Racial and Ethnic Differences in Women’s HIV Risk and Attitudes Towards Pre-Exposure Prophylaxis (PrEP) in the Context of the Substance Use, Violence, and Depression Syndemic

Abstract

Women with syndemic conditions, i.e., two or more co-occurring epidemics, are at elevated risk for HIV acquisition and are therefore prime candidates for pre-exposure prophylaxis (PrEP). However, PrEP uptake remains low among women, especially among Black and Hispanic women. This study examined associations of syndemic conditions with PrEP attitudes and HIV risk among women, and the moderating effect of race and ethnicity. In 2017, 271 non-Hispanic Black, non-Hispanic White, and Hispanic, PrEP-eligible women engaged in care at Planned Parenthood in the northeastern region of the U.S. completed an online survey. Participants reported syndemic conditions (i.e., intimate partner violence, depression, substance use), PrEP attitudes (e.g., PrEP interest), HIV sexual risk (e.g., multiple male sexual partners), and sociodemographics. Structural equation modeling was used to examine the effects of syndemic conditions on PrEP attitudes and HIV risk, and the moderating effect of race and ethnicity. Women with more syndemic conditions had a higher odds of reporting multiple male sexual partners. Syndemic conditions were positively associated with PrEP attitudes for Hispanic women than non-Hispanic Black and White women. Women with syndemic conditions, particularly Hispanic women, may be receptive to interventions promoting PrEP.

INTRODUCTION

In the context of HIV prevention and care, Syndemic Theory and associated frameworks have garnered attention over the past decade. A syndemic is the co-occurrence, interaction, and contribution of two or more epidemics that increase health risk for vulnerable populations (Singer, 2000, 2009; Singer & Clair, 2003). For example, the SAVA Syndemic, the concurrent and mutually reinforcing problems of substance abuse, violence, and HIV/AIDS, is a well-researched barrier to HIV care and treatment (Meyer, Springer, & Altice, 2011). Syndemics research has shown that co-occurring health problems are associated with greater HIV risk among women residing in urban areas (Koblin et al., 2015; W. Batchelder et al., 2016), Hispanic women (Martinez et al., 2017), and women experiencing intimate partner violence (IPV)(Peasant, Sullivan, Weiss, Martinez, & Meyer, 2017). For instance, Martinez et al. (2017) found that the co-occurrence of IPV, substance use, and depression had an additive effect on HIV-related sexual risk among pregnant Hispanic adolescents. Previous studies have illustrated the synergistic mechanisms linking IPV, substance use and depression, which often heightens HIV vulnerability among women (El-Bassel & Strathdee, 2015; Gilbert et al., 2015; Koblin A et al., 2015; Meyer et al., 2011). For example, substance use may be triggered by experiences and/or the mental health sequelae of IPV (Beydoun, Beydoun, Kaufman, Lo, & Zonderman, 2012), which can interfere with safe sex practices (Campbell & Soeken, 1999). Similarly, women who experience IPV are at elevated risk for HIV vulnerability due to forced sex and unequal power dynamics that thwart negotiation of safe sex practices (Stockman, Lucea, & Campbell, 2013). Due to this enhanced risk, the acceptability of safe and effective HIV prevention strategies are urgently needed for these populations of women (Willie T, Kershaw TS, Campbell JC, & Alexander KA, 2018), whose health needs may be greater than those not exposed to syndemic conditions.

One promising HIV prevention strategy is the use of pre-exposure prophylaxis (PrEP), an oral medication taken daily that prevents HIV acquisition among HIV-negative individuals (Centers for Disease Control and Prevention, 2017b). PrEP is a safe and effective method for men who have sex with men (MSM), heterosexually active men and women “at substantial risk for HIV acquisition”, and people who inject drugs (PWID) (Centers for Disease Control and Prevention, 2017b).

