Perceived HIV Risk,Barriers,and Preferences for HIV Testing in Structurally Vulnerable Communities in St. Louis: A Best-Worst Scaling Survey

Abstract

Background

Persistent gaps in efforts to end the HIV epidemic in the United States highlight the need for community-specific, evidence-based HIV testing strategies. This study investigated barriers and preferences around HIV testing among underserved populations, including Black, queer, and young adults in the St. Louis, Missouri area.

Methods

We conducted a Best-Worst Scaling (BWS) survey with adults (16+ years) recruited at community events, collecting responses electronically and providing a $25 incentive. The survey assessed the relative importance of 13 potential barriers to HIV testing such as stigma, structural barriers, and perceived HIV risk, chosen based on literature review and community input. Mean preference weights (MPWs) were derived from participants’ most and least selected barriers in a set of 13 choice tasks and rescaled using Hierarchical-Bayes estimation. We applied latent class conditional logit models to explore variability in preferences among participants.

Results

Among 152 participants, 53% identified as women, 39% as men, and 7% as nonbinary; 49% identified as Black, with a median age of 29 years (IQR: 23-37). Additionally, 49% identified as nonheterosexual. The most prominent barrier to testing was low perceived risk for HIV (MPW: 13.9; 95% CI: 12.5-15.2). Black participants reported lack of trust in testing organizations and the absence of peer encouragement for testing as significant barriers compared to White participants (P = .03, P = .003). Queer participants identified stigma, limited access, and fear of a positive result as more critical barriers than their heterosexual counterparts (P < .001, P = .023, P = .004). Latent class modeling identified 2 distinct groups: (1) “Low-risk perceivers” (67%) who emphasized low perceived risk of HIV (MPW: 16.7, 95% CI: 15.4-18.1) and lack of healthcare provider recommendation (11.9, 95% CI: 10.7-13.2), and (2) “Stigma avoiders” (33%) who were concerned about stigma (17.1, 95% CI: 16.0-18.3) and fear of a positive diagnosis (16.1, 95% CI: 14.5-17.8).

Conclusions

This BWS analysis provides insights into HIV risk perception and testing barriers among affected communities in the St. Louis region, revealing that low perceived HIV risk and social stigma are substantial deterrents. Tailored HIV testing initiatives should be developed to address these barriers and improve testing uptake across diverse community members.

Keywords

HIV testing implementation science stigma Best-Worst Scaling risk perception

Introduction

Significant gaps remain in the United States’ response to Ending the HIV Epidemic (EHE). As of 2022, 13% of people living with HIV (PLWH) remain undiagnosed.¹ People living with HIV who are undiagnosed or not linked to care account for approximately 80% of HIV transmissions in the United States.² Although expanded HIV testing has been integral in slowing the epidemic,^3,4 the continued national expansion of testing as recommended is unfeasible.⁵ Instead, locally targeted evidence-based testing strategies that align with community preferences and are well-integrated with the HIV care continuum is more viable.⁶ Furthermore, the heterogeneity in local epidemics complicates the selection of appropriate evidence-based testing strategies.⁷ The St. Louis Region (STL) in Missouri, a state targeted by the United States EHE initiative, illustrates this variability. In the region, Black or African American people account for a significantly larger proportion of new diagnoses compared to the national average (75% in STL vs 42% nationally), while gay, bisexual, and other men who have sex with men (GBMSM) account for a smaller proportion (50% in STL vs 69% nationally).^4,8 Current evidence demonstrates that the successful implementation of any given testing strategy depends on the local HIV system of care.⁹

Multiple evidence-based strategies exist for implementing more targeted HIV testing and overcoming various barriers.^10,11 These strategies differ by locus (e.g., clinical or nonclinical settings), timing (e.g., event-based testing or appointment testing), identification mechanisms (e.g., through peer networks or universal testing), and the individual conducting the test (e.g., clinician, community worker, or self-testing). Given the variability in local communities’ epidemiology and HIV service providers, the evidence supporting HIV testing approaches is not universally generalizable to specific locales. Emerging evidence shows varied preferences for testing among different target populations. Earlier studies have found that young adults strongly preferred HIV testing when combined with other STI screenings,¹² while GBMSM favored clinical locations, and some minorities preferred self-testing.¹³ A national study also found that GBMSM with lower education levels preferred couples’ counseling and testing,¹⁴ while others have reported that highly educated young Black GBMSM in Washington, DC, favored testing at medical facilities and schools.¹⁵ Transgender women in the US Southeast preferred at-home HIV and STI testing due to privacy concerns and the ability to avoid potential negative interactions with healthcare staff and providers.¹⁶ These findings underscore that a one-size-fits-all testing strategy is unlikely to be effective or generalizable across different populations and settings.

Overcoming barriers to HIV testing in regions with a disproportionate disease burden such as St. Louis requires strategies that align with the preferences of specific target communities, are conveniently located, and can be easily integrated with other aspects of the HIV care continuum. This paper reports findings from a cross-sectional study that utilized a Best-Worst Scaling (BWS) survey to understand perceived risk, barriers, and preferences for HIV testing among communities facing a disproportionate HIV burden in the STL. This study was conducted as part of the Clear Path Collaborative which brings together service providers, organizations, and individuals from affected communities in the STL to improve HIV care systems and increase access to HIV and STI testing for those who are often excluded.

