Applying the Social Ecological Model to Explore HIV-Related Stigma in the Southeastern United States: A Qualitative Study

Background: Although evidence suggests HIV-related stigma directly affects health and behaviors, we have a limited understanding of stigma’s inuence beyond the individual-level. We aimed to describe HIV-related stigma and strategies to reduce it in the Southeastern U.S. within the context of the social ecological model (SEM). Methods: Qualitative surveys were distributed in-person at community events, conferences, and via email to persons affected by HIV periodically over 8 months. The nal sample size included 87 participants (33 persons living with HIV, 54 without HIV). A directed content analysis was used to code responses into ve levels of the SEM (individual, interpersonal, community, institutional, and structural). Results: Multiple themes emerged within each level: Individual – knowledge, fear, internalization; Interpersonal – social network; Community – judgements, discrimination, community organizations, norms; Institutional – competent providers, healthcare services; and Structural – systemic barriers, language, education. Conclusions: The ndings exemplify the need for a multi-level approach to intervene and reduce HIV-related stigma. Based on the experiences and suggestions of people affected by HIV, future interventions should include substantial consideration from persons affected by HIV.


Background
Despite great progress that has been made over the past decade, the human immunode ciency virus (HIV) care continuum has not seen remarkable improvement [1]. One reason for this may be due to HIVrelated stigma, which has been cited in multiple studies as an impediment to successful HIV care engagement across the continuum [2][3][4]. In 2018, the Southern United States (U.S.) accounted for approximately 52% of new HIV diagnoses despite making up only 38% of the U.S. population [5,6]. Although the HIV diagnosis rate in the South (15.7 cases per 100,000) is not drastically higher than the national rate (11.4 per 100,000 people), some Southeastern states (e.g., Florida, Georgia) have diagnosis rates twice that of the national rate [5]. In early 2019, the U.S. Department of Health and Human Services announced a plan to end the HIV epidemic to reduce HIV transmission by 90% by 2030 [7]. This plan targets 48 counties that account for 50% of new HIV diagnoses, 48% of which are Southern states [7,8]. The plan calls to reduce HIV transmission through early diagnosis, treatment to achieve sustained viral suppression, protection for people at risk for acquiring HIV, and rapid response to HIV outbreaks [7].

Conceptual Framework
The conceptual model of this research incorporates concepts from the Social Ecological Model (SEM).
The SEM is based on the assumptions that health behaviors are shaped by the interactions across multiple levels of in uence (individual, interpersonal, community, institutional, and structural) and can provide a framework for integrating the multiple levels for studying and intervening on HIV-related stigma [15].
The complexity of HIV-related stigma can be attributed not only to its unique differences in cultural and geographical contexts, but also its variations on an individual, interpersonal, community, institutional, and structural level [16]. These differences have led to incongruencies in the de nition, assessment, and reduction of HIV-related stigma [16]. This is an important gap in the literature because the SEM suggests that stigma may have a much broader impact beyond the individual-level. As communities consider interventions to address HIV-related stigma, we need more information about what is contributing to stigma, and suggestions to intervene. Community insights on factors associated with HIV-related stigma in the Southeastern U.S. are unknown. To bridge this gap, we sought information to (1)

Study Population
Qualitative interviews were conducted among participants who attended the Rural HIV conference in Southern Georgia. Purposive sampling was used to recruit individuals affected by HIV (e.g. people living with and without HIV, individuals working in HIV prevention, and individuals interested in learning about HIV) in Florida and Georgia. We developed and distributed a survey at a rural HIV conference in southern Georgia (September 2018), a pride parade in Florida (October 2018), and an HIV-related comprehensive planning meeting in Florida (November 2018). In May 2019, we distributed an online link through the Florida Department of Health to share with potential participants. Individuals were eligible to participate if they were 18 years or older and uent in English. Participants were excluded if they were not from Florida or Georgia or had missing information on state of residence (n = 8).

