Medical Mistrust and the PrEP Cascade among Latino Sexual Minority Men

Abstract

Latino sexual minority men (SMM) have high HIV incidence rates but report low pre-exposure prophylaxis (PrEP) use. Medical mistrust predicts lower medical care use and could contribute to decreased PrEP engagement. This study examines how medical mistrust relates to the PrEP cascade among 151 Latino SMM aged 18–29. Logistic regressions were employed with medical mistrust predicting PrEP awareness, willingness, current use, and adherence. Greater medical mistrust was associated with decreased odds of all outcomes and may represent a barrier to PrEP engagement for Latino SMM. Further research should explore whether reducing medical mistrust among Latino SMM could increase PrEP engagement.

Introduction

In 2016, sexual minority men (SMM; non-exclusively heterosexual men) accounted for the majority (67%) of all new HIV diagnoses among men in the United States¹. Data suggest that HIV rates may be stabilizing or even decreasing among some racial/ethnic SMM populations, as, from 2011 to 2016, HIV diagnoses decreased 10% for White SMM and stabilized for African American SMM¹. Despite such decreases in HIV diagnoses, a CDC report found that, from 2011 to 2016, Latino SMM had an 18% increase in HIV diagnoses, and in 2017, Latinos represented 29% of new HIV diagnoses among SMM in the United States, second only to African American SMM¹. The disproportionate burden of HIV diagnosis among Latino SMM underlines the importance of research on which factors may be associated to their elevated rates of HIV transmission.

Pre-exposure prophylaxis (PrEP) is an efficacious HIV prevention approach which has been previously shown to reduce the risk of HIV by up to 99% when administered daily².

Despite promising evidence for PrEP to reduce HIV transmission, a study surveying SMM in California recruited through dating and sex apps found that Latino SMM between the ages of 18 and 29 were significantly less likely to use PrEP and reported significantly higher perceived PrEP need in comparison to White SMM². A PrEP cascade model, similar to that of an HIV continuum of care model, is a relevant tool used in the field to understand specific facilitators and barriers that individuals face in achieving PrEP efficacy³. One version of the cascade encompasses four steps: awareness of the existence and applications of PrEP, being willing to take PrEP, having taken PrEP, and adherence to PrEP (often defined as at least 60% adherence to a daily PrEP regimen)⁴. Given the low PrEP use and the high perceived PrEP need among Latino SMM, a PrEP cascade model may be useful in understanding specific factors associated with high and low PrEP awareness, willingness, use, and adherence among Latino SMM.

The concept of medical mistrust encompasses the lack of trust in medical providers, the information they supply, and the medical system within which they work. Lack of trust in physicians and in the medical system by racial and ethnic minority individuals contributes heavily to health disparities in this population⁵. The mistreatment of racial and ethnic minorities throughout history by the medical system, including the Tuskegee and Guatemalan Syphilis Studies and the harvesting of Henrietta Lack’s biological material post-mortem without her family’s consent⁵, contextualizes minority individuals sometimes express toward the medical establishment. One study examining a probability sample of adults throughout the US found that the endorsement of mistrust of the medical field among SMM of racial/ethnic minority identity (e.g., African American and Latino SMM) may in part be attributed to discrimination towards sexual orientation and cultural factors⁵. Compared with their White peers, Black and Latino individuals are more likely to report feeling disrespected by their physicians, are more likely to report negative interactions with medical professionals, and do not receive as high quality of health communication⁵. In addition, Latino and Black patients are less likely to report being satisfied with their care, trust their doctor, be involved in medical decision making, or believe that their health is more important to their doctor than profit⁵. However, Latino SMM remain an understudied population within the context of medical mistrust. Latinos are frequently either not included in medical mistrust research or included under a broad “other” category⁵. The majority of existing literature examining how elevated rates of medical mistrust relate to health outcomes has primarily focused on differences between heterosexual African American and White individuals, despite Latino individuals also reporting elevated rates of medical mistrust in comparison to White individuals⁵. Furthermore, research examining the association between medical mistrust and HIV care among SMM has mainly focused on African American SMM⁶ despite Latino SMM also being disproportionately affected by HIV¹.

Research examining how medical mistrust relates to HIV care and HIV prevention among racial/ethnic SMM suggests that medical mistrust may be a risk factor for poor HIV care and prevention. Among a mixed sample of young HIV-negative SMM in California (33% Latino, 25% African American, 21% White, 21% mixed or other), high medical mistrust was associated with lower PrEP use willingness. However, no effects were revealed when examining the effect by race/ethnicity⁷. Furthermore, among a convenience sample of HIV-negative African American SMM in the southeastern US, high medical mistrust attributed to race/ethnicity has also been associated with a lower PrEP use willingness⁵. When examining the association between medical mistrust and HIV antiretroviral (ART) adherence among a peer- and clinic-recruited sample in New York City, medical mistrust has been identified as a risk factor for low ART willingness and adherence among HIV-positive African American and Latino individuals⁸. Latino SMM with high levels of medical mistrust may be less likely than White SMM to trust any information given by medical providers regarding the efficacy of PrEP and therefore may be less willing to use PrEP and likely to use or adhere to PrEP adequately.