Methods

A growing approach in public health for understanding individual priorities, barriers, and preferences is BWS surveys.¹⁷ BWS surveys ask participants to select the best and the worst option from a set of 3 or more options,¹⁸ ultimately allowing for a quantifiable, relative ranking of the presented options.¹⁷ This survey method provides the advantages of a patient-centered approach to investigating preferences in health care and has a wide range of applications for policy, clinical practice, and public health.^17,18 When applied to HIV testing, this approach can help evaluate both underlying beliefs (e.g., perceived risk) about HIV testing and preferences for testing characteristics (e.g., location, strategy) within a specific community.

We administered a BWS survey to residents of the STL (city and county) from February 20, 2024, to September 20, 2024, to investigate preferences for, and underlying beliefs about HIV testing among communities facing a disproportionate HIV burden in the region. The study protocol and procedures of informed consent was reviewed and approved by the Institutional Review board of Washington University in St. Louis (IRB ID: 202305027). The IRB approved a waiver of the requirement to obtain informed consent under 45 CFR 46.116(f)(1). For minors included in the study (16-17-year-olds), we received a waiver of parental permission in accordance with 45 CFR46.408 (c), as the study posed minimal risk to children and did not involve sensitive information or information that would not necessarily be disclosed to parents. The Strengthening the Reporting of Observational Studies in Epidemiology checklist for cross-sectional studies was used as a guide to report the results and develop this manuscript¹⁹ (Supplemental File 1).

Participants and Recruitment

Our target population was structurally vulnerable adults in the STL, with particular focus on individuals who identify as Black, queer, or younger adults. These groups experience a disproportionate HIV burden and persistent barriers to HIV testing. Eligibility criteria were intentionally broad to allow participation from adults across the lifespan. Residents of St. Louis City or St. Louis County who were 16 years of age or older were eligible. This broad eligibility allowed us to oversample priority groups through targeted recruitment while still including the wider community to understand how experiences and preferences related to HIV testing may differ across the lifespan.

We collected data electronically using computer tablets at in-person events or provided participants with an anonymous online link to complete the survey later. Participants received a $25 USD gift card as an incentive for participating in the study. We recruited participants in 2 ways: through local organizations that provide sexual health care services to individuals within the target population and at public and community events dedicated to ethnic and sexual minorities, including Juneteenth celebrations and PRIDE parades. Before participating, individuals had opportunities to ask questions. All participants provided online consent at the start of the survey and had the option to download a copy of their completed survey responses for their records.

Sample Size Considerations

No formal power calculation exists for best worst scaling combined with latent class analysis. We followed published methodological guidance indicating that samples of approximately 100 to 300 participants typically provide stable estimation of preference weights and reliable identification of latent classes.^17,20 Based on recommendations from Louviere et al,²⁰ and Cheung et al,¹⁷ we targeted a sample size of approximately 150 participants, which falls within established norms for BWS studies.

Survey Design

The survey took approximately 15 minutes to complete electronically. Participants provided information on demographics, basic needs, HIV testing history, medical care, and health insurance status. Surveys were conducted in REDCap (Research Electronic Data Capture),²¹ a secure, web-based platform. The BWS portion of the survey was designed and conducted using Lighthouse Studio Sawtooth Software (version 9.15.6).²²

We searched the literature for empirical studies and systematic reviews of HIV testing barriers and facilitators within the United States with attention to articles focused on marginalized populations (e.g., Black, queer, young).^10,23–26 Based on our review, we created a table of common barriers and facilitators across studies which we organized into 3 groups: risk perception and external motivation (e.g., low HIV risk perception, knowledge of HIV and testing options), stigma and psychosocial barriers (e.g., stigma and discrimination, patient/provider trust), and structural barriers (e.g., service deserts, cost). We consulted HIV service professionals working with marginalized communities to develop BWS survey items based on identified barriers and facilitators. After initial feedback, we piloted the BWS items with additional professionals and refined the survey accordingly.

The BWS items used in the study included 13 potential barriers or concerns related to HIV testing. The full list of barriers is in Supplemental Table 1. For the BWS survey, participants completed a series of 13 of questions or “choice tasks,” from which participants were presented with 4 of the 13 barriers and selected which one was the most and which was the least influential in their personal decisions about whether to get an HIV test (Supplemental Figure 1). Lighthouse Studio Sawtooth Software (version 9.15.6) was used to generate 300 versions of the survey using a balanced incomplete block design to ensure barriers were presented an equal number of times for each participant and that participants received different random combinations of barriers in each question. The survey instrument with a sample of the BWS questions is attached as Supplemental File 2.

Statistical Analyses

For overall survey findings, we report summary statistics proportions of demographics, HIV testing history, medical care history, and perceptions of HIV risk across all survey participants. We restricted the BWS analysis to participants that passed a data quality check designed to identify those selecting BWS choice tasks at random or completing the survey too quickly to have considered the questions carefully. For the quality check, we calculated the commonly used root likelihood (RLH) fit statistic for each participant which measures how well a model predicts the participant's choice.²⁷ To test for participants who responded to the BWS choice tasks completely at random, we used a simulation of 500 computer-generated responses to determine an RLH cutoff from those expected of random respondents. We then excluded participants who (1) fell below the 80th percentile of the RLH cutoff or (2) sped through the BWS (<40% of the median BWS time across participants; ie, completed the survey in 2 min or less) and fell below the 95th percentile of the RLH.²⁸

Among those who passed the quality check, we conducted a count analysis, calculating BWS scores according to the frequency with which each barrier was selected as the most or least important, relative to the total number of times the item appeared in the survey. We then used a hierarchical Bayes model to generate rescaled mean preference weights (MPWs). Rescaled MPWs represent a share of preference out of 100, where a higher MPW represents higher relative importance of the barrier and a lower score indicates lower relative importance (out of 100). This rescaling allows for a direct quantitative comparison of the importance of HIV testing barriers, where for example, an MPW of 8.0 is twice as important as an MPW of 4.0.