Knowledge about attitudes towards PrEP among women – particularly women facing syndemic conditions – is growing (Auerbach, Kinsky, Brown, & Charles, 2015; Bond & Gunn, 2016; Flash et al., 2014; Garfinkel, Alexander, McDonald-Mosley, Willie, & Decker, 2016; Goparaju et al., 2017; Willie T et al., 2018), but research to date has instead focused largely on MSM. Further, emerging research indicates significant relationships between PrEP acceptability and intentions relative to the individual conditions comprising the IPV-substance use-depression syndemic (i.e., IPV, substance use, OR depression) (Braksmajer, Leblanc, El-Bassel, Urban, & McMahon, 2018; Garfinkel et al., 2016; Roberts et al., 2016; Willie et al., 2018; Willie, Stockman, Overstreet, & Kershaw, 2017) but few examine them in combination (Bazzi et al., 2019; Jain et al., 2019). Notably, Bazzi et al. (2019) found that sex workers in Kenya had high PrEP acceptability and need given the co-occurring contexts of substance use and violence. To date, extant research provides a promising avenue for understanding the feasibility of PrEP for women who experience syndemic conditions, but a more comprehensive investigation considering mutually-reinforcing syndemic conditions such as IPV, substance use and depression is critically needed because the entanglement of IPV, substance use, and depression may influence women’s attitudes towards PrEP.

The current study addresses the aforementioned gaps and limitations in two ways. First, to our knowledge, no study has examined the effects of syndemics on women’s PrEP attitudes, despite the fact that PrEP can expand HIV prevention options for women (Willie, Kershaw, Campbell, & Alexander, 2017; Willie, Stockman, et al., 2017) and early research suggests a significant association between syndemics and PrEP engagement among men (Blashill et al., 2019). This point is particularly salient because women experiencing syndemic conditions represent a population at greatest risk for HIV acquisition (Meyer et al., 2011). Second, few studies on syndemics relative to other forms of HIV prevention among women use gender, race, and ethnicity to contextualize their research (Wyatt, Gómez, et al., 2013). Given the disproportionate burden of HIV infections among African American women (Centers for Disease Control and Prevention, 2018), recent research efforts have called for the application of a syndemic intersectional approach to understand HIV risk, prevention, and interventions for African American and Hispanic communities (Wyatt, Gomez, et al., 2013). Better understanding how race and ethnicity may influence the effects of syndemics on women’s attitudes towards PrEP could help inform gender-specific, culturally-tailored, and effective PrEP interventions, programs, and policies.

The primary objective of this study was to examine the association of syndemic conditions (i.e., IPV, substance use and depression) with (a) PrEP-related attitudes (i.e., interest, intentions, comfort discussing PrEP with provider) and (b) HIV risk indicators (i.e., condomless sex, multiple sexual partners) among women engaged in care at U.S. Planned Parenthood centers. The secondary objective was to explore differences in these associations among non-Hispanic Black, non-Hispanic White, and Hispanic women.

METHODS

Methods described below have also been reported elsewhere (Calabrese, Dovidio, et al., 2018; Calabrese, Willie, et al., 2019; Tekeste et al., 2019).

Procedures

Participants were recruited for an anonymous online survey distributed to 11,238 patients in February of 2017. A survey invitation was sent via email to patients 18 years of age or older who had recently engaged in care at Planned Parenthood centers in New Haven, Bridgeport, and Hartford—the three cities reporting the highest incidence of HIV infections in Connecticut (State of Connecticut Department of Public Health, 2017). Patients were defined as “recently engaged in care” if they had one or more visits documented in their medical record in the past ten months, and the survey link was only sent to patients who agreed to receive email communication from Planned Parenthood (77%). Interested patients clicked a survey link in the recruitment email, completed an online consent form, and responded to survey questions (median time = 36 minutes). Participants were compensated with $10 gift cards. The survey closed to new participants within 100 hours of sending the recruitment email, once 973 had enrolled, in order to avoid exceeding the recruitment maximum of 1000. All procedures were approved by the Yale University Human Investigation Committee institutional review board.