Study instrument
We created an anonymous 11-item survey consisting of 4 open-ended questions ( Table 1)

Ethics
The study was approved by the University of Florida Institutional Review Board with a waiver of obtaining written informed consent.

Analysis
A total of 87 surveys were collected from persons living in Florida and Georgia. We obtained qualitative data using surveys and a directed content analysis approach for data analysis. With a directed approach, an established theory guides initial coding. This method is appropriate when an existing theoretical framework (i.e., SEM) exist to explain the phenomenon (i.e., HIV-related stigma) in a new context [17]. All text was initially coded by two researchers (VR and RW) independently using the ve levels of the SEM (individual, interpersonal, community, institutional, structural). Following this, each response was coded to address speci c areas within each level ( Table 2). Given the complexity of HIV-related stigma, responses could be coded in more than one level and were grouped by PLWH and people not living with HIV (PNLWH). In the event of a discrepancy in coding, a third researcher (NS or SC) resolved con icts. We used a deductive process of group discussion and recoding and developed a codebook collaboratively. Finally, an audit trail was generated by VR, and RW checked the audit trail to capture thought processes and re ection of the text and to interpret personal reactions or biases. While each question was analyzed separately, the results from questions 1 and 2 were combined to describe: Experiences and examples of HIV-related stigma and questions 3 and 4 were combined to describe: Strategies to reduce HIV-related stigma ( Table 1). Table 1 Community Stigma Survey Open-Ended Questions

Fear
Participants tended to describe HIV-related stigma as a fear to disclose status. One person wrote, "...fears of disclosing status because of potential violence", another said, "people are afraid that if people nd out they have HIV they will suffer negative consequences such as isolation; loss of job or housing; physical harm". Participants also described fears related to risk perceptions, "fear of contracting HIV simply by being close to an HIV+ person…" Internalization Participants brought up similar examples of internalization, including, "HIV-related stigma is the stigma that we put upon ourselves. We have to accept ourselves as who we are with HIV rst." Interpersonal Social Network Participants described instances of family members and friends not wanting to share items such as food or utensils with a PLWH. One respondent remarked, "families are still giving people who are HIV+ different things to eat on". Two PLWH also brought up the concept of dating being a source of HIV-related stigma, for example, "I've seen people still afraid to date someone HIV+".
Other instances involved rejection by people close to them. Another example was, "an older person who has been going to her church for many years and eventually she felt comfortable enough to share her diagnosis with one of her closest friends. Soon after that the friend stopped coming to her house and stop hugging her and uses hand sanitizer after she awkwardly shakes her hand." Community Judgements Judgements was the most commonly cited example of HIV-related stigma in our sample. Many responses involved stereotyping: "that the person a icted was promiscuous or on drugs" and "only gay people get HIV", religion: "you got what you deserved, God's wrath on gays", and more general negative opinions: "prejudice against people infected with HIV in all facets of life" and, "the negative and often hurtful attitudes and ideas that people have about those of us living with HIV." Judgments came mostly from interactions with others. Only one person (without HIV) described media as an example of HIV-related stigma, writing, "[HIV medication] commercial. Although it depicts multiple races, it still insinuates HIV is a white, gay man's disease."

Discrimination
Responses ranged from avoidance to violence. Examples included, "…hate crimes against people living with HIV and AIDS", "not allowing an individual to participate in sports events", and "people avoiding, not talking and not wanting to be with you."

Institutional Competent Providers
Examples described providers who were not su ciently able to treat PLWH. There was not one predominant provider type in which this occurred. One participant shared an experience that exempli ed competency in providers in which they described as, "an orthopedic surgeon saying he cannot perform a total hip replacement on me because he has never done one on a person with my condition." Another person described their experience as someone who works with PLWH, "When I was giving birth the nurse had to get another person to start my IV because she couldn't get it started, I heard her down the hall telling the other nurse that I was high-risk because I work in the HIV department."