HIV testing barriers fell into one of 3 mutually exclusive categories: (1) Structural barriers related to cost and access of testing, (2) Risk perception and external motivation barriers related to a participant's personal health beliefs and interactions with healthcare providers, and (3) Psychosocial barriers related to internal and external stigma and psychosocial factors.

We explored heterogeneity in preferences among participants to determine if there were differences in the importance of HIV testing barriers in specific subgroups that could help inform future, targeted testing strategies. To do this, we compared MPWs across participant characteristics using ANOVA (analysis of variance) tests.²⁹ Demographic characteristics were collapsed into fewer categories where appropriate for purposes of subgroup analysis. To better understand potential differences in preferences across the intersection of race and gender, we also examined preferences across these subgroups. To further explore heterogeneity, we fit latent class conditional logit models with up to 5 classes. We estimated the probability of a participant belonging to each class and used fit statistic criteria, including Akaike information criterion and Bayesian information criterion, and qualitative comparison to choose a class model with the best fit. We examined demographic characteristics associated with latent class membership using Wald tests. All analyses were conducted in Stata 16.1 and Lighthouse Studio Sawtooth Software (version 9.15.6).

Results

Participant Characteristics

Among those surveyed, 129 (84.5%) met data quality checks for inclusion in the BWS analysis. Those who were excluded due to data quality checks all had a low RLH (<80%) fit statistic, indicating answering BWS questions at random (n = 23). After comparing, these participants did not differ significantly in characteristics from those that did not answer the BWS at random, and therefore we still report their survey responses prior to completing the BWS to understand their health and HIV testing history (Supplemental Table 7).

Among all 152 participants, about half identified as women (52%), and half identified as black or African American (49%). The median age was 29 (IQR 23-37) years. Most participants had attended at least some college or postsecondary school (65%). The majority identified as single (61%). Half of the participants identified as straight or heterosexual (50%), 23% identified as gay or lesbian, and 20% identified as bisexual. More than half (54%) of participants indicated running out of money for basic needs over the past 6 months. More details about participants can be found in Table 1.

Table 1.

Full Sample Participant Demographic Characteristics.

	N	(%)
Age
17-24	40	(26.3)
25-34	59	(38.8)
35-44	22	(14.5)
45+	14	(9.2)
Missing	17	(11.2)
Gender identity
Man	59	(38.8)
Woman	80	(52.6)
Nonbinary or third gender	10	(6.6)
Prefer not to say	3	(2.0)
Transgender
Yes	9	(5.9)
No	134	(88.2)
Prefer not to answer	9	(5.9)
Education
Less than high school	10	(6.6)
High school graduate or GED	39	(25.7)
Some college/Associate's/vocational	53	(34.9)
Bachelor's degree	28	(18.4)
Graduate or professional degree	18	(11.8)
Prefer not to answer	4	(2.7)
Race
White	60	(39.5)
Black or African American	75	(49.3)
American Indian or Alaska Native	1	(0.7)
Asian	4	(2.6)
Native Hawaiian or Other PI	5	(3.3)
Other	3	(2.0)
Prefer not to answer	4	(2.6)
Hispanic
Yes	3	(2.0)
No	143	(94.1)
Prefer not to answer	6	(3.9)
Relationship
Single	93	(61.2)
In a relationship	34	(22.4)
Married or domestic partnership	17	(11.2)
Widowed/Divorced/Separated	7	(4.6)
Prefer not to answer	1	(0.7)
Sexual orientation
Straight or heterosexual	77	(50.6)
Queer/gay/questioning/bi/pan	74	(48.7)
Prefer not to say	1	(0.7)
Basic needs
Did not run out of money	62	(40.8)
Less than once a month	14	(9.2)
Once a month	23	(15.1)
2-3 times a month	20	(13.2)
Once a week	8	(5.3)
Many times a week	18	(11.8)
Prefer not to answer	7	(4.6)
Total	152	(100%)

Medical Care and Health Insurance status

Only 56% of participants had a primary care provider. Forty-one percent indicated that they usually received their medical care from a medical doctor's office, another 23% indicated they usually received care at urgent care services, and 15% at a community clinic. Most (62%) participants reported having health insurance for the entirety of the past year. A third of participants (33%) reported experiencing discrimination while receiving medical care once in the past year. More details can be found in Table 2.

Table 2.

