What Will It Take to End the HIV/AIDS Epidemic? Linking the Most Disenfranchised Into Care Through Outreach

Network infrastructure

The unique aspect of the BANPH network was the collaboration of agencies from three counties: San Francisco, Alameda, and Contra Costa. These counties are joined by the San Francisco/Oakland Bay Bridge and the Bay Area Rapid Transit system. Some PLWHA seek services outside of their county of residence; therefore, coordinated efforts across counties are essential.

The Health Equity Institute (HEI) at San Francisco State University served as the lead agency for BANPH. HEI coordinated all network activities, including facilitating linkages between network partners and HIV care providers, leading evaluation activities, and providing ongoing technical assistance and trainings for partners. Local health departments and direct service providers conducted outreach to out-of-care PLWHA, identified PLWHA in tenuous care (i.e., those at risk of falling out of care), provided health education, and linked individuals to care and ancillary services. Formally contracted partners signed a Chain of Trust agreement to enable interagency sharing of client data for coordinating linkage to care and other services. Informal partners provided referral services and other resources for PLWHA.

In the planning stage, network agencies proposed an overall goal of linking 1000 PLWHA (one-eighth of the estimated 8000 out-of-care PLWHA in San Francisco and O-TGA) into medical care per year. This lofty goal was later reduced to 600 PLWHA per year when it became clear that locating and engaging the most hard-to-reach PLWHA would require intensive efforts and more resources. BANPH placed heavy emphasis on the collective target goal over individual agency goals to reduce interagency competition that is common among agencies that often competed for limited HIV program dollars. This collective model was key in creating an effective network.

Street outreach was selected as the main outreach method to reach the most marginalized PLWHA. HEI provided trainings on its street outreach model and protocols for fifteen outreach staff. Staff comprised near peers and peer navigators with no formal education beyond high school as well as case managers and nursing professionals with college or graduate-level education training. Outreach staff attended monthly case management meetings to share lessons learned from outreach activities and to discuss strategies to overcome challenges. During the course of our project, we saw a 20% turnover rate among agency outreach staff, which was considered typical among our partner agencies.

Participant recruitment

From June 2010 to June 2013, we conducted purposive identification of PLWHA who were out of care or were experiencing tenuous care in San Francisco, Oakland, and Richmond, California. Individuals were considered out of care if they had less than two HIV care visits in the past 12 months. Individuals experiencing tenuous care had routine visits, but had difficulty getting to medical appointments due to unstable circumstances in their lives. Our target populations included racial/ethnic minorities, adolescents and young adults, low-income individuals, and persons with a history of homelessness, incarceration, substance abuse, and/or mental illness.

Participants were recruited in several ways. Correctional facilities referred postincarcerated PLWHA to BANPH agencies upon release. Clinics examined internal patient data systems to identify PLWHA who have missed appointments or were at risk of falling out of care. Outreach workers, peer advocates, and patient navigators conducted street outreach and recruited in areas where our target populations frequented, including churches, homeless shelters, soup kitchens, needle exchange programs, and drug treatment centers. BANPH also established relationships with these highly frequented service organizations and recruited the organizations to refer known HIV-infected clients to the network.

Potential clients were given English/Spanish flyers about BANPH services and a toll-free number to speak with the HEI outreach coordinator for more information. When necessary, outreach workers had multiple contacts with clients to build rapport. New clients who were eligible and interested to use our services were asked to complete contact information forms, which documented their frequented locations and contact information for family members and friends. Contact was made in person or through phone. Client consent was obtained to have family members relay messages from outreach workers when the client could not be reached. Clients were also asked to complete our intake survey and to sign release of medical information forms. They were then referred to HIV care and/or ancillary services. Outreach workers followed up with individuals who were referred to HIV care to assess whether they had attended their medical appointment. If clients attended their appointment, then they were considered linked to care. If the clients did not attend their appointment, outreach workers tried to reengage those individuals to link them to care. For individuals who were referred to ancillary services, outreach workers followed up with them to assess if they had received services and if they were ready for HIV care. Clients who were ready for HIV care were referred to care. Outreach workers tried to reengage clients who were not yet ready for care or did not receive ancillary services.

