Disparities in HIV knowledge and attitudes toward biomedical interventions among the non-medical HIV workforce in the United States

ABSTRACT

Non-medical, community-based workers play a critical role in supporting people living with (or at risk of acquiring) HIV along the care continuum. The biomedical nature of promising advances in HIV prevention, such as pre-exposure prophylaxis and treatment-as-prevention, requires frontline workers to be knowledgeable about HIV science and treatment. This study was developed to: measure knowledge of HIV science and treatment within the HIV non-medical workforce, evaluate workers’ familiarity with and attitudes toward recent biomedical interventions, and identify factors that may affect HIV knowledge and attitudes. A 62-question, web-based survey was completed in English or Spanish between 2012 and 2014 by 3663 US-based employees, contractors, and volunteers working in AIDS service organizations, state/local health departments, and other community-based organizations in a non-medical capacity. Survey items captured the following: respondent demographics, HIV science and treatment knowledge, and familiarity with and attitudes toward biomedical interventions. An average of 61% of HIV knowledge questions were answered correctly. Higher knowledge scores were associated with higher education levels, work at organizations that serve people living with HIV/AIDS or who are at a high risk of acquiring HIV, and longer tenure in the field. Lower knowledge scores were associated with non-Hispanic Black or Black race/ethnicity and taking the survey in Spanish. Similarly, subgroup analyses showed that respondents who were non-Hispanic Black or Hispanic (versus non-Hispanic white), as well as those located in the South (versus other regions) scored significantly lower. These subpopulations were also less familiar with and had less positive attitudes toward newer biomedical prevention interventions. Respondents who took the survey in Spanish (versus English) had lower knowledge scores and higher familiarity with, but generally less positive attitudes toward, biomedical interventions. In summary, low knowledge scores suggest the need for additional capacity-building efforts and training for non-medical HIV workers, particularly those who provide services in the communities most affected by HIV.

KEYWORDS:

• HIV/AIDS
• survey
• workforce
• knowledge
• biomedical interventions
• attitudes

Introduction

Despite recent clinical advancements in preventing new human immunodeficiency virus (HIV) infections and improving health-related outcomes for individuals living with HIV and acquired immunodeficiency syndrome (AIDS), the rate of new infections in the United States (US) remains stable at approximately 50,000 per year (Centers for Disease Control and Prevention, 2015). The epicenter of new infections lies in the Southern US (accounting for 45% of new infections), and this region also has the highest proportion of HIV-related deaths (46%) (CDC, 2012; University of Virginia School of Medicine® and Med-IQ, 2014). HIV disproportionately affects Black/African American (hereinafter referred to as “Black”) and Hispanic/Latino communities; while these groups represent 12% and 16% of the US population, respectively, they account for approximately 44% and 21% of new HIV infections annually (CDC, 2013a, 2013b).

With the US Department of Health and Human Services (DHHS) treatment guidelines supporting antiretroviral treatment (ART) initiation for all people living with HIV (DHHS Panel on Antiretroviral Guidelines for Adults and Adolescents, 2016) and with the advent of biomedical prevention interventions such as pre-exposure prophylaxis (PrEP) (Baeten et al., 2012; Grant et al., 2010; Thigpen et al., 2012) and treatment-as-prevention (TAP) (Cohen et al., 2011), there are new opportunities for the non-medical HIV workforce to facilitate access to treatment and prevention measures. However, substantial gaps remain in the HIV care continuum, with only 37% of people living with HIV initiating ART nationally, and even fewer (30% of all people living with HIV) achieving viral suppression (CDC, 2014); evidence also suggests that these deficiencies may be more severe among Blacks compared with whites and Hispanics/Latinos (Dasgupta, Oster, Li, & Hall, 2016). The non-medical workforce, which includes employees of health departments, AIDS service organizations (ASOs), and other community-based organizations (CBOs) that serve people living with HIV, provides a critical link between medical providers and the individuals most impacted by HIV. In Black and Hispanic/Latino communities where histories of medical racism and mistrust are prevalent (Brooks, Newman, Duan, & Ortiz, 2007; Earl et al., 2013), integration and implementation of biomedical HIV interventions may rely on advocacy from trusted, non-medical support staff.