Participants’ Medical Care and Health Insurance Status.

	n	(%)
Primary Care Provider
Yes	86	(56.6)
No	61	(40.1)
Prefer not to answer	5	(3.3)
Primary care location
Private medical doctor's office	67	(44.1)
Emergency department	7	(4.6)
Community clinic	23	(15.1)
Urgent care	35	(23.0)
Pharmacy	3	(2.0)
Other	7	(4.6)
Prefer not to say	10	(6.6)
Health insurance in past 12 months
No	34	(22.4)
Yes—health insurance part of the year	23	(15.1)
Yes—health insurance the whole year	95	(62.5)
Experienced discrimination at medical care
None	92	(60.5)
Once	29	(19.1)
2-3 times	19	(12.5)
4 or more times	2	(1.3)
I haven't gotten medical care.	9	(5.9)
Missing	1	(0.7)

History of HIV Testing and Perceptions of HIV Risk

More than a quarter of participants (27%) reported never having received an HIV test in their lifetime. Only 11% indicated feeling the need to get tested for HIV. About 72% of participants had been tested for HIV at least once, with nearly half (49%) having been tested within the past year. Among those who had been tested, most (49%) received their most recent HIV test at a community health clinic or free clinic, while 26% were tested at a private doctor's office, and 7% at a hospital. When asked about their perceived likelihood of contracting HIV in their lifetime, approximately half (51%) felt they were not at all likely to contract HIV. When asked about their level of concern regarding contracting HIV, the majority (57%) reported being not at all worried about contracting HIV. More details about HIV testing and perceptions of risk can be found in Table 3.

Table 3.

Participants’ History of HIV Testing and Perceptions of HIV Risk (N = 152).

	n	(%)
Do you feel that you need an HIV test?
Yes	18	(11.8)
No	119	(78.3)
Not sure	15	(9.9)
When was the last time you had an HIV test?
I've never received an HIV test	41	(27.0)
Less than 6 months ago	63	(41.4)
6 months to a year ago	12	(7.9)
More than a year ago	35	(23.0)
Missing	1	(0.7)
Last HIV test location
Community health/free clinic	54	(35.5)
Private doctor's office	29	(19.1)
Mobile unit	4	(2.6)
Hospital	12	(7.9)
I got a home test and took it myself	6	(3.9)
Other	5	(3.3)
Missing	42	(27.6)
How likely do you think you are to get HIV in your lifetime?
Not at all likely	77	(50.7)
Slightly likely	35	(23.0)
Moderately likely	13	(8.6)
Very likely	2	(1.3)
Extremely likely	3	(2.0)
This does not apply to me	22	(14.5)
How worried are you about getting HIV in your lifetime?
Not at all worried	87	(57.2)
Slightly worried	26	(17.1)
Moderately worried	16	(10.5)
Very worried	2	(1.3)
Extremely worried	5	(3.3)
This does not apply to me	16	(10.5)

Overall Best Worst Scaling Scores

Participants identified the factors that mattered most and least when deciding to get an HIV test (n = 129 after data quality checks). Across all barriers to HIV testing, MPWs ranged from 4.6 to 13.9, and the highest (most important) were factors related to risk perception and external motivation. Specifically, the most important barriers were of participants who did not think they were at risk for HIV (MPW = 13.9) and those who had not received a testing recommendation from a medical professional (MPW = 9.9) (Figure 1). Psychosocial barriers such as fear of stigma if testing positive (MPW = 9.7), and fear of being HIV positive (MPW = 9.2) were also highly rated. Among structural barriers, the most significant factors included HIV testing not being a priority (MPW = 8.6), not knowing what to do next if they tested for HIV (MPW = 7.9), and not having access to testing (MPW = 7.9). Less relevant barriers (ie, lowest MPWs) for HIV testing decisions included participants being worried that people close to them will find out they took a test, participants not being able to trust the person or organization doing the test,and participants feeling ashamed or embarrassed if they took a test. MPWs for HIV testing barriers are detailed in Figure 1.

Figure 1.

Best-Worst Scaling scores: relative importance of barriers to HIV testing. CI, confidence interval.

Differences in Preferences Across Subgroups

Analysis of variance tests revealed some significant differences in HIV testing barriers across demographic groups, highlighted below (Supplemental Table 3). Figure 2 displays preferences disaggregated by race, sexual orientation, and gender identity, with statistically significance (P ≤ .05) indicated in asterisks. Black participants placed greater importance than White participants on lack of trust in testing organizations (MPWs of 5.8 vs 4.3, P = .03) and not receiving peer recommendations (MPWs of 7.8 vs 5.4, P ≤ .01). In contrast, White participants rated stigma of testing positive for HIV higher than Black participants (MPWs of 11.6 vs 7.8, P = .02) along with shame around testing (MPWs of 5.7 vs 3.8, P = .04) (Figure 2 panel A, Supplemental Table 3). Heterosexual participants were more likely to emphasize not feeling at risk for HIV (MPW = 16.5, P < .001) and lack of peer encouragement to get tested (MPW = 8.5, P < .001) compared to queer participants (MPW = 10.8 and 4.9). Conversely, queer participants rated stigma (MPW = 11.8, P < .01) and fear of a positive result (MPW = 11.1, P < .01) higher than heterosexual participants (MPW = 7.8 and 7.6) (Figure 2 panel B, Supplemental Table 3). Women more often cited not knowing the next steps after testing than men (MPWs of 8.4 vs 6.7, P = .03) and lack of peer suggestions (MPWs of 7.8 vs 6.2, P < .01). Men, however, placed more emphasis on stigma than women (MPWs of 10.8 vs 8.1, P < .01) and on fear of others finding out (MPWs of 5.7 vs 3.9, P < .01) (Figure 2 panel C, Supplemental Table 3). Participants with less than a high school education (MPW = 11.1), some college or a vocational degree (MPW = 8.5), or a bachelor's degree (MPW = 8.7) identified not knowing what to do after testing positive as a key barrier (P = .02) (Supplemental Table 3).