For this present study, we were interested in examining the impact of syndemic conditions among PrEP-eligible women. As a result, we restricted the analytic sample to include only participants who identified as women, reported no prior HIV diagnosis, reported having vaginal sex with a male sexual partner in the past 6 months, and responded to all PrEP eligibility questions based on the guidance summary criteria located in the 2017 CDC PrEP Clinical Guidelines (Centers for Disease Control and Prevention, 2017b). Specifically, women were included if they reported 1 or more of the following in the past 6 months: (a) sexual partner living with HIV, (b) bacterial STI, (c) multiple sexual partners (2 or more), (d) history of inconsistent or no condom use, (e) commercial sex work (Calabrese et al., 2019). The other criterion in the guidelines, high area/network HIV prevalence, was not directly measured or considered for inclusion. It is important to note while both cisgender and transgender women were eligible to participate in the parent study, none of the women in the parent study identified as a “transgender woman” and all identified as a “woman”; thus, we presume most or all of the women in our sample were cisgender.” Additionally, because IPV was one of the syndemic conditions of interest and was operationalized over a 1-year relationship timespan according to the measure used, the sample was restricted to women who reported having a sexual or romantic partner with whom they had been in a relationship for at least one year. We also restricted the analyses to women with complete data for all study variables; of those who otherwise qualified for inclusion, 8 participants were excluded due to incomplete data on 1 or more of the syndemic measures. The final analytic sample included 271 women.

Measures

Sociodemographic Characteristics included gender, race/ethnicity, sexual orientation, education, employment, household income, age, and geographic location.

Syndemic conditions were measured through questions about IPV, depression, and substance use. IPV was examined using a short form of the revised Conflict Tactics Scale (CTS2S) (Straus & Douglas, 2004). An IPV binary variable was created from four forms of violence involving physical assault or sexual coercion, and participants who reported any of these forms of violence at least once in the past year with their main partner were assigned the value of (1) experienced IPV. Creating and analyzing IPV as a binary variable has been utilized in a number of studies (Garfinkel et al., 2016; Willie, Powell, & Kershaw, 2016; Willie et al., 2019; Willie, Stockman, et al., 2017). Participants who reported experiencing none of these events in the past year within their current relationship were assigned the value of (0) IPV not experienced. Depression was examined using the Patient Health Questionnaire 9 (PHQ-9) (Kroenke, Spitzer, & Williams, 2001). Participants reported how often they experience nine symptoms of depression (0=not at all to 3=nearly every day) and the frequencies were summed to determine depression severity. The total score was used to create a binary depression variable (0–9=no current depression; 10+=current depression) (Manea, Gilbody, & McMillan, 2012). Substance use included injection drug use (past six months) and recreational use of non-injection drugs (past six months). Participants were asked to respond as Yes/No; I don’t know; and I prefer not to say. Due to the low frequency of women reporting injection drug use (1%), the substance use variable combined women reporting injection drug use and non-injection drug use. Thus, participants who answered in the affirmative were coded as using substances. A count variable was constructed for syndemic conditions. The count variable was created by summing the affirmative responses from each syndemic condition, with possible scores ranging from 0 (no syndemic conditions) to 3. Due to the low frequency of women reporting 3 syndemic conditions, the syndemic count variable combined women reporting 2 and 3 syndemic conditions. As a result, the final syndemic condition variable was coded as 0, 1, or (2 or 3) conditions.

