlates in MLR was 184; our sample of 171 may therefore have not been sufficiently powered for the MLR. Additionally, the sample size may have been too small to detect interaction effects that often have low statistical power.⁵³ Third, the gender and racial discrimination scales did not explore the multiple dimensions of stigma that were explored in the HIV-related stigma scale. When all HIV-related stigma subscales were included in MLR, only the negative self-image subscale was associated with depression, highlighting both the salience of examining stigma subtypes and the need for more complex racial/gender discrimination measures in future research with this population. Given these limitations, future research could utilize larger sample sizes, different sampling techniques to enhance generalizability (e.g., respondent driven sampling), and employ measures of racism and sexism that address internalized, enacted, and perceived domains.

Our study has several strengths despite these limitations. First, this study contributes to theoretical development by assessing the inclusion of HIV-related stigma, racial discrimination, and gender discrimination as stressors in the intersectional stigma model. To our knowledge, this is the first study that assessed the associations between these three forms of stigma/discrimination; our findings therefore contribute to understanding the importance of an intersectional approach to research with WLWH. Second, findings highlight the salience of examining depression among HIV-positive ACB women who appear to bear a disproportionate burden of depression. Third, the current study highlighted the important role negative self-image associated with HIV, indicative of internalized stigma, may play in contributing to depression. Furthermore, the roles social support and coping styles play in moderating/mediating depression have been under-examined in HIV research and our analyses revealed resilient coping may partially mediate the relationship between HIV-related stigma and depression.

Enhanced understanding of the mental health impacts of stigma and discrimination experienced by HIV-positive ACB women can guide the development and evaluation of multi-level interventions to promote health and reduce stigma. Micro-level interventions may include counseling to challenge negative self-image and strategies to build resilient coping skills and address depression; meso-level interventions can include campaigns to challenge social norms and values underpinning intersectional stigma and to enhance social support for WLWH.⁶ Structural interventions could include training health care providers and AIDS service organizations to address racial/gender discrimination as well as HIV-related stigma and to integrate depression assessments and education into treatment, support, and care programming. Implementing strategies that concomitantly build resilient coping, address depression, and challenge intersectional stigma can promote mental health among diverse WLWH.

Footnotes

Author Disclosure Statement

No competing financial interests exist. This study was funded by the Canadian Institutes of Health Research (CIHR). Funders had no role in study design, data collection, analysis, or interpretation.

References

Public Health Agency of Canada (PHAC). HIV and AIDS in CanadaSurveillance Report toDecember 31, 2008 Ottawa: Public Health Agency of Canada, Surveillance and Risk Assessment Division, 2009.

Sherr

, Clucas

, Harding

, Sibley

, Catalan

. HIV and depression—A systematic review of interventions. Psychol Health Med, 2011; 16:493–527.

Newman

, Williams

, Massaquoi

, Brown

, Logie

. HIV prevention for Black women: Structural barriers and opportunities. J Health Care Poor Underserved, 2008; 19:829–841.

Williams

, Newman

, Sakamoto

, Massaquoi

. HIV prevention and the social organization of risk for Black women in Canada. Soc Sci Med, 2009; 68:12–20.

Logie

, James

, Tharao

, Loutfy

. HIV, gender, race, sexual orientation, and sex work: A qualitative study of intersectional stigma experienced by HIV-positive women in Ontario, Canada. PLoS Med, 2011; 8:e1001124.

Mahajan

, Sayles

, Patel

et al. Stigma in the HIV/AIDS epidemic: A review of the literature and recommendations for the way forward. AIDS, 2008; 22:S67–79.

Stuber

, Meyer

, Link

. Stigma, prejudice, discrimination and health. Soc Sci Med, 2008; 67:351–357.

Jones

. Levels of racism: A theoretic framework and a gardner's tale. Am J Public Health, 2000; 90:1212–1215.

Williams

. Racial/ethnic variations in women's health: The social embeddedness of health. Am J Public Health, 2002; 92:588–597.

10.

Brondolo

, Gallo

, Myers

. Race, racism and health: Disparities, mechanisms, and interventions. J Behav Med, 2009; 32:1–8.

11.

Williams

, Neighbors

, Jackson

. Racial/ethnic discrimination and health: findings from community studies. Am J Public Health, 2003; 93:200–208.

12.

Logie

, Gadalla

. Meta-analysis of health and demographic correlates of stigma towards people living with HIV. AIDS Care, 2009; 21:742–753.

13.

Lee

, Henderson

, Patel

. A UN summit on global mental health. Lancet, 2010; 376:516.

14.

W.H.O Depression. Geneva: World Health Organization (WHO), 2010.

15.

Leserman

. The effects of stressful life events, coping, and cortisol on HIV infection. CNS Spectr, 2003; 8:25–30.

16.

Whetten

, Reif

, Ostermann

et al. Improving health outcomes among individuals with HIV, mental illness, and substance use disorders in the Southeast. AIDS Care, 2006; 18:S18–26.

17.

Bravo

, Edwards

, Rollnick

, Elwyn

. Tough decisions faced by people living with HIV: A literature review of psychosocial problems. AIDS Rev, 2010; 12:76–88.

18.

Shorter-Gooden

. Multiple resistance strategies: How African American women cope with racism and sexism. J Black Psychol, 2004; 30:19.

19.

Borrell

, Artazcoz

, Gil-Gonzalez

, Perez

, Rohlfs

, Perez

. Perceived sexism as a health determinant in Spain. J Women's Health, 2010; 19:741–750.

20.

Paradies

. A systematic review of empirical research on self-reported racism and health. Int J Epidemiol, 2006; 35:888–901.

21.