Scientific and treatment knowledge among the non-medical HIV workforce has not been evaluated previously. This survey-based study assessed knowledge about HIV science, treatment, and biomedical interventions; attitudes toward and beliefs about biomedical interventions; and how scientific knowledge levels correlate with attitudes toward using newer treatment and prevention tools, in order to identify areas for future training, capacity-building, and health policy efforts. In this report, we describe results from a large, national survey, with a specific focus on the communities most affected by HIV.

Methods

Study design

This research entailed a de-identified, 62-question, web-based survey that included the following categories: demographic and screening information (21 questions), HIV/AIDS knowledge (26 questions), and attitudes toward biomedical interventions (15 questions). The knowledge questions were divided into 3 categories: basic science and terminology, treatment, and biomedical interventions. Respondents were asked to rate their level of familiarity with various biomedical interventions (5-point interval scale), and their level of agreement with various statements related to the potential effectiveness of biomedical interventions and other aspects of community-based implementation (referred to as “attitudes and beliefs”; 5-point interval scale).

The survey was conducted in 4 waves: (1) in-person for attendees of the 2012 US Conference on AIDS (USCA); (2) online (in 2013; national rollout); (3) online in Spanish for Spanish-speaking, but not necessarily self-identified, Hispanic/Latino workers (in 2013; Spanish-language cohort); and (4) online for state AIDS directors and state/local department of health (DOH) workers (in 2014; DOH survey). Additional details are provided in the Supplementary Materials.

Inclusion criteria

All respondents met the following criteria: employee, contractor, or volunteer in an ASO, state/local health department, or CBO; HIV-focused work based primarily in the US or US territories; aged ≥18 years; able to receive an incentive for participation or willing to participate without receiving an incentive; did not participate in paid research studies on HIV treatment education in the past 12 months; and did not previously complete the survey. Duplicate participation was prevented by collecting respondents’ email addresses.

Subanalyses

In addition to overall findings, epidemiologically relevant subanalyses are reported according to geographic region, race/ethnicity, sexual orientation, and survey language (see Supplementary Materials for category descriptions).

Statistical analysis

Descriptive statistics are provided for respondent demographics. Contingency tables and ordinary, stepwise least squares regression analyses were used for statistical comparisons (performed on SPSS v.22). A chi-square test compared response percentages and a t test compared mean values. The net associations between the various respondent characteristics and knowledge score (shown in Figure 2) were evaluated based on the results of linear regression. Adjustments were made for multiple comparisons. The significance cutoff was P < 0.05.

Results

Population

A total of 3663 completed surveys were included in the analysis: USCA (n = 643), national rollout (n = 1523), Spanish-language cohort (n = 300), and DOH survey (n = 1197). Ninety-one percent of surveys were taken in English and 9% were taken in Spanish (Table 1). Overall, 48% of respondents were male, 25% were homosexual/gay/lesbian/same-gender loving, and 15% were HIV positive. Respondents from the South were well represented (45%), as were non-Hispanic Black (32%) and Hispanic (19%) respondents. Overall, 13% of respondents were considered men-who-have-sex-with-men (MSM) of color, 10% white/other MSM, and 77% non-MSM (i.e., women and heterosexual men).

Table 1. Demographic characteristics of respondents (N = 3663).