To the best of our knowledge, no previously published studies have examined the association between medical mistrust and steps of the PrEP cascade among Latino SMM. The aim of the current study is to explore how medical mistrust relates to four stages of the PrEP cascade: PrEP awareness, willingness, use, and adherence among a sample of HIV-negative/unknown Latino SMM, a population highly susceptible to HIV infection. We hypothesized that greater medical mistrust would be associated with lower odds of PrEP awareness, willingness, use, and adherence.

Methods

From April to June of 2017, we recruited 151 Latino SM cisgender men aged 18–29 (M = 24.18 years of age, SD = 3.19) via paid advertisements on the dating websites Grindr, Squirt.org, and Scruff, as well as from a registry of participants of other studies who had expressed interests in future studies. Surveys were approximately 30 minutes in length and were available in Spanish and English. Inclusion criteria for the current study were 1) being between the ages of 18–29, 2) identifying as both Hispanic/Latino and a man, 3) identifying as either a Spanish or English speaker, 4) live in San Diego County, 5) identifying as a gay or bisexual man or as a man attracted to men, and 6) self-reporting HIV-negative/unknown status. Of the participants in the study, 71 (47%) expressed PrEP awareness, 70 (46.4%) expressed willingness to take PrEP, 23 (15.2%) reported current PrEP use, and 22 (14.6%) reported at least 60% PrEP adherence in the past thirty days. Participants provided consent and all procedures were approved by the San Diego State University Institutional Review Board.

Measures

PrEP Awareness

Awareness of PrEP was assessed by the following item: “Before this study, had you ever heard of PrEP?” Response options were: “Yes = 1,” “No = 2,” and “Not sure = 3.” Responses were dichotomized to “Yes” vs “No” or “Not sure”. A response of “yes” was coded as positive for PrEP awareness.

PrEP Willingness

Willingness to take PrEP was assessed by the following item: “How likely would you be to use PrEP?” Response options ranged from 1 = Extremely unlikely to 5 = Extremely likely. Consistent with previous research⁴, a score of 4 (Likely) or greater was indicated as positive for willingness to take PrEP.

PrEP Use

Current use of PrEP was defined as PrEP use in the last month, and was assessed by the following item: “Over the past month have you taken PrEP?” Response options were “Yes = 1” or “No = 0.” As this question was only displayed to participants who endorsed lifetime PrEP use (n = 28), the participants who were not shown the question were coded as “No.”

PrEP Adherence

Adherence to PrEP was assessed with the following item: “Thinking about the past 30 days, what percent of the time did you take all of your PrEP medications as your doctor prescribed?” Response options were on a scale from 0–100%. Consistent with previous research⁴, selection of 60% or higher was coded as being adherent to PrEP.

Medical Mistrust

Medical mistrust was measured using the Group Based Medical Mistrust Scale (GBMMS)⁹. The GBMMS is a 12-item scale that measures one’s mistrust in medical services provided by health care systems and health care providers based on one’s racial or ethnic group, with response options ranging from 1 = Strongly disagree to 5 = Strongly agree. A mean score was calculated with higher scores indicating a higher level of medical mistrust. The GBMMS has been previously validated in Latino men¹⁰ and previous research has found an internal consistency of α = .83. The internal consistency in the current study was a = .83.

Translation

In order to ensure that the study would be open to Latino SMM regardless of English fluency, the survey was made available to participants in English and Spanish. For measures that had a validated Spanish version, this version was included in the Spanish iteration of the survey. Measures without a validated Spanish version were translated by a researcher and undergraduate research assistant, both of whom are bilingual Spanish-speakers and members of the study team. Translated measures were reviewed by two bilingual community members who met study eligibility criteria in order to verify their appropriateness for use in this study.

Planned Analysis

Four separate unadjusted logistic regressions were employed with medical mistrust entered as the predictor variable and awareness of PrEP, willingness to take PrEP, current PrEP use, and PrEP adherence entered as the respective outcome variables. All logistic regressions were tested for assumptions of linearity and multicollinearity. Odds ratios (OR), their 95% confidence intervals (CIs), and Nagelkerke pseudo R² are reported. Sensitivity adjusted models were also performed with sociodemographic variables (age and sexual identity) entered as covariates and are reported as well.