PrEP Attitudes was a latent variable comprising three measured constructs: 1) PrEP interest, 2) PrEP intentions, and 3) Comfort discussing PrEP with a provider. Before viewing the PrEP attitudes measures, participants were presented with a brief introduction to PrEP, including basic facts about dosing, effectiveness, and side effects. PrEP interest was assessed by asking participants: “How interested are you in learning more about PrEP (daily HIV prevention pill)?”. Participants were able to respond on a 5-point Likert scale ranging from (1) Not At All Interested to (5) Extremely Interested. PrEP intention was assessed by asking participants: “How likely would you be to take PrEP (daily HIV prevention pill) if it were available for free?” Participants were able to respond on a 5-point Likert scale ranging from (1) Definitely Would Not Take PrEP to (5) Definitely Would Take PrEP. Comfort discussing PrEP with a provider was assessed by asking participants: “How comfortable would you be talking with a health care provider about PrEP (daily HIV prevention pill)?” Participants were able to respond on a 5-point Likert scale ranging from (1) Not At All Comfortable to (5) Extremely Comfortable. The measurement model for the PrEP attitudes latent variable was assessed to confirm that the observed variables loaded adequately. The measurement model was just-identified, $X^2\left(0\right)=0$, RMSEA = 0.00, CFI = 1.00, TLI = 1.00, with factor loadings greater than 0.50 and a Cronbach’s alpha of 0.60.

HIV sexual risk indicators included condomless vaginal or anal sex with a male partner (past six months); and vaginal or anal sex with multiple (two or more) male partners (past six months).

Data Analysis

Descriptive statistics (frequencies, means) were conducted to describe sample characteristics. PrEP attitudes was operationalized as a latent variable composed of three observed indicators (i.e., PrEP interest, PrEP intentions, Comfort discussing PrEP with a provider). Structural equation modeling was conducted to examine syndemic conditions as a predictor of the 3 outcomes (PrEP attitudes, condomless sex, and multiple male sexual partners). Next, a separate multigroup model was conducted to determine if the relationship between syndemic conditions, HIV risk behavior, and PrEP attitudes were different for non-Hispanic Black, non-Hispanic White, and Hispanic women (effect modification). Standard goodness of fit statistics for confirmatory factor analysis were used to determine the model’s fit: factor loadings ≥ 0.40. The structural equation model predicting HIV risk behavior and PrEP attitudes was evaluated for good fit with: 1) a RMSEA less than .08, and 2) a CFI and TFI around .90. All analyses were conducted using SAS 9.4 (SAS Institute, 1990) and Mplus 7.2. (Muthén & Muthén, 2012).

RESULTS

Sociodemographic Characteristics

Table 1 describes the women in the present study. As noted previously, we presume that most or all of the women in our sample were cisgender because none of the women in the sample identified as transgender women. Approximately one-fourth of women self-identified as Hispanic (28.3%), while slightly more than one-third of women were non-Hispanic White (36.4%) and non-Hispanic Black (35.3%). The majority of women (72.1%) reported full-time or part-time employment. Three-fourths of women self-identified as heterosexual (75.0%), 14.3% (n=39) self-identified as bisexual, and 10.7% (n=29) self-identified as lesbian or reported another identity.

Table 1.

Sample Characteristics of 271 PrEP-Eligible Women Engaged in Care

Study Variables	N (%)
Race and Ethnicity
Hispanic	77 (28.3)
Non-Hispanic White	98 (36.4)
Non-Hispanic Black	96 (35.3)
Employment
Not working	30 (11.0)
Not in workforce (e.g., student, homemaker)	46 (16.9)
Working full-time or part-time	19 (72.1)
Sexual Orientation
Bisexual	39 (14.3)
Lesbian or another identity	29 (10.7)
Heterosexual	203 (75.0)
Age
Greater than 25 years of age	167 (61.8)
Less than 25 years of age	104 (38.2)
Syndemic Conditions
Intimate partner violence	54 (19.9)
Depression	51 (18.8)
Substance use	43 (15.9)
# of Syndemic Conditions
No syndemic conditions	158 (58.3)
1 syndemic condition	80 (29.5)
2 or more syndemic conditions	33 (12.2)
Comfort discussing PrEP with provider
Not at all comfortable	22 (8.1)
A little bit comfortable	25 (9.2)
Somewhat comfortable	66 (24.4)
Very comfortable	89 (32.8)
Extremely comfortable	67 (24.7)
PrEP interest
Not at all interested	97 (35.8)
A little bit interested	69 (25.5)
Somewhat interested	63 (23.3)
Very interested	22 (8.1)
Extremely interested	18 (6.6)
PrEP intention
Definitely would not take PrEP	65 (23.9)
Probably would not take PrEP	54 (19.9)
Might take PrEP	68 (25.1)
Probably would take PrEP	38 (14.0)
Definitely would take PrEP	44 (16.2)
HIV Risk Behaviors
Multiple sexual partners	57 (21.0)
Condomless sex	260 (95.9)

Note. Column percentages rounded to the nearest tenth and may not equal 100.