Parameter	n (%)
Current gender orientation
Male	1750 (48)
Female	1874 (51)
Transgender	39 (1)
Age, years
18–34	1575 (43)
35–44	699 (19)
45–54	746 (20)
≥55	643 (18)
Race/ethnicity^a,b
Non-Hispanic Black	1184 (32)
Non-Hispanic white	1576 (43)
Hispanic	705 (19)
Other	198 (5)
Survey language
English	3324 (91)
Spanish	339 (9)
Sexual orientation
Heterosexual/straight	2498 (68)
Homosexual/gay/lesbian/same-gender loving	915 (25)
Bisexual/other	250 (7)
HIV status
Negative	3012 (82)
Positive	541 (15)
Do not know	40 (1)
Decline to answer	70 (2)
Geographic region
Northeast	557 (15)
Midwest	506 (14)
South	1644 (45)
West	556 (15)
Other/no answer	400 (11)
Education level
Associate degree or below	1100 (30)
Bachelor’s degree	1131 (31)
Some post-graduate education or higher	1432 (39)
Current organization
ASO	1639 (45)
State/local health department	779 (21)
Other CBO	1245 (34)
Current position
Consultant	539 (15)
Employee^c	2634 (72)
Volunteer	490 (13)
Tenure in the HIV field, years
0–2	779 (21)
3–5	975 (27)
6–10	748 (20)
11–15	477 (13)
≥16	684 (19)
Number of staff members^b
0–10	1065 (29)
11–20	740 (20)
21–30	598 (16)
31–50	510 (14)
51–100	403 (11)
≥101	347 (9)

Note: HIV, human immunodeficiency virus; ASO, AIDS service organization; CBO, community-based organization; AIDS, acquired immunodeficiency syndrome.

^aRespondents were classified during analysis of the survey results based on their responses to a question about primary race and a question about Hispanic ethnicity.

^bPercentages may not total 100% due to rounding.

^cFull- or part-time employee.

Overall survey results

Among all respondents, an average of 61% (standard deviation, 19%) of knowledge questions were answered correctly. Questions related to basic science and terminology were answered correctly most often, while treatment questions and questions related to biomedical interventions were answered correctly less often (73%, 54%, and 45% answered correctly, respectively).

Less than half of all respondents were “extremely familiar” or “very familiar” with research on PrEP, experimental topical microbicides, experimental HIV vaccines, and TAP (Figure 1a). There was variability in attitudes toward and beliefs about biomedical interventions; the number of respondents choosing “strongly agree” or “somewhat agree” was highest when asked if they were interested in learning about new biomedical prevention methods (86%), and far lower when asked if they had proper knowledge and training to advocate for PrEP (54%) or TAP (55%) within their local communities (Figure 1b).

Figure 1. Summary of survey results (total population; N = 3663). (a) Familiarity with biomedical interventions. (b) Attitudes toward biomedical interventions. (c) Overall knowledge scores according to attitudes toward biomedical interventions.

Notes: Figure (c) the n values in each bar represent the number of respondents who answered strongly/somewhat agree (black bars), or strongly/somewhat disagree or neither agree nor disagree (gray bars), for each survey question. PrEP, pre-exposure prophylaxis; HIV, human immunodeficiency virus; TAP, treatment-as-prevention; ART, antiretroviral treatment; US, United States.

When respondents were categorized according to their level of agreement with various statements about biomedical interventions, those who had more positive attitudes had, on average, higher overall knowledge scores than those with less positive attitudes (Figure 1c). The largest disparity was seen for a statement about reducing the risk of HIV transmission by suppressing viral load with ART, which is the scientific concept behind TAP; there was also a large disparity for a statement about interest in learning about new biomedical prevention methods.

Factors associated with HIV knowledge score

Higher levels of education, providing services broadly to people living with HIV/AIDS, and longer tenure in the field of HIV were most strongly associated with above-average HIV knowledge (Figure 2). In contrast, factors most strongly associated with below-average HIV knowledge were Spanish survey language, Black race or Hispanic ethnicity, working for a smaller organization, and location in the South. Associations according to respondent geographic region and race/ethnicity are shown in Figures S1 and S2, respectively.

Figure 2. Factors associated with HIV knowledge score (total population; N = 3663).

Note: Positive and negative values refer to unstandardized regression coefficients (average change in survey scores per unit change of the independent variable, while controlling for other variables). All associations were statistically significant and the larger the value (whether positive or negative), the more significant the impact of that variable. HIV, human immunodeficiency virus; organization; AIDS, acquired immunodeficiency syndrome; MSM, men-who-have-sex-with-men; LGBT, lesbian, gay, bisexual, transgender.