Results

Of the participants included in analyses, 55.0% of the sample identified as White, 26.5% as Black, 4.0% as Native American, 2.6% as Asian or Pacific Islander, and 11.3% as another race, while 0.7% did not identify a race. For sexual identity, 68.9% identified as gay, 30.5% identified as bisexual, and 0.7% identified as pansexual. In terms of sexuality, 60.3% of participants reported attraction to only men, 12.3% were attracted mostly to men, 24.5% were attracted equally to men and women, and 2.6% were attracted mostly to women. Means and standards of deviation for age and medical mistrust, as well as frequencies and percentages for sexual identity (gay vs. bisexual), PrEP awareness, willingness, use, and adherence, are reported in Table I, in addition to the bivariate correlations.

Table I:

Descriptive Statistics and Bivariate Correlations

	Mean/Frequency	Standard Deviation/Percent	Age	GBMMS	Sexual Identity	PrEP Awareness	PrEP Willingness	Current PrEP Use	PrEP Adherence
Age	24.18	3.19	1
GBMMS	2.66	0.64	0.028	1
Sexual Identitya
Gay	104	68.9%	−0.024	0.292**	1
Bisexual	46	30.5%
PrEP Awarenessb
No	80	53.0%	−0.041	−0.431**	−0.370**	1
Yes	71	47.0%
PrEP Willingnessb
No	81	53.6%	0.035	−0.381**	−0.043	0.401**	1
Yes	70	46.4%
Current PrEP Useb
No	128	84.8%	0.179*	−0.239**	−0.203*	0.376**	0.308**	1
Yes	23	15.2%
PrEP Adherenceb
No	6	4.0%	0.347	−0.620**	−0.101	0.284	0.576**	0.438*	1
Yes	22	14.6%

^aGay = 0, Bisexual = 1; of the 151 participants, n = 1 participant identified as pansexual. As this did not fit within the binary variable used on the adjusted model (gay vs. bisexual), this individual was excluded from the adjusted analyses. This participant was retained in the unadjusted analyses.

^bNo = 0, Yes = 1

^*p < 0.05 level (2-tailed)

^**p < 0.01 level (2-tailed)

Adjusted Analyses

PrEP Awareness

There was no significant association with age. Sexual identity was associated with PrEP awareness such that bisexual individuals were less likely to be aware of PrEP than gay individuals (Wald χ²(1) = 10.611, p = 0.001, R²_Nagelkerke = 0.36, OR = 0.23 95% CI [0.10, 0.56]). Medical mistrust was significantly associated with lower odds of PrEP awareness (Wald χ²(1) = 18.426, p < 0.0001, R²_Nagelkerke = 0.36, OR = 0.19, 95% CI [0.09, 0.40]).

PrEP Willingness

There was no significant association with age or sexual identity. Medical mistrust was significantly associated with lower odds of PrEP willingness (Wald χ²(1) = 21.177, p < 0.0001, R²_Nagelkerke = 0.22, OR = 0.20, 95% CI [0.10, 0.40]).

Current PrEP Use

There was no significant association with sexual identity. Age was significantly associated with current PrEP use such that older participants were more likely to be currently using PrEP (Wald χ²(1) = 4.370, p = 0.037, R²_Nagelkerke = 0.19, OR = 1.19, 95% CI [1.01, 1.40]). Medical mistrust was significantly associated with lower odds of current PrEP use (Wald χ²(1) = 5.175, p = 0.023, R²_Nagelkerke = 0.19, OR = 0.43, 95% CI [0.21, 0.89]).

PrEP Adherence

No significant associations emerged with age or sexual identity. Medical mistrust was significantly associated with lower odds of PrEP adherence (Wald χ²(1) = 4.511, p = 0.034, R²_Nagelkerke = 0.58, OR = 0.03, 95% CI [0.00, 0.77]).

Unadjusted Analyses

Medical mistrust was significantly associated with lower odds of PrEP awareness (Wald χ²(1) = 22.877, p < 0.0001, R²_Nagelkerke = 0.25, OR = 0.19, 95% CI [0.10, 0.38]), PrEP willingness (Wald χ²(1) = 18.800, p < 0.0001, R²_Nagelkerke = 0.19, OR = 0.25, 95% CI [0.14, 0.47]), current PrEP use (Wald χ²(1) = 7.965, p = 0.005, R²_Nagelkerke = 0.09, OR = 0.38, 95% CI [0.20, 0.75]), and PrEP adherence (Wald χ²(1) = 7.863, p = 0.005, R²_Nagelkerke = 0.53, OR = 0.06, 95% CI [0.01, 0.43]).