Data collection

A $10 gift card to Safeway or Subway was given to clients as an incentive to complete an intake survey. Client intake data were collected using interviewer-administered surveys on Apple iPod Touch devices. Participants' names and addresses were collected on paper forms, but participants were assigned identification numbers for electronic data entry to avoid storing patient identifiable information on iPod devices. Fifteen peer advocates, outreach workers, and case managers from 10 BANPH sites used these devices to access the iForm™ app for data entry. When the iPod was connected to the internet through WiFi, survey data were automatically uploaded to a centralized cloud server that was compliant with the Health Insurance Portability and Accountability Act (HIPAA). Once uploaded, client responses were deleted from the iPod devices. The cloud server allowed for real-time data management by the HEI evaluation coordinator. Surveys containing data entry errors were flagged and electronically sent back to outreach staff for correction. Follow-up forms were generated and sent to iPods to prompt outreach workers to follow up and record client attendance of HIV care appointments.

There were two versions of the survey: one for PLWHA referred by correctional facilities and one for all other participants (in English and Spanish). The surveys documented client demographics with response categories adapted from the Centers for Disease Control and Prevention. ¹⁹ Using national standard measures, we collected data on income, housing status, history of incarceration, health status, support services accessed and needed, and barriers to care. Support service needs included drug and alcohol abuse treatment, housing/shelter, food, dental services, HIV-related medical services, general medical services, and medication services. The survey for postincarceration PLWHA listed additional support services that could ease transition from jail to community.

In addition to measuring stigma as a barrier, BANPH developed a stigma scale to measure the frequency that PLWHA thought about stigma. The following four items comprised the stigma summative rating scale: (1) I've felt that people avoided me because I have HIV, (2) I've feared I would lose friends if they learned that I was HIV positive, (3) I've thought other people were uncomfortable being with me because I am HIV positive, and (4) I've avoided getting treatment because someone might find out that I am HIV positive. Clients rated these items by frequency and the following four ordinal measures were summed: (1) not at all, (2) rarely, (3) sometimes, and (4) often. Minimum summated score was 4 and maximum was 16.

Data analyses

All analyses were performed using SPSS, Version 22 (SPSS Analytics; IBM, Inc., Armonk, NY). Data from incomplete surveys and surveys with unmatched client identification numbers were excluded from analysis. We examined the associations between linkage to care (indicated by a scheduled medical appointment) and client characteristics, barriers to care, and stigma.

Intensive intentional outreach is key to successful linkage to care

In a review, McKenzie et al. found nine essential strategies to track and follow-up with marginalized populations for longitudinal study cohorts, including (1) collection of contact information, (2) thorough organization of tracking efforts, (3) attention to staff training and support, (4) use of phone and mail follow-up, (5) use of incentives, (6) establishing rapport with participants, (7) assurance of confidentiality, (8) use of agency tracking, (9) use of field tracking, and (10) attention to safety concerns. ²⁹ These strategies had informed our outreach protocol design. Although these strategies were identified more than a decade ago for retaining participants in longitudinal studies, we found that many of these were applicable to our outreach efforts with PLWHA. However, we also found that written communications did not work as well with PLWHA. Our strategies of emailing clients, hand-delivering letters to their homes, and posting flyers with a toll-free number in frequented areas elicited few responses. These forms of communication require clients to make initial contact and thus may not be effective to engage hard-to-reach PLWHA who had little or no intention of receiving HIV care in the first place. Outreach strategies that use repeated, staff-initiated direct contact with hard-to-reach PLWHA may be more appropriate for linkage-to-care programs.

Diversity of organization types within a network can be more effective

Some PLWHA in our target populations did not attend ASOs. For this reason, we extended BANPH funding, trainings, and support to include CBOs who had never focused on PLWHA before, but had access to our target population. Each BANPH agency favored different outreach methods at the onset of the project. Agencies needed reminders and reinforcements to adopt street outreach as the preferred network approach. In addition, with the varying levels of experience working with PLWHA between agencies, a significant amount of time and effort was dedicated to training and technical assistance to build network capacity.