Results by geographic region

Respondents from states with the highest HIV prevalence, especially in the South, tended to score below average for overall knowledge (Figure S3). Consistent with these findings and the association analysis, knowledge scores were significantly lower among respondents in the South (60%) compared with the Northeast (64%) and Midwest (63%; Figure S4A).

Respondents from the South and the Northeast tended to have less familiarity with biomedical interventions compared with those from other regions (Figure S4B), and respondents from the South had less positive attitudes toward biomedical interventions (Figure S4C). The largest disparity was observed for a statement about having proper knowledge and training to advocate for the use of PrEP; respondents from the Midwest tended to have the strongest levels of agreement. For all regions, respondents who were strongly or somewhat familiar with biomedical interventions had above-average knowledge scores (Figure S4D).

Results by race/ethnicity

Non-Hispanic Black and Hispanic respondents had significantly lower average knowledge scores than non-Hispanic white respondents (58%, 54%, and 67% of questions answered correctly, respectively; Figure S5A). Non-Hispanic whites were more familiar with biomedical interventions than non-Hispanic Blacks and Hispanics (Figure S5B), and they were also more likely to have positive attitudes toward biomedical interventions (Figure S5C). The largest disparities in attitudes and beliefs were observed for questions related to the concept of TAP and the belief that PrEP/TAP can decrease viral loads in the US. Despite these differences, all racial/ethnic groups were equally interested in learning about new biomedical prevention methods. Among respondents who had positive attitudes toward biomedical interventions, the trend of non-Hispanic whites having higher overall knowledge scores remained (Figure S5D).

Results for MSM

MSM of color had significantly lower average knowledge scores than did white/other MSM (61% and 70% of questions answered correctly, respectively; Figure S6A). Non-MSM respondents also had significantly lower knowledge scores (60%) than white/other MSM. MSM as a group were more familiar with research supporting PrEP and TAP than were non-MSM and, overall, familiarity was greater for the 2 prevention interventions that are currently available (PrEP and TAP; Figure S6B). In general, MSM had similar attitudes toward biomedical interventions regardless of race/ethnicity, and these attitudes tended to be more positive than those of non-MSM; moreover, all groups showed a similarly high interest in learning more about new biomedical prevention methods (Figure S6C). Among respondents who had more positive attitudes toward biomedical interventions, race/ethnicity remained a factor in knowledge scores, with higher knowledge levels for white/other MSM compared with MSM of color and non-MSM (Figure S6D).

Results by survey language

Fifty-five percent of respondents to the Spanish-language survey self-identified as Hispanic. Average knowledge scores were significantly lower for respondents who took the survey in Spanish versus English (40% and 63% of questions answered correctly, respectively; Figure S7A). While respondents who took the survey in Spanish had significantly higher familiarity regarding biomedical interventions compared with those who took it in English (Figure S7B), they also tended to have less positive attitudes toward biomedical interventions (Figure S7C). Some of the largest disparities in attitude concerned questions related to the concept of TAP and interest in learning about new biomedical prevention methods, with more positive attitudes held by respondents who took the survey in English. Notably, the largest attitudinal disparity, which concerned having the proper knowledge and training to advocate for the use of PrEP, showed that respondents who took the survey in Spanish had more positive attitudes. However, even among respondents who had positive attitudes, average knowledge scores for those who took the survey in English were consistently higher than for those who took it in Spanish (Figure S7D).

Discussion

Historically, the non-medical, community-based HIV workforce has played a critical role in ensuring that people living with HIV have access to life-saving treatments and that people at risk of acquiring HIV have access to effective prevention measures. In this study, we evaluated the technical knowledge and attitudes of the non-medical HIV workforce to understand their preparedness to implement and advocate for newer, ART-based prevention interventions within their local communities.