Most of the women had no syndemic condition (58.3%), 80 women (29.5%) had 1 syndemic condition, 31 women (11.4%) had 2 syndemic conditions, and 2 women (0.7%) had 3 syndemic conditions. 91.8% of women expressed some level of comfort in discussing PrEP with a healthcare provider (i.e., were “a little bit”, “somewhat,” “very,” or “extremely” comfortable). 64.0% of women expressed some level of interest in learning more about PrEP (i.e., were “a little bit”, “somewhat,” “very,” or “extremely” interested). 30.5% of women reported they “probably” or “definitely” would take PrEP if it were freely available. More than one-fifth of women (21.0%) reported having multiple sexual partners in the past six months. The majority of women (95.9%) reported having condomless vaginal or anal sex in the past six months.

Structural Equation Modeling Results of Syndemic Conditions, PrEP Attitudes, and HIV Risk

Assessing Fit for Initial and Multigroup Models

The overall fit for the initial and multigroup (i.e., effect modification) models was assessed for PrEP attitudes and HIV risk while controlling for covariates. A satisfactory fit was found for both models: initial model (CFI=.98, TLI=.95, RMSEA=.03) and multigroup model (CFI=.94, TLI=.89, RMSEA=.02).

Within Overall Sample

In the unadjusted model among all women in the sample, syndemic conditions were associated with PrEP attitudes [B=.15; P=.03] and with having multiple sexual partners [B=.29; P=.009], such that women with syndemic conditions expressed more positive attitudes toward PrEP and were more likely to have multiple partners. No association was observed between syndemic conditions and condomless sex [B=−.05, P=.60]. The final adjusted model is presented as Figure 1. After adjustment for sociodemographic characteristics (i.e., employment, sexual orientation, age), the association between syndemic conditions and having multiple sexual partners remained among all women in the sample [B= .28; P=.01]. In the adjusted model, there were no associations of syndemic conditions with condomless sex [B=.14; P=.73] or PrEP attitudes [B=.13; P=.09].

Figure 1.

Final adjusted model of relationships between syndemic conditions, PrEP attitudes, condomless sex, and multiple sexual partners among non-Hispanic Black, non-Hispanic White, and Hispanic women. Standardized regression coefficients are shown. Significant paths (solid lines) and non-significant paths (dashed lines) are shown. Variables adjusted for in the model included employment, sexual orientation, and age. Model Fit: RMSEA = .03, CFI = .98,TLI = .95. The assessment of effect modification of the adjusted model (not shown) indicated syndemics and PrEP attitudes were associated among Hispanic women.

Effect Modification of Race and Ethnicity

The combined race and ethnicity variable was an effect modifier only for the association of syndemic conditions with PrEP attitudes. Specifically, syndemic conditions were associated with PrEP attitudes among Hispanic women [B=.50, P=.01]. Syndemic conditions were not associated with PrEP attitudes among non-Hispanic Black [B=.10, P=.64] and non-Hispanic White women [B=−.04, P=.83]. Results from the chi-square difference test illustrate that the structural paths of syndemic conditions to PrEP attitudes was different across the three racial and ethnic groups (chi-square difference=14.78, P=.02).

