Impart: findings from a prison-based model of HIV assisted partner notification in Indonesia

Abstract

Introduction: Assisted partner notification (APN) safely and effectively increases partner awareness of HIV exposure, testing and case identification in community settings. Nonetheless, it has not been specifically developed or evaluated for use in prison settings where people with HIV often are diagnosed and may have difficulty contacting or otherwise notifying partners. We developed Impart, a prison-based APN model, and evaluated its efficacy in Indonesia to increase partner notification and HIV testing.

Methods: From January 2020 to January 2021, 55 incarcerated men with HIV were recruited as index participants from six jail and prison facilities in Jakarta in a two-group randomized trial comparing the outcomes of self-tell notification (treatment as usual) versus Impart APN in increasing partner notification and HIV testing. Participants voluntarily provided names and contact information for sex and drug-injection partners in the community with whom they had shared possible HIV exposure during the year prior to incarceration. Participants randomized to the self-tell only condition were coached in how to notify their partners by phone, mail or during an in-person visit within 6 weeks. Participants randomized to Impart APN could choose between self-tell notification or anonymous APN by a two-person team consisting of a nurse and outreach worker. We compared the proportion of partners in each group who were notified of exposure by the end of 6 weeks, subsequently tested and HIV diagnosed.

Results: Index participants (n = 55) selected 117 partners for notification. Compared to self-tell notification, Impart APN resulted in nearly a six-fold increase in the odds of a named partner being notified of HIV exposure. Nearly two thirds of the partners notified through Impart APN (15/24) completed HIV testing within 6 weeks post notification compared to none of those whom participants had self-notified. One-third of the partners (5/15) who completed HIV testing post notification were diagnosed as HIV positive for the first time.

Conclusions: Voluntary APN can be successfully implemented with a prison population and within a prison setting despite the many barriers to HIV notification that incarceration presents. Our findings suggest that the Impart model holds considerable promise to increase partner notification, HIV testing and diagnosis among sex and drug-injecting partners of HIV-positive incarcerated men.

Trial registration: ClinicalTrials.gov NCT04155320.

Keywords: HIV testing; Indonesia; contact tracing; partner notification; prisons; substance use.

Figure 1

Study flow diagram. Abbreviation: APN, assisted partner notification ^a We excluded men (n = 175) who were incarcerated more than 3 years (42%) or within 6 months of their planned release date (8%). We also excluded men who were unable to provide names for partners (22%), had already notified all partners (21%), reported no partners during the contact tracing period (18%) or reported that their partners were deceased (11%) or residing outside Jakarta (9%). We excluded men (n = 175) who were: incarcerated more than 3 years (42%) or within 6 months of their planned release date (8%). We also excluded men who were unable to provide names for partners (22%), had already notified all partners (21%) or reported no partners during the contact tracing period (18%), or reported that their partners were deceased (11%) or residing outside Jakarta (9%). Some men were ineligible for more than one reason. ^b Most men who met the study's eligibility criteria (55/66) agreed to participate; however, 11 declined to participate for reasons that included: feeling anxious (n = 4) or undecided (n = 3), lack of time (n = 2) or interest (n = 1) and illness (n = 1). Collectively, the 55 index participants identified 117 partners for notification. Participants (n = 28) randomized to the self‐tell only condition identified 66 partners. Impart‐assigned participants (n = 27) identified 51 partners, of whom six were selected to be personally notified and 45 (88%) through Impart APN. Table 2 shows the characteristics of partners whom index participants chose to notify.