Discussion

Based on the results of the current study, greater endorsement of beliefs consistent with medical mistrust is associated with decreased odds of outcomes along the PrEP cascade. As such, medical mistrust may present a barrier to PrEP efficacy. Medical mistrust has been studied as a barrier to HIV care, with mistrust of physicians correlating with decreased ART readiness among Black and Latino individuals⁸, and medical mistrust mediating the association between reports of discrimination and ART adherence among Latino men¹¹. Based on the findings of this study, in addition to its well documented negative effects on HIV treatment, medical mistrust may also negatively impact HIV prevention. Addressing medical mistrust among Latino SMM may be helpful in increasing PrEP awareness and engagement among this vulnerable population.

The findings of this study indicate several potential pathways for future research. Increasing awareness of PrEP among Latino SMM is vital, but interventions to increase knowledge of and engagement with PrEP may not be effective without complimentary interventions addressing factors contributing to medical mistrust among Latino. Interventions aimed at improving the quality of interactions with the medical care system for Latino SMM would also be important to examining if medical mistrust has a causal association to decreased PrEP engagement, as well as potentially promoting PrEP engagement. Developing interventions to better address the bias experienced by Latino SMM in care settings and promote a standard of care that is both culturally sensitive and identify affirming would be important next step in the current body of research. Such studies would help clarify the directionality of the association between medical mistrust and outcomes along the PrEP cascade and could be helpful in promoting PrEP uptake in this vulnerable population. In addition, future studies would do well to examine the role of PrEP persistence, through measures of regular STI/HIV testing and regular contact with the PrEP prescriber, in order to better understand PREP effectiveness.

In the field of HIV treatment, culturally informed interventions have shown promise in promoting positive treatment outcomes among marginalized populations. For instance, Rise, a program for Black men and women living with HIV designed to address cultural concerns associated with decreased medication adherence, including medical mistrust, helped participants maintain adherence¹². Related work with HIV positive Latino SMM found that medical mistrust mediated the association between perceived discrimination for both their Latino identity and HIV and ART adherence¹¹. This finding, in conjunction with those in this study, suggest that medical mistrust may also mediate the associations between ethnicity and HIV-related discrimination and healthcare decisions in other settings. Should further research bear out this hypothesis, providing a culturally adept intervention to address the underlying concerns of medical mistrust could serve to help Latino SMM access PrEP.

Similar culturally informed interventions for Latino MSM could be effective in addressing concerns related to medical mistrust, promoting PrEP uptake, and examining whether there is a causal association between higher rates of medical mistrust and decreased engagement with PrEP. In addition, interventions within the medical system which actively seek to prevent discrimination against Latino and SMM patients and promote culturally competent standards of care among non-minority medical practitioners would be an important step in addressing the root causes of medical mistrust, and potentially contributing to increased engagement along the PrEP cascade among Latino SMM. While providing culturally competent resources that supplement medical care is helpful in addressing health disparities, the onus should be equally, if not more so, placed on the medical system to address its own ingrained bias and history of discrimination.

While informative, the current study is not without its limitations. Data collected was geographically limited to the San Diego area, which limits the generalizability of the sample. In addition, due to the online-only nature of the study, HIV status could not be corroborated through testing. PrEP adherence was also measured through self-report, and therefore estimates may have been reported inaccurately, whether intentionally (e.g., social desirability bias) or not (e.g., incorrect memory). Future research could improve upon these self-report measures through technological monitoring (e.g., Medication Even Monitoring System, or MEMS Caps). While this study does provide useful information in regard to the association between medical mistrust and the PrEP cascade, the cross-sectional design prevents inferring temporal ordering. In future research, it would be prudent to also examine medical mistrust and the full PrEP cascade among African American SMM, as this group has also been shown to be at heightened risk for HIV infection¹ and previous research has indicated that medical mistrust may negatively impact their engagement with HIV care in HIV positive individuals⁸. In addition, future studies may benefit by measuring patterns of PrEP use in addition to adherence rate, ideally through technological monitoring, such as MEMS Caps, in order to better understand PrEP efficacy. Finally, future research would also benefit from measuring PrEP persistence by measuring the frequency at which PrEP users visit their provider and receive HIV/STI testing³, as PrEP persistence is an important step in assessing if individuals are able to maintain PrEP efficacy over time.

Conclusion

Considering the risk of contracting HIV that Latino SMM face, it is important to explore avenues to promote their engagement with HIV prevention, including PrEP. This study found that throughout the PrEP cascade, higher levels of medical mistrust were associated with decreased engagement with PrEP, suggesting that medical mistrust may be a barrier to PrEP knowledge and use. The negative association between medical mistrust and outcomes along the PrEP cascade suggests that it is important to consider how Latino SMM experience their interactions with medical care professionals and cultural attitudes towards the healthcare system. Further research will be needed to examine if providing culturally sensitive interventions aimed at addressing medical mistrust will improve PrEP engagement and uptake among Latino SMM.