DISCUSSION

Nearly 19% of all new HIV diagnoses in the U.S. are among women and heterosexual contact accounts for the majority of HIV diagnoses (Centers for Disease Control and Prevention, 2017a), yet less than 5% of women are using PrEP (Siegler et al., 2018). In addition, research highlights racial and ethnic disparities in both women’s HIV diagnoses (Centers for Disease Control and Prevention, 2017a) and PrEP uptake rates in the U.S. (Ya-lin, Zhu, Smith, Harris, & Hoover, 2018). The current study utilized the syndemic intersectional approach to better understand associations between syndemic conditions, HIV risk, and PrEP attitudes within the context of race and ethnicity. Results from our study illustrate the complex role of syndemic conditions in women’s lives. Overall, women experiencing syndemic conditions appeared to be more likely to engage in HIV-related sexual risk behavior (i.e., have multiple sex partners). Moreover, our examination revealed that the association between syndemic conditions and PrEP attitudes was stronger for Hispanic women than non-Hispanic Black and White women. These findings add to the nascent literature on syndemic conditions and PrEP – which focuses primarily on men’s experiences (Blashill et al., 2019; Tan, Leon-Carlyle, Mills, Moses, & Carvalhal, 2016; Wilton et al., 2018). Collectively, these findings highlight opportunities for future PrEP engagement strategies designed for at-risk women and future research on PrEP engagement in the context of sociocultural factors.

Expanding current research, our findings suggest that women experiencing syndemic conditions are more likely to have multiple sexual partners and to have similarly positive attitudes towards PrEP as compared to those who are not. Although our study did not examine potential mediating factors between syndemic conditions and HIV risk, previous research suggests that psychological consequences of violence such as depression, could explicate this relationship (Meyer et al., 2011). While our findings are consistent with extant research on syndemic conditions and women’s HIV risk (Brennan et al., 2012; Koblin et al., 2015; Martinez et al., 2017; Pitpitan et al., 2013; W. Batchelder et al., 2016), this is one of the first studies to examine the effects of syndemic conditions on PrEP attitudes among U.S. PrEP-eligible women. In our study, the unadjusted and adjusted associations between syndemic conditions and PrEP attitudes were in the positive direction. Thus, it is possible that women experiencing syndemic conditions have similarly or more positive attitudes towards PrEP as compared to women without syndemic conditions. Coupled with previous research (Brennan et al., 2012; Koblin et al., 2015; Martinez et al., 2017; Pitpitan et al., 2013; W. Batchelder et al., 2016), our findings provide some evidence that women experiencing syndemic conditions are not only appropriate candidates for PrEP, but may be receptive to public health messages on PrEP as an HIV prevention option.

It is noteworthy that the positive association between syndemic conditions and PrEP attitudes attenuated in the adjusted model. The adjusted model, employment status was significantly associated with PrEP attitudes and may be an important social identity that shapes women’s attitudes towards PrEP. While few studies to date have explicitly examined the impact of employment status on women’s attitudes towards PrEP, previous research on the social ecological model and the theory of gender and power have documented how social, cultural and economic factors contribute to women’s HIV vulnerability (Aholou, Murray, & Sutton, 2016; Wingood & DiClemente, 2000) and possibly their attitudes towards PrEP. For example, socioeconomic characteristics such as employment and poverty status, can increase women’s susceptibility to economic hardships, sexual risk taking and HIV infection (Aholou et al., 2016). It is possible that women who are unemployed may not be able to afford PrEP or have limited knowledge of PrEP payment assistance programs which could impact their attitudes towards PrEP. Also women who are economically dependent on their male partners may few options to financially access PrEP, and thus view PrEP as unattainable. Altogether, our findings underscore the potential utility of PrEP engagement interventions tailored to women with syndemic conditions, however future interventions may need to address the structural and social factors that influence women’s attitudes towards PrEP.