Figure 2.

Mean preferences weights by subgroup. (A) By race; (B) By sexual orientation; (C) By gender.

Groups Based on Preference Profiles – Latent Class Analysis

While the ANOVA analyses demonstrated differences based on participant characteristics, latent class analysis revealed 2 distinct groups of participants based on their responses to the BWS exercise. The first group, “Low-Risk Perceivers” (67%), was primarily influenced by factors related to risk perception and external motivation. For these participants, the most important barriers were not feeling at risk for HIV (MPW = 16.7) and not having a healthcare provider recommend testing (MPW = 11.9). The second group, “Stigma Avoiders” (33%), was more affected by psychosocial concerns, particularly fear of stigma if testing positive (MPW = 17.1) and fear of being HIV positive (MPW = 16.1; Figure 3, Supplemental Table 4).

Figure 3.

Latent class analysis of relative importance of barriers to HIV testing. (A) Relative ranking of barriers between groups, with barrier category (ladder plot); (B) Mean preference weight (MPW) (95% CI) across groups (forest plot).

Discussion

In this study, we used a BWS survey and latent class analysis to examine perceived risks and barriers to HIV testing among communities disproportionately affected by HIV in the STL. Our findings indicate that the most prominent barrier to testing was low perceived risk for HIV, followed by healthcare workers not recommending testing, and fear of stigma associated with an HIV-positive result. While demographic characteristics (e.g., race or sexual orientation) are often used as proxies for testing barriers, our latent class analysis suggests that different types of barriers, structural, social, and psychological, are more likely to co-occur within and across demographic groups. Our findings identified 2 distinct latent class groups with unique testing barriers and perceived risks. The first group, “low-risk perceivers,” generally believed they were at minimal risk for HIV, leading to limited motivation for testing, often coupled with the absence of testing recommendations from healthcare providers. The second group, “stigma avoiders,” made up about one-third of the participants and reported heightened HIV risk perceptions but faced significant concerns around stigma and fear of receiving a positive test result. These findings demonstrate that even within a relatively small community that different barriers figure more prominently for different groups of the community. Lastly, these findings underscore the importance of tailored HIV testing strategies that account for the unique needs of specific subgroups,⁹ especially in high-burden regions like St. Louis.

Perceived risk plays a fundamental role in health behavior change models,^30–32 and previous research has shown that enhancing risk perception can lead to increased health-related behaviors.³³ However, our results suggest that increasing HIV risk perception alone may be insufficient in this context. For instance, “stigma avoiders” indicated stigma and fear of a positive diagnosis were about twice as important as their perceived risk of HIV in influencing their decision to get tested. Additional subgroup analysis showed that Black participants often cited distrust in testing organizations as a barrier, while queer participants reported stigma as a central concern. Gender differences also emerged, with women emphasizing the importance of someone like them recommending HIV testing, whereas men reported more concerns about stigma and the fear of disclosure. These findings strongly support the push for community-specific, tailored strategies that address the unique needs and concerns of local populations.⁶

Our study has several implications for enhancing HIV testing strategies. First, targeted testing interventions need to address the distinct barriers faced by different groups. For “stigma avoiders,” education campaigns should aim to reduce stigma and provide resources on treatment and mental health to help address fears related to a positive result.^34,35 Scholarship on intersectional stigma, or the interrelated nature of various forms of stigma and discrimination (e.g., HIV stigma, racism, homophobia) has emphasized the importance of addressing stigma at multiple levels including stigma and discrimination by providers, and within healthcare spaces and policies. In contrast, “low-risk perceivers” would benefit from education that emphasizes the importance of routine HIV testing, even for those who consider themselves at minimal risk. Healthcare providers should also be encouraged to recommend testing more broadly, and nonrisk-based opt-out testing policies could be more widely adopted in healthcare settings where feasible.

Addressing structural barriers to testing is also essential for improving access. Young adults in our study reported challenges in finding testing locations and understanding next steps after a positive result, suggesting that targeted advertising could help inform them of available testing options. Moreover, to address existing structural barriers with a paucity of testing services available in predominantly Black economically divested areas of the city (i.e., North County), more modes of testing in these geographic areas are desperately needed.³⁶ Additionally, most participants who had recently tested for HIV did so at community health centers or free clinics, indicating a preference for accessible, no-cost testing services. Although many participants reported a private doctor's office as their primary care location, only a fifth received their last HIV test in such a setting. This discrepancy could stem from limited provider recommendations for testing, and psychosocial barriers, such as fear of stigma, shame, or lack of trust in providers which are common among marginalized groups. Indeed, about a third of participants reported experiencing discrimination in healthcare, which may further deter them from seeking testing in traditional healthcare settings.^10,25,26

Several evidence-based strategies have shown promise for increasing testing uptake in similar contexts. Nonrisk-based opt-out HIV testing in clinical settings has been effective in increasing acceptance rates and testing uptake,^37,38 as patients may feel less singled out when testing is routine. There is also much evidence indicating that expanding HIV testing access through colocated services and nontraditional settings can help reduce barriers and increase testing uptake.^11,39 By offering HIV testing alongside services such as sexual health, mental health, and dental care, testing becomes a routine, normalized part of healthcare, potentially reducing stigma and improving convenience. Additionally, providing testing in community-based, nonmedical locations such as libraries, places of worship, and community events, makes testing more accessible to those who may not visit traditional healthcare settings.¹¹ These strategies bring testing closer to communities, enhancing accessibility and promoting earlier diagnosis and care, especially for marginalized populations who face racism, homonegativity, and other interwoven oppressions which exacerbate HIV risks and barriers to healthcare.