Healthcare Services
Participants described healthcare services as an example of HIV-related stigma, but one person described the setup of a health department: "…having to walk into a county health facility to get my meds at a special window in a crowded room full of people."

Structural Systemic Barriers
Participants discussed HIV-criminalization laws, citing the, "increased penalties in the 'justice system'… levied at people who are HIV positive". Multiple respondents also mentioned challenges with our laws, such as, ".... I would also like to see Florida modernize its HIV-related and sex work laws. I am excited that syringe exchange may nally be an option in

Community Judgements
Respondents generally described that a barrier to HIV-related stigma interventions refers to judgement from and within speci c communities. For example, "Reducing HIV-related stigma needs to focus on the religious community where the language used by religious leaders need to be more inclusive, so people don't keep on perpetuating hate language in their homes." Media was also described as a mode of decreasing judgments. Mediums such as diverse websites, commercials, and billboards would be useful to increase awareness about HIV. Respondents detailed, "Increasing your voice, being seen by way of billboards, brochures at every medical facility.......depicting all kinds of people old, young, white, black, brown, Asian, the disabled in brochures and prevention messages HIV does not discriminate anyone."

Community Organizations
Respondents agreed that inclusive, respectful organizations such as peer-led community-based organizations as well as government-funded agencies would decrease HIV-related stigma. "Agencies that make an effort to let people know they treat all persons with respect and dignity, and then actually do so."

Norms
Participants noted that HIV should be normalized as a chronic illness where people could feel comfortable disclosing their status or get tested without being judged, mentioning: "Messaging here is di cult but not impossible. Too many people still think HIV is a death sentence. It isn't. I may very well outlive many of my HIV negative friends. HIV is a treatable, manageable condition. We need to get that out there. At the same time, we don't want people to let their guard down too much. HIV is still serious".
PLWH also noted this approach could normalize HIV if peers living with HIV could tell their story.

Institutional Competent Providers
Respondents also explained how HIV-related stigma could be perpetuated among providers. Although the type of provider was not speci ed, participants tended to agreed that support, education, and communication could be improved, in other words, "HIV providers and Case Managers provide education using simple terms, in everyday words that people can understand and empathize rather than sympathize."

Healthcare Services
Respondents agreed that resources such as increased testing, providers, and healthcare centers are needed. Respondents noted that facilities should not be identi ed as HIV-speci c, rather they should provide a gamut of services. Speci cally, respondents commented, "Create a safe non-descriptive clinic that is not just for PLWH so they can access care without feeling singled out."

Structural Systemic Barriers
Participants discussed criminalization laws, discriminatory practices, and policies that could increase testing efforts, citing, "For example, not having a patient sign special consent forms saying they want HIV testing" and, "Modernize the outdated, unscienti c, discriminatory HIV speci c laws." PLWH speci cally mentioned engaging communities as a whole such as, "….Allow them [people living with HIV] to sit at the decision-making table." Language PLWH described language as meaningful, stating, "People or patient rst language being used to take the impact of harm from the intent behind describing people living with HIV."

Education
Participants often cited education focusing on HIV transmission, particularly geared towards school-aged children, "For my state/city, I think it could be helpful if we provided more information to the youth about HIV, in a conducive way. Implement a course on it." One respondent noted, "...Renewed community HIV/AIDS education prevention efforts targeting high risk communities utilizing real leaders in the community."

Several Approaches
It should be noted that a couple of respondents indicated that strategies to reduce HIV-related stigma should not be a single approach citing, "Any strategy to overcome HIV-related stigma must be multipronged, it must involve policy, media, speci c education, and health services integration".

