The FANMI (“my FAMILY” in Creole) study to evaluate community-based cohort care for adolescent and young women living with HIV in Haiti: protocol for a randomized controlled trial

The FANMI model of care was designed based on individual interviews and focus groups with patients, providers, and GHESKIO’s adolescent community advisory board (CAB), which identified key barriers to adolescent retention in care. Figure 1 illustrates the contextual, social, and individual factors influencing retention among adolescents and youth living with HIV using the adapted social action theory. This theory emphasizes the context in which behavior occurs and the confluence of social interaction and self-regulatory processes that affects health outcomes [15]. Young girls living with HIV reported profound social isolation, family rejection, and stigma associated with attending a specialized HIV clinic. They also felt “dread” coming to the medical clinic as it reminded them of their disease. During clinic visits, they reported that their visits consisted of a series of short encounters with multiple health care providers – i.e. various combinations of a nurse, a clinician, a counselor, a phlebotomist, and a pharmacist. The FANMI model of care was developed to address the following barriers: 1) social isolation and lack of family support; 2) stigma associated with clinic-based care; and 3) disjointed care from multiple providers. The adolescent CAB named this model of care “FANMI,” which is Haitian Creole for “FAMILY.”

FANMI is an unblinded randomized controlled trial of 160 females 16–23 years old who are randomized to the FANMI model (intervention) or to standard care (control). The study objective is to evaluate the efficacy of the FANMI model compared to standard care on the primary outcome of retention in HIV care at 12 months after enrollment. Secondary outcomes include HIV-1 RNA viral load suppression, sexual risk behaviors, FANMI acceptability among participants and providers, and health care utilization and costs (Table 1).

Study eligibility criteria include: female; 16–23 years of age; not currently on ART or on ART ≤ 3 months; participant knowledge of HIV infection; willing to receive care at the clinic or in the community; and willing to provide consent or assent. Exclusion criteria include: pregnancy at enrollment; a severe HIV/AIDS-related illness requiring hospitalization or intensive medical follow-up; and clinician determination of a developmental stage not suited for study participation. Enrollment was limited to adolescent girls and young women ages 16–23 years because this demographic group represents 80% of AYLWH in Haiti [16‐18].

Participants are recruited from those who self-present for HIV testing at GHESKIO, those referred to GHESKIO for HIV testing by a community health worker, and those who receive an HIV test at GHESKIO’s mobile clinic. Potential participants are referred to the GHESKIO Adolescent Clinic, where clinic staff provide post-test counseling, medical assessment, same-day ART initiation, and management of opportunistic infections, per national guidelines, for adolescents and young adults up to 23 years of age [19, 20]. Research staff provide an overview of the study, invite eligible patients to enroll, and assess understanding of the study.

Young adults 18 years and older provide written consent; adolescents 16–17 years old provide written assent with written parental or guardian consent. Following enrollment, participants are randomized to the FANMI model or standard care by computer-generated randomization with block size of 10. Participants may deconsent to participating in the study at any time during the duration of study follow-up.

HIV services are provided to participants in both study arms according to World Health Organization (WHO) and Haitian guidelines [19, 20]. Newly diagnosed adolescents receive post-test counseling, are screened for opportunistic infections including TB, and initiated on ART. At monthly visits, participants are monitored for new symptoms, assessed for medication toxicity, and receive one-month medication refills. Prior to October 2018, patients were initiated on first-line ART regimens consisting of tenofovir, lamivudine, and efavirenz. Based on changes to Haiti’s national guidelines, patients were switched to first-line regimens of tenofovir, lamivudine, and dolutegravir starting in November 2018. Second-line ART regimens include tenofovovir and lamivudine with a ritonavir-boosted protease inhibitor, either lopinavir or atazanavir. Plasma HIV-1 viral load is measured at 3, 6 and 12 months after ART initiation; individuals with plasma HIV-1 RNA level > 1000 copies/ml meet with a nurse and social worker for an adherence assessment, development of an individualized adherence plan, and repeat viral load testing 3 months later. Family planning counseling and contraceptive methods are offered to all participants. Girls who become pregnant after enrollment continue to receive HIV services at the Adolescent Clinic or in FANMI groups, in addition to three antenatal care visits at GHESKIO’s Obstetric Clinic.