Overall, the results of this survey research showed that knowledge of HIV science and treatment among the non-medical workforce is poor (61% average knowledge score). It is unclear if the current level of knowledge has any impact on the delivery of care and services, or how high knowledge scores would need to be to correlate with positive outcomes; however, the average knowledge score was only 73% for the most basic questions, which indicates that there are knowledge gaps. Knowledge scores decreased as the questions progressed from basic to more technical concepts, consistent with the observation that treatment-related engagement is the point on the HIV continuum of care at which individuals tend to have the sharpest decline in success (CDC, 2014).

Workers who may serve subpopulations that are disproportionately impacted by HIV – people living in the South, Blacks, and Hispanics/Latinos – had significantly lower HIV knowledge scores than respondents in other regions and white respondents (except respondents from the West; the relatively lower knowledge levels in the West, versus Northeast and Midwest, may be related to high levels of intra-region variability in the West). Southern, Black, and Hispanic/Latino workers were also less familiar with and had generally less positive attitudes toward biomedical interventions. This may be due, in part, to cultural, societal, and/or structural barriers. For example, there are high levels of medical mistrust among Black and Hispanic/Latino communities in the US, which, in turn, is associated with skepticism about the efficacy of medications (Lopez-Cevallos, Harvey, & Warren, 2014; Washington, 2007). HIV stigma is also pervasive in Black and Hispanic/Latino communities, particularly in the South (Lichtenstein, 2003), and may be further fueled by intervention methods that do not focus on reducing the number of sexual partners or “risky behavior.” This stigma has the potential to influence workers’ attitudes and beliefs, the care they provide, and, ultimately, health outcomes (Institute of Medicine of the National Academies, 2003; Sheehan et al., 2016; Wong et al., 2013). In addition, limited resources in the South and structural barriers, such as poor healthcare infrastructure, lack of providers, lack of transportation, and clinic/appointment wait times, could potentially lead the local healthcare workforce to focus their efforts on more basic issues related to access to care (Freed, Hansberry, & Arrieta, 2013; Hefner, Wexler, & McAlearney, 2015). In this context, it is especially important to ensure that the Black and Hispanic/Latino HIV workforce has strong knowledge of and value the most effective HIV interventions, as this is paramount for uptake in disproportionately impacted communities. Cultural concordance is important in these communities, and while the technical knowledge of the Black and Hispanic/Latino HIV workforce was relatively low in this study, we are not able to draw conclusions about other skills, such as the ability to engage and relate to people living with HIV.

Familiarity with biomedical interventions was highest for the 2 interventions that are currently available, PrEP (US approval in 2012) and TAP (first manuscript supporting TAP published in 2011) (Cohen et al., 2011; US Food and Drug Administration, 2012). Notably, there were significantly lower levels of familiarity with all interventions for Black versus white respondents; thus, the community that needs these prevention interventions the most (i.e., the community with the higher risk of infection) was the least familiar with them. However, it is important to view these results in the context of the timeframe in which the survey was administered (2012–2014); familiarity would be expected to increase over time and, consistent with this, a large proportion of respondents (86%) expressed interest in learning about new biomedical prevention methods. This hypothesized trend towards increasing familiarity may be observed in the future by using results from the current study as a baseline for comparison with new studies of workers’ familiarity with, as well as knowledge of and attitudes toward, biomedical interventions.

Study findings showed that respondents with more positive attitudes tended to have higher knowledge scores than those with less positive attitudes. While causality cannot be determined due to the nature of the analysis, trends seen with certain questions suggest that a lack of scientific knowledge translates to a lack of belief or faith in a particular intervention or principle. For example, respondents who agreed that suppressing viral load with ART reduces the risk of transmitting HIV (i.e., the concept of TAP) had higher knowledge scores than those who did not (65% versus 43% correct answers). Of note, respondents who knew more about the science and treatment of HIV had a larger appetite to learn more about new biomedical prevention methods in the future. Interestingly, respondents with positive attitudes toward biomedical interventions had generally similar knowledge scores regardless of region, whereas racial/ethnic differences in knowledge scores were maintained. This pattern was also apparent for MSM; while MSM as a group tended to have more positive attitudes than non-MSM, MSM of color had consistently lower knowledge scores than white/other MSM, even among those with positive attitudes. Results based on survey language were more complex. Respondents who took the survey in Spanish (versus English) tended to have lower knowledge scores and higher familiarity with, but less positive attitudes toward, biomedical interventions (including less interest in learning about new biomedical prevention methods). These results are difficult to reconcile; it is possible that the small number of Spanish survey–takers influenced the results or that there are other covariates not previously considered.