While greater HIV-related sexual risk emerged among women experiencing syndemic conditions, our effect modification results revealed that, among Hispanic women, syndemic conditions were also associated with more positive attitudes toward PrEP, a prevention strategy that could help to offset risk. The burgeoning literature on syndemics and HIV risk among Hispanic women in the context of sociocultural factors may provide better understanding of this phenomenon. While this body of work provides mixed results, some research suggest that acculturation to the US is associated with greater sexual risk behaviors among Hispanic women (Gonzalez-Guarda et al., 2012). Specifically, Hispanic women with higher levels of acculturation may have more permissive attitudes towards sexual relationships such as multiple sexual partners (Organista, Carrillo, & Ayala, 2004; Rojas-Guyler, Ellis, & Sanders, 2005). Future research should replicate our study and be inclusive of sociocultural factors such as acculturation in order to better explicate women’s experiences of syndemic conditions in the context of their HIV prevention attitudes and behaviors.

Although our findings suggest that syndemic conditions may be more relevant to PrEP attitudes among Hispanic women, this should not be interpreted as an indication that syndemic conditions are irrelevant to the HIV prevention behaviors of Black and White women. Instead our findings illustrate a potential area for future research. To date, research on PrEP among U.S. women is growing (Auerbach et al., 2015; Bradley et al., 2019; Calabrese et al., 2019; Garfinkel et al., 2016; Goparaju et al., 2017; Tekeste et al., 2018; Willie, Kershaw, et al., 2017; Willie et al., 2018), however, few studies have specifically focused on the needs of women with syndemic conditions, especially ethnic minority women. It would be advantageous for future research to utilize qualitative research methods in order to identify social, structural and cultural factors shaping PrEP attitudes among women with syndemic conditions. Based on the syndemic intersectional approach (Wyatt, Gómez, et al., 2013), future research could distinguish how gender roles and identities in ethnic minorities communities influence PrEP attitudes in order to identify recommendations for culturally-tailored biomedical HIV intervention development.

It may be beneficial to incorporate a syndemic screener as part of routine care in clinical settings administering PrEP. Beyond the relevance of syndemic conditions to HIV prevention, research has suggested that syndemic conditions negatively impact HIV treatment adherence (Blashill et al., 2015), and PrEP adherence (Storholm, Volk, Marcus, Silverberg, & Satre, 2017). A study among MSM currently using PrEP found that a self-administered syndemic screener was useful in identifying patients that could benefit from additional mental and community resources while engaging in PrEP (Tan et al., 2016). Extending this concept, a routine syndemic screener for women could be beneficial in clinical settings and could coincide with conversations about PrEP and sexual health more broadly. In particular, a syndemic screener could be a point of entry to other services that may support women’s overall wellbeing.

These findings should be interpreted in the context of the following study limitations. The data were based on participant self-report, which is susceptible to social desirability bias. The findings were based on a cross-sectional study design and causality cannot be inferred. To address this common design limitation, recent research on women’s engagement in PrEP care has used innovative causal inference methods such as doubly robust estimation to estimate the causal effect of IPV (Willie et al., 2019). Future research could use causal inference methods to strength empirical claims of cause-and-effect relationships in non-randomized samples. While these results may be more generalizable to cisgender women than transgender women given the presumed composition of our sample, there is an opportunity for future research to investigate the association of syndemics with HIV risk and PrEP attitudes among transgender women, an important population to prioritize for HIV prevention initiatives given disproportionate rates of infection. Additionally, we did not explore potential variability by type of relationship (e.g., casual, stable primary, sex trade); reported findings may not be generalizable to women in all types of relationships. Also, women were recruited from Connecticut Planned Parenthood centers and may not be representative of women in other geographic areas or with different levels of healthcare engagement.

CONCLUSIONS

Women with syndemic conditions tend to report greater HIV-related sexual risk vs. those who did not and tended to have positive attitudes towards PrEP, particularly Hispanic women with syndemic conditions. Although the majority of PrEP research and marketing focuses on men, there is an opportunity to design PrEP materials to focus on the needs of women with syndemic conditions. Further, additional research is needed to provide better understanding of the influence of sociocultural factors in the context of women’s HIV prevention. Some cultural beliefs and values may empower women and protect sexual health. Finding ways to embody and disseminate these empowering, culturally-congruent messages could be important to engage women in racial and ethnic communities in PrEP interventions and care.