All participants receive GHESKIO’s routine retention support services, which include transportation fees (~ 100 Haitian gourdes/~$1.40 USD per visit), phone call reminders for scheduled visits, and home visits for missed appointments. When a patient turns 24 years old, Adolescent Clinic staff initiate a transition toGHESKIO’s adult HIV clinic through 3 counseling sessions with a social worker.

Figure 2 compares the characteristics of the FANMI model and standard care.

The primary outcome is retention in care defined as being alive and retained in care at 12 months with a visit between 11 and 13 months after study enrollment. Secondary outcomes include: viral suppression as defined by plasma HIV-1 RNA viral load <1000 copies/mL at 12 months after enrollment; sexual risk behavior, including self-reported sexual activity, condom use, incidence of sexually-transmitted infections, and pregnancy; acceptability of the FANMI intervention; and health care utilization and costs .

All participants complete a baseline questionnaire at the time of study enrollment, which includes socio-demographic, clinical, and behavioral data. Socio-demographic data includes parent/guardian information, schooling and level of education completed, employment, marital and housing status, and time and cost to travel to GHESKIO. Clinical variables include height, weight, WHO staging, ART regimen, history of pregnancy, sexually transmitted infections (STIs) (e.g. gonorrhea, chlamydia, and syphilis), CD4 count, and viral load. Behavioral measures listed in Table 1 include assessment of HIV testing history, HIV knowledge (HIV Prevention Information Deficits Questionnaire (AHRB)), HIV Stigma Scale for Children (HSSC-8 plus 4 items) [21], Module 75 Disclosure Questionnaire (Special Projects of National Significance Initiative) [22], Multidimensional Scale of Perceived Social Support (MSPSS) [23‐25], Center for Epidemiology Studies – Depression Tool (CES-D) [26, 27], Modified Alcohol Use Disorders Identification Test (AUDIT) [27, 28], Short Form Household Food Security Scale [29], Adolescent Sexual Activity Index (ASAI) [30], and a Modified AIDS Clinical Trials Group (ACTG) Adherence Assessment [31].

Six months after enrollment, participants are reassessed for WHO staging, sexual risk behaviors, alcohol and drug use and newly assessed for ART adherence, and health care utilization and costs. Viral load, CD4 count, tenofovir diphosphate levels, pregnancy, and incident gonorrhea and chlamydia are also measured 6 months after enrollment. Twelve months after enrollment, participants are administered the same assessments as were performed at enrollment and 6 months after enrollment, with the addition of tenofovir diphosphate levels (Table 1).

A subset of 30 participants in the FANMI arm will be recruited for qualitative interviews at 6 and 12 months to assess acceptability of the FANMI model. These interviews will be facilitated by a qualitative research assistant not involved in the study. Participants not retained in care at 12 months will be invited to return for a qualitative interview to explore reasons for non-retention and to identify further barriers.

We will also evaluate the fidelity of the intervention by evaluating a subset of 20 randomly selected FANMI group visits for adherence to components of the FANMI model of care. The selected FANMI sessions will be recorded and transcribed and then reviewed using a checklist to evaluate completion and time spent on each component of the intervention including peer socialization, group counseling and clinical care (symptom review, medication adherence assessment, medication refills, and phlebotomy when needed).

We based our sample size estimate on a retrospective review of retention from the electronic medical record for the standard care arm and data from a cohort care pilot project for the FANMI arm [32]. We estimate that 85% of participants in the FANMI arm will be alive and retained in care at 12 months, compared to 60% of participants in the standard arm. We calculated the required sample size for this trial using Fisher’s exact test. With a sample size of 160 (80 participants randomized to each arm), we will have ~ 93% power to detect this 25% difference, with an alpha of 5% (for a single primary outcome). Approximately 370 adolescent girls are newly diagnosed with HIV infection at GHESKIO each year. We estimate that 240 of 370 (65%) adolescent girls will satisfy the eligibility criteria and be willing to participate in the study each year. Our sample size also provides us sufficient power to analyze the secondary outcome of achieving viral suppression. With a sample size of 160, we will have > 80% power to detect a 30% difference between the number of participants achieving a viral level < 1000 copies/ml between the two arms. We estimate that 60% of participants in FANMI and 30% of participants in standard care will achieve viral suppression based on pilot data [32].