McLaughlin

, Hatzenbuehler

, Keyes

. Responses to discrimination and psychiatric disorders among Black, Hispanic, female, and lesbian, gay, and bisexual individuals. Am J Public Health, 2010; 100:1477–1484.

22.

Szymanski

, Stewart

. Racism and sexism as correlates of African American women's psychological distress. Sex Roles, 2010; 63:226–238.

23.

Parker

, Aggleton

. HIV and AIDS-related stigma and discrimination: A conceptual framework and implications for action. Soc Sci Med, 2003; 57:13–24.

24.

Sandelowski

, Barroso

, Voils

. Gender, race/ethnicity, and social class in research reports on stigma in HIV-positive women. Health Care Women Int, 2009; 30:273–288.

25.

Wingood

, Diclemente

, Mikhail

et al. HIV discrimination and the health of women living with HIV. Women Health, 2007; 46:99–112.

26.

Canady

, Bullen

, Holzman

, Broman

, Tian

. Discrimination and symptoms of depression in pregnancy among African American and White women. Womens Health Issues, 2008; 18:292–300.

27.

West

, Donovan

, Roemer

. Coping With racism: What works and doesn't work for Black women? J Black Psychol, 2010; 36:18.

28.

Greer

. Coping strategies as moderators of the relationship between individual race-related stress and mental health symptoms for African American Women. Psychol Women Quart, 2011; 35:215–226.

29.

Brondolo

, Brady ver Halen

, Pencille

, Beatty

, Contrada

. Coping with racism: A selective review of the literature and a theoretical and methodological critique. J Behav Med, 2009; 32:64–88.

30.

Heckman

, Anderson

, Sikkema

, Kochman

, Kalichman

. Emotional distress in non-metropolitan persons living with HIV disease enrolled in a telephone-delivered, coping improvement group intervention. Health Psychol, 2004; 23:94–100.

31.

Szymanski

, Owens

. Group-level coping as a moderator between heterosexism and sexism and psychological distress in sexual minority women. Psychol Women Quart, 2009; 33:197–205.

32.

Galvan

, Davis

, Banks

, Bing

. HIV stigma and social support among African Americans. AIDS Patient Care STDs, 2008; 22:423–436.

33.

Lincoln

, Chatters

, Taylor

, Jackson

. Profiles of depressive symptoms among African Americans and Caribbean Blacks. Soc Sci Med, 2007; 65:200–213.

34.

Schulz

, Israel

, Zenk

et al. Psychosocial stress and social support as mediators of relationships between income, length of residence and depressive symptoms among African American women on Detroit's eastside. Soc Sci Med, 2006; 62:510–522.

35.

Song

. Social capital and psychological distress. J Health Soc Behav, 2011; 52:478–492.

36.

Shields

. Gender: An intersectionality perspective. Sex Roles, 2008; 59:301–311.

37.

Collins

. Black Feminist Thought: Knowledge, Consciousness, and the Politics of Empowerment. New York: Routledge, 1991.

38.

Flicker

, Guta

, Larkin

et al. Survey design from the ground up: Collaboratively creating the Toronto Teen Survey. Health Promot Pract, 2010; 11:112–122.

39.

Wright

, Naar-King

, Lam

, Templin

, Frey

. Stigma scale revised: Reliability and validity of a brief measure of stigma for HIV+ youth. J Adolesc Health, 2007; 40:96–98.

40.

Clark

, Coleman

, Novak

. Brief report: Initial psychometric properties of the everyday discrimination scale in black adolescents. J Adolesc, 2004; 27:363–368.

41.

Sherbourne

, Stewart

. The MOS social support survey. Soc Sci Med, 1991; 32:705–714.

42.

Sinclair

, Wallston

. The development and psychometric evaluation of the brief resilient coping scale. Assessment, 2004; 11:94–101.

43.

Beck

, Guth

, Steer

, Ball

. Screening for major depression disorders in medical inpatients with the Beck Depression Inventory for Primary Care. Behav Res Ther, 1997; 35:785–791.

44.

Preacher

KJ HA

. Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behav Res Methods, 2008; 40:879–991.

45.

Shrout

, Bolger

. Mediation in experimental and nonexperimental studies: New procedures and recommendations. Psychol Methods, 2002; 7:422–445.

46.

Government of Canada. The human face of mental health and mental illness in Canada. Minister of Public Works and Government Services: Ottawa, 2006.

47.

Meyer

. Prejudice, social stress, and mental health in lesbian, gay, and bisexual populations: Conceptual issues and research evidence. Psyc Bulletin, 2003; 129:674–697.

48.

Tsega

, Udeagu

, Begier

. A comparison of partner notification effectiveness in African-, Caribbean-, and United States-born HIV-infected Blacks in New York City. AIDS Pt Care STDs, 2012; 26:406–410.

49.

Link

BGP

. Conceptualizing stigma. Ann Rev Sociol, 2001; 27:363–385.

50.

Deacon

. Towards a sustainable theory of health-related stigma: Lessons from the HIV/AIDS literature. J Commun Appl Social Psychol, 2006; 16:418–425.

51.

Collins

, von Unger

, Armbrister

. Church ladies, good girls, and locas: Stigma and the intersection of gender, ethnicity, mental illness, and sexuality in relation to HIV risk. Soc Sci Med, 2008; 67:389–397.

52.

Bowleg

. When Black+Lesbian+Woman ≠Black Lesbian Woman: The methodological challenges of qualitative and quantitative intersectionality research. Sex Roles, 2008; 59:312–325.

53.

AIken

. Multiple Regression: Testing and Interpreting Interactions. Newbury Park, CA: Sage, 1991.