Limitations

Our study findings should be considered within the context of certain limitations. First, we did not collect data on participants’ HIV status. Consequently, responses reflect testing preferences rather than experiences with an HIV diagnosis, although it is possible that PLWH were involved in our study population. Additionally, limitations inherent to the BWS method impose certain constraints on our results. Best-Worst Scaling prioritizes attributes solely among those included in the task, constraining our findings to the selected attributes and excluding any factors not addressed in the BWS questions. Despite conducting a thorough literature search and gathering feedback from community partners, the list of attribute statements we used was not exhaustive. However, attribute development was an iterative process, incorporating feedback from multiple individuals to confirm the relevance and clarity of the statements. Additionally, the questionnaire used was not validated, which may affect the reliability of the findings. Lastly, selection bias is an important limitation of this study. Recruitment at community events and through sexual health organizations may have attracted individuals who are more connected to community resources, more engaged in health-seeking behaviors, or more comfortable disclosing aspects of identity. Adults who do not participate in such events or who avoid identity-affirming spaces due to stigma may experience different barriers to HIV testing. This limits generalizability to the broader St. Louis population. Also, the small sample size for transgender participants limits statistical power. While we report findings from these observations, direct comparisons to better-represented gender groups should not be interpreted as statistically significant differences.

Conclusions

Our study findings highlight the need for diverse, contextually sensitive HIV testing approaches to address barriers specific to different community subgroups in the STL. Tailoring interventions to the unique needs of “stigma avoiders” and “low-risk perceivers” could enhance testing uptake and reduce HIV-related stigma. Opt-out testing strategies and self-testing options offer promising avenues for increasing testing rates, especially among groups with low perceived risk or concerns about privacy. These findings contribute to a growing body of evidence supporting the importance of community-informed, flexible testing approaches to help achieve national goals for EHE.

Supplemental Material

sj-docx-1-jia-10.1177_23259582261425288 - Supplemental material for Perceived HIV Risk, Barriers, and Preferences for HIV Testing in Structurally Vulnerable Communities in St. Louis: A Best-Worst Scaling Survey

Supplemental material, sj-docx-1-jia-10.1177_23259582261425288 for Perceived HIV Risk, Barriers, and Preferences for HIV Testing in Structurally Vulnerable Communities in St. Louis: A Best-Worst Scaling Survey by Emmanuel K Tetteh, Noelle Le Tourneau, Gregory Gross, McKenzie Swan, Tyrell Manning, Justin Cole, Katie Wolf, Lawrence Hudson-Lewis, Julia D López, Todd Combs and Virginia R McKay in Journal of the International Association of Providers of AIDS Care (JIAPAC)

Supplemental Material

sj-pdf-2-jia-10.1177_23259582261425288 - Supplemental material for Perceived HIV Risk, Barriers, and Preferences for HIV Testing in Structurally Vulnerable Communities in St. Louis: A Best-Worst Scaling Survey

Supplemental material, sj-pdf-2-jia-10.1177_23259582261425288 for Perceived HIV Risk, Barriers, and Preferences for HIV Testing in Structurally Vulnerable Communities in St. Louis: A Best-Worst Scaling Survey by Emmanuel K Tetteh, Noelle Le Tourneau, Gregory Gross, McKenzie Swan, Tyrell Manning, Justin Cole, Katie Wolf, Lawrence Hudson-Lewis, Julia D López, Todd Combs and Virginia R McKay in Journal of the International Association of Providers of AIDS Care (JIAPAC)

Supplemental Material

sj-docx-3-jia-10.1177_23259582261425288 - Supplemental material for Perceived HIV Risk, Barriers, and Preferences for HIV Testing in Structurally Vulnerable Communities in St. Louis: A Best-Worst Scaling Survey

Supplemental material, sj-docx-3-jia-10.1177_23259582261425288 for Perceived HIV Risk, Barriers, and Preferences for HIV Testing in Structurally Vulnerable Communities in St. Louis: A Best-Worst Scaling Survey by Emmanuel K Tetteh, Noelle Le Tourneau, Gregory Gross, McKenzie Swan, Tyrell Manning, Justin Cole, Katie Wolf, Lawrence Hudson-Lewis, Julia D López, Todd Combs and Virginia R McKay in Journal of the International Association of Providers of AIDS Care (JIAPAC)

Footnotes

Acknowledgments

The authors would like to acknowledge members of the Clear Path Collaborative for their support on this work, and Drs. Gifty Aboagye-Mensah, Nwaliweaku Anidi, and Harry Adu Obeng for their assistance with data collection.

Ethical Approval

The study protocol and procedures of informed consent was reviewed and approved by Institutional Review board of Washington University in St. Louis (IRB ID: 202305027). The IRB approved a waiver of the requirement to obtain informed consent under 45 CFR 46.116(f)(1). For minors included in the study (16-17-year-olds), we received a waiver of parental permission in accordance with 45 CFR46.408 (c), as the study posed minimal risk to children and did not involve sensitive information or information that would not necessarily be disclosed to parents.