This research supports the need for a coordinated effort to ensure that members of the non-medical HIV workforce have sufficient knowledge to implement effective HIV interventions (Tobias, Downes, Eddens, & Ruiz, 2012), particularly as the medical provider workforce faces constraints (Office of National AIDS Policy, 2015). A set of core competencies should serve as a standard that can be communicated and promoted, especially among the young and/or inexperienced non-medical workers that made up a large portion of the study cohort. There has been limited research to determine the requisite skills and training, and the current study demonstrates that knowledge and attitudes can vary greatly based on geographic region and race/ethnicity. Such data may help to inform decisions related to initial training and continuing education. With knowledge and the associated positive attitudes, as well as self-efficacy (i.e., capacity building, competencies), the non-medical HIV workforce will be enabled to promote positive outcomes.

This study is limited in some respects. There is no publicly available information on the number of HIV workers in the US or their demographics and, given the large number of volunteers and high rates of employee turnover, national figures are expected to be highly fluid. Therefore, the large convenience sample may not be representative of the general non-medical HIV workforce in the US. As with any “opt in” or incentivized research, there is a potential bias associated with volunteering to take the survey. In addition, the population categorized as Hispanic may not have been the same as the Hispanic/Latino population that is defined by the US Census. The lack of a baseline or a standard with which to compare respondent scores limits interpretation of the results, and the cross-sectional nature of the data prevents examination of causality between variables. It is also unknown if the statistically significant differences in knowledge scores and attitudes observed in this study will be meaningful in practice. Additionally, we did not assess the quality of care and services delivered by the respondents, and so we were not able to draw any conclusions related to their skills and effectiveness beyond a measure of scientific and technical knowledge.

The HIV/AIDS field has made significant clinical, social, and political advancements in the past 20 years. Much of the success has been due to the advocacy and commitment of the HIV/AIDS workforce and community. To continue that legacy, there remains an important need to provide the workforce with the skills, knowledge, and training to ensure that the newest advancements reach the people who need them most.

Acknowledgments

We thank all of the HIV workers who participated in this survey; staff members from the Black AIDS Institute, the Latino Commission on AIDS, and the National Alliance of State & Territorial AIDS Directors for supporting the work of this manuscript; and Alisha Bermudez (Added Value) for statistical support. Medical writing support for this manuscript was provided by Courtney St. Amour, PhD, of MedErgy, and was funded by Janssen Scientific Affairs, LLC.

Disclosure statement

In accordance with Taylor & Francis policy and the authors’ obligations as researchers, the authors report that Janssen Scientific Affairs, LLC, and the Black AIDS Institute provided funding for this survey-based study. R. M. Copeland and P. Wilson are employed by the Black AIDS Institute, which provides training and capacity building to the HIV workforce, including through its flagship training programs, the African American HIV University (AAHU) and the Black Treatment Advocates Network (BTAN); the Black AIDS Institute has received funding from Merck, Bristol-Myers Squibb, ViiV, and Gilead to support its HIV training programs. M. Penner is employed by the National Alliance of State & Territorial AIDS Directors, which receives funding from Janssen Therapeutics. E. Y. Wong and L. D. Parisi are employees and shareholders of Janssen Scientific Affairs, LLC. G. Betancourt, D. Garcia, and R. Abravanel report no potential conflict of interest. Medical writing support for this manuscript was provided by Courtney St. Amour, PhD, of MedErgy, and was funded by Janssen Scientific Affairs, LLC. These interests have been disclosed fully to Taylor & Francis, and an approval plan is in place for managing any potential conflicts arising from the stated interests.

Additional information

Funding

This study was funded by Janssen Scientific Affairs, LLC, and the Black AIDS Institute.