In any given year, one out of seven people living with HIV will pass through one of America's jails or prisons. When incarcerated people with HIV are released from prison, only about half remain engaged in care three years after release.
In 2001, the Centers for Disease Control and Prevention (CDC) launched a study designed to see if an evidence-based counseling program would work to improve HIV risk-reduction outcomes for incarcerated people post-release. They studied two groups of men -- an "intervention" group that received multiple pre- and post-release counseling sessions and a "comparison" group that received a single individual session prior to release from prison.
The program, called Project START, was a success, supporting the idea that thoughtfully designed counseling and rehabilitation programs can work to lower HIV risk among recently incarcerated people. Since then, copycat Project START programs have been adopted in other correctional settings for people living with HIV and those at risk of contracting HIV.
But despite nearly two decades of work on Project START pilot programs, HIV researchers still don't have a clear picture of what specific types of interventions work best for recently incarcerated people, according Christina McDonald, M.P.H., behavioral scientist in CDC's Division of HIV/AIDS Prevention, who presented a systematic review of post-release interventions at the 2019 National HIV Prevention Conference (NHPC) in Atlanta.
McDonald and her colleagues identified 14 studies that evaluated care programs designed to bolster engagement among people with HIV recently released from prison or jail. Among those 14 studies, 12 included strategies that implemented both pre- and post-release counseling. The most common counseling strategies included discharge planning, case management, patient navigation, and HIV education. Five of the studies used a one-group research design, meaning there was no comparison group.
Those five studies had promising results, demonstrating linkage-to-care outcomes that ranged from 70% to 100%. However, studies that compared one intervention against a control group are generally considered to be higher-quality evidence than one-group studies.
Among the nine two-group studies, results generally trended positively for the interventions, but results were inconclusive. Five of those studies reported positive benefits but didn't reach statistical significance, three found a significant impact but only along specific measures of care, and one study didn't find any impact at all. Assessing the impact of these programs as a whole was made even more complicated by the fact that only three of the studies reported retention in care and viral suppression rates.
In summary, the literature review found that the majority of high-quality, comparator studies hinted at a benefit, but that benefit did not reach statistical significance.
One big limitation of the review, McDonald said, is that many Project START–style pilot studies are summarized in unpublished abstracts, which are typically not included in literature reviews and therefore not incorporated in the overall compendium of evidence-based medical interventions.
"There are probably a lot of great, effective programs that are not included in this review because they are not published," she said, speaking at the conference.
In fact, the next speaker to take the podium after McDonald presented such a program. Barry Zack, M.P.H., associate clinical professor at the University of California, San Francisco, presented data from Girlfriends Connect, an intervention program modeled after Project START but adopted to serve transgender women incarcerated in men's penitentiary settings.
The program, implemented at prisons in San Francisco and Oakland, included coaching sessions that focused on post-release linkage to care, housing needs, and risk behavior education. The first two sessions were conducted pre-release and the last four sessions were conducted post-release.
According to data presented by Zack, 100% of the people with HIV who were enrolled in Girlfriends Connect were linked to care post-release, versus 70% of those who were not in the study arm with that program.
However, Zack cautioned that the number of people was small, and the study wasn't powered to show statistical significance.
"The research component has ended (with a small 'n' of less than 20), but the program piece continues with local funding," Zack said in an email.
Though the Girlfriends Connect project may not have met McDonald's inclusion criteria for a literature review, it's yet another hint that Project START interventions are on the right track by helping formerly incarcerated people navigate HIV care post-release.