Consent to Participate

The IRB approved a waiver of the requirement to obtain informed consent under 45 CFR 46.116(f)(1). For minors included in the study (16-17-year-olds), we received a waiver of parental permission in accordance with 45 CFR46.408 (c), as the study posed minimal risk to children and did not involve sensitive information or information that would not necessarily be disclosed to parents.

Author Contributions

Virginia R. McKay and Gregory Gross conceptualized the study and Virginia R. McKay obtained funding. Emmanuel K. Tetteh, McKenzie Swan, Virginia R. McKay, and Gregory Gross conducted the research and collected data. Noelle Le Tourneau performed the formal analysis of the study data. Emmanuel K. Tetteh and Noelle Le Tourneau drafted the initial version of the manuscript. All authors (Emmanuel K. Tetteh, Noelle Le Tourneau, McKenzie Swan, Gregory Gross, Justin Cole, Tyrell Manning, Katie Wolf, Lawrence Hudson-Lewis, Julia D. López, Todd Combs, Virginia R. McKay) reviewed, provided feedback, and approved the manuscript before submission for publication.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Funding for the study was provided by the Opportunity Fund from the Missouri Foundation for Health (grant number n/a), Washington University Institute of Clinical and Translational Sciences Grant (UL1TR002345), and the Midwest Developmental Center for AIDS Research Grant (P30AI176532).

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

ORCID iDs

Emmanuel K Tetteh

Noelle Le Tourneau

Julia D López

Todd Combs

Supplemental Material

Supplemental material for this article is available online.

References

CDC. HIV surveillance report: Diagnoses, deaths, and prevalence of HIV in the United States and 6 territories and freely associated states, 2022. Accessed February 7, 2025. https://stacks.cdc.gov/view/cdc/156509

Purcell

Sansom

Hayes

Hall

. Vital signs: HIV transmission along the continuum of care—United States, 2016. MMWR Morb Mortal Wkly Rep. 2019;68(11):267-272. doi:10.15585/mmwr.mm6811e1

Cooley

Wejnert

Rose

Paz-Bailey

; National HIV Behavioral Surveillance Study Group. Increases in recent HIV testing among men who have sex with men coincide with the centers for disease control and prevention’s expanded testing initiative. Clin Infect Dis. 2015;60(3):483-485. doi:10.1093/cid/ciu851

Pitasi

. HIV Testing in 50 local jurisdictions accounting for the majority of new HIV diagnoses and seven states with disproportionate occurrence of HIV in rural areas, 2016–2017. MMWR Morb Mortal Wkly Rep. 2019;68(25):561–567. doi:10.15585/mmwr.mm6825a2

Granich

Gilks

Dye

De Cock

Williams

. Universal voluntary HIV testing with immediate antiretroviral therapy as a strategy for elimination of HIV transmission: a mathematical model. Lancet. 2009;373(9657):48-57. doi:10.1016/S0140-6736(08)61697-9

Geng

Holmes

Moshabela

Sikazwe

Petersen

. Personalized public health: an implementation research agenda for the HIV response and beyond. PLoS Med 2019;16(12):e1003020. doi:10.1371/journal.pmed.1003020

Zang

Mah

Quan

AML

, et al. Human immunodeficiency virus transmission by HIV risk group and along the HIV care continuum: a contrast of 6 US cities. J Acquir Immune Defic Syndr. 2022;89(2):143-150. doi:10.1097/QAI.0000000000002844

Missouri Department of Health and Senior Services D of C and PH. Epidemiologic Profiles of HIV, STD, and Hepatitis in Missouri. 2018, Accessed November 14, 2024. http://archive.org/details/2018MOHIVSTDEpiProfiles

McKay

Cambey

Combs

Stubbs

Pichon

Gaur

. Using a modeling-based approach to assess and optimize HIV linkage to care services. AIDS Behav 2020;25(3):886. doi:10.1007/s10461-020-03051-5

10.

Zanoni

Elliott

Neilan

Haberer

. Screening for HIV and linkage to care in adolescents: insights from a systematic review of recent interventions in high- versus low- and middle-income settings. Adolesc Health Med Ther. 2018;9(211):235. doi:10.2147/AHMT.S153204

11.

Delaney

DiNenno

. HIV Testing strategies for health departments to end the epidemic in the U.S. Am J Prev Med 2021;61(5):S6-S15. doi:10.1016/j.amepre.2021.06.002

12.

Llewellyn

Sakal

Lagarde

Pollard

Miners

. Testing for sexually transmitted infections among students: a discrete choice experiment of service preferences. BMJ Open. 2013;3(10):e003240. doi:10.1136/bmjopen-2013-003240

13.

Miners

Nadarzynski

Witzel

, et al. Preferences for HIV testing services among men who have sex with men in the UK: a discrete choice experiment. PLoS Med. 2019;16(4):e1002779. doi:10.1371/journal.pmed.1002779

14.

Sharma

Stephenson

White

Sullivan

. Acceptability and intended usage preferences for six HIV testing options among internet-using men who have sex with men. SpringerPlus. 2014;3(1):109. doi:10.1186/2193-1801-3-109

15.

Parchem

Molock

. HIV testing preferences, barriers and facilitators to receiving HIV services among young black sexual minority men. AIDS Care. 2022;34(7):839-846. doi:10.1080/09540121.2021.1975627

16.