As part of our collaborative efforts, a correctional facility referred PLWHA to BANPH for our linkage to care program. Of those previously incarcerated PLWHA who completed our intake survey, 78.0% were linked to care. In a randomized control trial, Wohl et al. found that comprehensive prerelease discharge planning to establish referrals for HIV-positive incarcerated individuals was just as effective as an intensive case management intervention in linking individuals to care. ³⁰ Although a referral system might be sufficient to link previously incarcerated PLWHA to care, more could be done to increase linkage rates. In a study by Booker et al., it was reported that of the 1021 HIV-infected individuals who were previously incarcerated, 79% received clinical services through its multi-site linkage-to-care project. ³¹ The authors found several factors that were significantly associated with linkage to care for HIV-positive jail detainees within 30 days postrelease, including receiving HIV or medication education in the jail setting, having a completed discharge plan upon release, and having stable housing at 30 days postrelease. Linkage to care programs should explore collaborations with correctional facilities to establish a referral system to facilitate continuity of care for individuals postrelease. Further, researchers can utilize national corrections data and clinical data collected from the Ryan White Services Report to assess and improve linkage to care for individuals leaving corrections facilities. ³²

Clinics that were involved in our linkage-to-care program identified individuals who were experiencing tenuous care by analyzing their electronic health records. However, patients who appeared to be out of care might have been receiving care from other healthcare providers and did not require outreach for reengagement in care. Reengagement efforts could be further enhanced by integrating state HIV surveillance data to avert unnecessary outreach efforts to PLWHA who are receiving care elsewhere. ³³ The use of HIV surveillance data and correction facilities data could strengthen identification of individuals who are linked to care or not linked to care and help inform programmatic interventions.

Do not underestimate the resources and capacity needed for successful engagement

As part of a network of agencies located in one of the most resource-rich areas for HIV care and prevention, we expected that we had the capacity and resources to link 600 PLWHA into medical care per year. However, we overestimated our outreach capacity and underestimated the level of commitment required to engage the most marginalized populations. Outreach to PLWHA was time-consuming and labor-intensive. Engagement with PLWHA required multiple contacts over a period of a few months. Even with an additional grant-funded outreach worker at each partner agency, we linked less than 600 PLWHA into care over the course of 3 years. Funders should be cognizant that linkage-to-care programs could require more resources for longer periods of time to engage hard-to-reach PLWHA.

Our linkage-to-care project and analysis have limitations. First, since the category of PLWHA experiencing tenuous care was introduced in the second year of the project, some PLWHA did not receive BANPH resources in the previous year and might have fallen out of care. Second, the BANPH network was funded to carry out linkage-to-care activities up until the first medical appointment; postlinkage data on retention in care were not obtained. Last, depending on the makeup of agencies in a geographical area, our network outreach model might not be replicable.

Based on our findings, we recommend that (1) linkage-to-care programs collaborate with CBOs to target the most disenfranchised PLWHA and plan community-level interventions to reduce HIV stigma; (2) lead agencies of outreach networks establish an outreach definition for all partner agencies to act on; (3) outreach agencies take into account the time and labor required to engage PLWHA to realistically set target goals; (4) outreach workers conduct street outreach to out-of-care PLWHA, identify PLWHA who are experiencing tenuous care, connect clients to support services, coordinate and accompany clients to medical appointments, and provide health education and motivational counseling; and (5) funders consider providing long-term grants for networks to build capacity to engage the hardest-to-reach PLWHA.

PLWHA face insurmountable barriers to HIV care. With the ongoing treatment-as-prevention paradigm, intensive outreach activities targeting the most disenfranchised PLWHA for linkage to care and retention in care might be a promising strategy to finally end the HIV/AIDS epidemic in the United States. Our network outreach model demonstrated that agencies could be given the necessary training and tools to engage the most disenfranchised PLWHA, and interagency collaboration could help to facilitate outreach capacity building for regional linkage-to-care efforts. More resources and collaborative funding mechanisms are needed for long-term linkage-to-care interventions.