Discussion
This is one of the rst studies to qualitatively describe examples of HIV-related stigma and strategies to reduce HIV-related stigma in the Southeastern U.S. within the context of the SEM. In this study, participants identi ed that HIV-related stigma encompasses individual (knowledge, fears, internalizations) interpersonal (social networks), community (judgement, discrimination, community organizations, norms), institutional (competent providers, healthcare services) and structural levels (systemic barriers, language, education) of the SEM which have been supported in previous studies [18] , [19]. These ndings demonstrate that a multi-component intervention that encompasses each of the levels of the SEM simultaneously could be useful to reduce HIV-related stigma in the Southeastern U.S.
Examples and experiences of HIV-related stigma were most commonly noted as instances of enacted stigma. Many of these experiences occurred on a community and institutional level which entailed judgments and discrimination from friends, family, and the community at large highlighting fears of sharing utensils, stereotyping, and violence. These ndings are consistent with a similar qualitative study, in which a representative sample of families in care for HIV in the US described enacted stigma as the most common theme [20]. A smaller number of respondents in our sample described structural stigma related to education, HIV-criminalization laws, and housing discrimination. Examples of HIV-related stigma were also experienced at an individual level (i.e., internalized negative attitudes about HIV, misconceptions about HIV transmission), and institutional level (i.e., anticipated stigma from healthcare providers). The present study's results have implications for the design of HIV-related stigma interventions. Although researchers mainly consider individual-level interventions, our results demonstrate that further research is needed on interventions that include all levels of the SEM.
Our ndings were similar to results of studies done in other settings which further support that a crosscutting approach may support an e cient, effective response to reduce HIV-related stigma in the Southeastern U.S. [21][22][23][24]. Most interventions to-date focus on a single social ecological level targeting individual levels of in uence [25]. While these studies address the knowledge, fears, and attitudes of individuals, these interventions may not adequately address HIV-related stigma manifestations on a community (e.g., community attitudes) interpersonal (e.g., social isolation from family and friends), institutional (e.g., healthcare worker discrimination of PLWH), or structural-level (e.g., discriminatory laws). Speci cally, fewer community, institutional, and structural-level interventions have been tested in a North American context. [25] HIV-related stigma interventions that fail to address all of the social ecological levels may not sustain long-term health seeking or preventative behaviors [25]. More research is needed to explore combinations of strategies at multiple social ecological levels for PLWH affected by HIV-related stigma in North America.
The use of purposive sampling served as a major limitation in this study, as our sample does not re ect the full diversity of the Southeastern U.S. Purposive sampling, however, is a common method for recruiting participants into qualitative research [26]. Furthermore, the sample was predominantly White and non-Hispanic, and more input from Black and other racial/ethnic groups could provide additional examples. Although, it is possible that the survey did not reach the people who experience the greatest amounts of stigma who may be less likely to participate in HIV-related research studies, we felt that including the perspective of both PLWH and PNLWH who worked with a larger population would help us to understand those experiences. Additionally, because this was a self-administered survey that only allowed short answers, we were unable to follow-up on responses that lacked clarity or warranted further explanation. Lastly, because this was a qualitative study, we are not able to make formal comparisons in attitudes or speci c suggestions in PLWH compared to PNLWH.
Despite these limitations, the study has several strengths. Participants were able to describe their experiences in an open-answer format, allowing them to expand on their thoughts. Additionally, this study recruited in different environments to attempt to include a variety of persons affected by HIV, including both people with and without HIV. This was one of the rst studies to qualitatively assess HIV-related stigma in the Southeastern U.S., thus provides a novel explanation of the experiences of HIV-related stigma and recommendations to combat it.

Conclusions
HIV-related stigma exists within every level of the SEM. Our ndings exemplify the need and potential bene ts from a multi-level approach to intervene and reduce HIV-related stigma in the Southeastern U.S. The data suggest that such an approach includes the following: (1) increasing knowledge by addressing fears, attitudes and beliefs about HIV (2) talking openly with family, friends, and partners to correct misconceptions (3) normalizing HIV through media campaigns and inclusive community organizations (4) structured education and training, speci cally for school or healthcare settings and (5)  All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. A waiver of informed consent was received for this study.
Availability of Data and Material: The full survey used for this study is included within the article.