Van Gerwen

Austin

Bethune

Sullivan

Muzny

. Sexual healthcare and at-home STI test collection: attitudes and preferences of transgender women in the Southeastern United States. Front Public Health. 2023;11:1187206. doi:10.3389/fpubh.2023.1187206

17.

Cheung

Wijnen

BFM

Hollin

, et al. Using best–worst scaling to investigate preferences in health care. Pharmacoeconomics. 2016;34(12):1195-1209. doi:10.1007/s40273-016-0429-5

18.

Mühlbacher

Zweifel

Kaczynski

Johnson

. Experimental measurement of preferences in health care using Best-Worst Scaling (BWS): theoretical and statistical issues. Health Econ Rev. 2016;6(1):5. doi:10.1186/s13561-015-0077-z

19.

von Elm

Altman

Egger

, et al. The Strengthening The Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet. 2007;370(9596):1453-1457. doi:10.1016/S0140-6736(07)61602-X

20.

Louviere

Flynn

Marley

AAJ.

Best-Worst Scaling: Theory, Methods and Applications. Cambridge University Press; 2015. doi:10.1017/CBO9781107337855

21.

Harris

Taylor

Minor

, et al. The REDCap consortium: building an international community of software platform partners. J Biomed Inform 2019;95:103208. doi:10.1016/j.jbi.2019.103208

22.

Lighthouse Studio. Sawtooth Software version 9.15.6. Accessed December 16, 2024. https://sawtoothsoftware.com/lighthouse-studio

23.

Bagchi

Davis

. Clinician barriers and facilitators to routine HIV testing: a systematic review of the literature. J Int Assoc Provid AIDS Care 2020;19:2325958220936014. doi:10.1177/2325958220936014

24.

Leblanc

Flores

Barroso

. Facilitators and barriers to HIV screening: a qualitative meta-synthesis. Qual Health Res. 2016;26(3):294-306. doi:10.1177/1049732315616624

25.

Levy

Wilton

Phillips

, et al. Understanding structural barriers to accessing HIV testing and prevention services among black men who have sex with men (BMSM) in the United States. AIDS Behav. 2014;18(5):972-996. doi:10.1007/s10461-014-0719-x

26.

Tan

Black

. A systematic review of health care provider-perceived barriers and facilitators to routine HIV testing in primary care settings in the southeastern United States. J Assoc Nurses AIDS Care. 2018;29(3):357-370. doi:10.1016/j.jana.2017.12.006

27.

Jonker

Roudijk

Maas

. The sensitivity and specificity of repeated and dominant choice tasks in discrete choice experiments. Value Health 2022;25(8):1381-1389. doi:10.1016/j.jval.2022.01.015

28.

Orme

. Consistency cutoffs to identify “Bad” respondents in CBC, ACBC, and MaxDiff.

29.

Kim

. Statistical notes for clinical researchers: post-hoc multiple comparisons. Restor Dent Endod. 2015;40(2):172-176. doi:10.5395/rde.2015.40.2.172

30.

Rogers

. A protection motivation theory of fear appeals and attitude change. J Psychol 1975;91(1):93-114. doi:10.1080/00223980.1975.9915803

31.

Noar

Zimmerman

. Health behavior theory and cumulative knowledge regarding health behaviors: are we moving in the right direction? Health Educ Res 2005;20(3):275-290. doi:10.1093/her/cyg113

32.

Ferrer

Klein

. Risk perceptions and health behavior. Curr Opin Psychol 2015;5:85. doi:10.1016/j.copsyc.2015.03.012

33.

Sheeran

Harris

Epton

. Does heightening risk appraisals change people’s intentions and behavior? A meta-analysis of experimental studies. Psychol Bull. 2014;140(2):511-543. doi:10.1037/a0033065

34.

Lutete

Matthews

Sabounchi

, et al. Intersectional stigma and prevention among gay, bisexual, and same gender-loving men in New York city, 2020: system dynamics models. Am J Public Health. 2022;112(S4):S444-S451. doi:10.2105/AJPH.2022.306725

35.

Stangl

Atkins

Leddy

, et al. What do we know about interventions to reduce intersectional stigma and discrimination in the context of HIV? A systematic review. Stigma and Health. 2023;8(3):393-408. doi:10.1037/sah0000414

36.

For the Sake of All: A report on the health and well-being of African Americans in St. Louis and why it matters for everyone | Health Equity Works | Washington University in St. Louis. Accessed February 7, 2025. https://healthequityworks.wustl.edu/items/for-the-sake-of-all-a-report-on-the-health-and-well-being-of-african-americans-in-st-louis-and-why-it-matters-for-everyone/

37.

Haukoos

. The impact of nontargeted HIV screening in emergency departments and the ongoing need for targeted strategies. Arch Intern Med. 2012;172(1):20-22. doi:10.1001/archinternmed.2011.538

38.

Soh

LYJ

Chow

EPF

Johnson

Jamil

Ong

. HIV Testing uptake according to opt-in, opt-out or risk-based testing approaches: a systematic review and meta-analysis. Curr HIV/AIDS Rep 2022;19(5):375. doi:10.1007/s11904-022-00614-0

39.

Campos-Outcalt

Mickey

Weisbuch

Jones

. Integrating routine HIV testing into a public health STD clinic. Public Health Rep 2006;121(2):175. doi:10.1177/003335490612100212

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB

0.22 MB

0.91 MB