Leveraging family-based assets for Black men who have sex with men in House Ball Communities: Protocol for a cluster randomized controlled trial
PMCID: PMC10490903
PMID: 37683036
Abstract
Black men who have sex with men (MSM) continue to have the highest incidence of new human immunodeficiency virus (HIV) diagnoses in the United States but are least likely to be engaged in care or to be virally suppressed. Many Black MSM face multiple stigmas, but some have found refuge in the House Ball Community (HBC)—a national network of Black lesbian, gay, bisexual, and transgender kinship commitments that provide care-giving, affirmation, and survival skills-building for its members. We propose to modify a skills-building and HIV prevention best-evidence, group-level intervention for HIV- negative Black MSM (Many Men Many Voices) into a family-based intervention to focus on asset-building for both HIV-negative and HIV-positive Black MSM within HBC families. The adapted intervention will be re-branded as Our Family Our Voices (OFOV). We proposed a mixed-methods study to test the feasibility and preliminary efficacy of OFOV adapted for HIV status-neutral use with HBC families. First, we will develop the intervention protocol using the ADAPT-ITT model for modifying behavioral interventions. Then, we will conduct a cluster randomized controlled trial with six HBC families in New York City. Families will be randomized to the OFOV intervention or waitlist control arm. Primary outcomes will be HIV testing, HIV pre-exposure prophylaxis use, currently in HIV care and on HIV treatment. Secondary outcomes will be the number of family-based assets, resilience, number of sexual partners, and relative frequency of condomless anal intercourse. The results of the formative research, including the pilot trial, will contribute to the evidence-base regarding the development of HIV status-neutral interventions that respond to the diversity and complexities of HBC families and that recognize the importance of asset-building for facilitating HBC resilience to stigma as a part of the United States’ domestic policy objective of ending the HIV epidemic by 2030.
Full Text
HIV continues to disproportionately affect Black communities in the United States (US). Current epidemiologic data indicate that Black individuals represent 42% of new HIV diagnoses, while representing only 13% of the US population [1]. Most new HIV diagnoses among Black individuals have been observed in men who have sex with men (MSM) [2]. Among Black males, male-to-male sexual contact accounted for 79% of new HIV diagnoses in 2019. It is estimated that intragroup transmission accounts for over half of the new HIV diagnoses via condomless anal intercourse (CAI) among Black males [1]; however, the likelihood of HIV risk exposure is influenced by a variety of socioecological factors [3–9].
For many Black MSM, the increased exposure to HIV risk is driven by multiple socioecological factors [9–12]. Some Black MSM seek refuge from environments that disaffirm and/or attempt to dismantle essential aspects of their personhood, including their gender expression(s) and sexualities [13, 14]. Nonetheless, exposure to disaffirming environments has been reported to have residual negative psychosocial impacts [15–17]. In our previous work, we found that early-life (before age 18) exposure to environments that disaffirmed same-gender sexual behaviors was associated with an increased number of sex partners and HIV diagnosis in adulthood [18]. In a longitudinal study [19], we also found that financial crisis was associated with reporting a greater number of male sexual partners in the past 6 months; a recent conviction was associated with a sexually transmitted infection (STI) diagnosis at 6-month follow-up; and unstable housing was associated with an STI diagnosis at 12-month follow-up. Thus, the absence of “behavioral risk factors” does not confer protection against other economic and social hardships that still exert influence on HIV outcomes for Black MSM. It is therefore necessary to identify and strengthen assets to support resilience to hardships that can undermine the effects of HIV interventions.
The House Ball Community (HBC)—a national network of Black lesbian, gay, bisexual, and transgender persons (LGBT) kinship commitments (families)—provides caring, affirmation, and survival skills-building for its members [14]. The HBC is a thriving community within the larger Black and Latinx LGBT communities. The HBC is organized to meet the needs of its members for social solidarity and mentoring in a society largely hostile to diversity in sexual and gender expression [20]. The term “houses” refers to the actual familial networks, which are generally not localized in any one place or residence. The “ballroom” or “balls” are the common terms for competitive dance and performance events that occur at regular intervals throughout the year and constitute an event circuit [21]. Each House structure has a “father” and a “mother” who assume the traditional roles performed by parents of origin regardless of gender, sex or age, and act as surrogate parents to the House. Generally, parents provide a safe haven and source of support for their house members in situations where they have been rejected by their families of origin or other social institutions [20]. The HBC population size is estimated to be approximately 8,000 nationwide, with the largest concentration in New York City [21]. Despite the emerging visibility and celebration of the HBC’s contributions to culture in the Black LGBT communities, alarming disparities in HIV persist. Previous studies with HBCs identified HIV prevalence as high as 20% [22], as well as a high prevalence of stigma and history of stressful life events [22, 23]. Moreover, Black HBC members were 2.5 times more likely than non-Black Latinx members to be unaware of their HIV status [23].
Many Men Many Voices (3MV) is a multi-session, group-level behavioral intervention developed to prevent HIV and STIs among Black MSM. A summary of the 3MV sessions is presented in Table 1. The intervention is designated as a best-evidence intervention by the U.S. Centers for Disease Control and Prevention for its demonstrated efficacy in improving rates of HIV testing and reducing the frequency of CAI [24, 25]. Although 3MV has strong evidence for reducing HIV/STI behavioral risk in the general population of Black MSM, it was not designed to specifically respond to the social and cultural dynamics of HBC family life that are realities for some Black MSM. To optimize the HIV prevention outcomes for Black MSM in HBC families, we sought to go beyond identifying and reducing risk factors towards simultaneously identifying, increasing, and supporting assets that are present in HBC family life.
Internal and external assets are key influencing factors in the health-seeking behaviors of MSM. Building internal social assets as part of family-based socioecological assets is critical to facilitating engagement in health-promoting behaviors [26–32]. External assets are defined as relationships, support systems (e.g., family support), and opportunities such as positive peer influence and creative activities [33–35]. Internal assets are skills and personal values such as self-esteem, a sense of purpose, positive identity, and planning/decision-making capacities [35, 36]. Although it is important to have interventions that focus on reducing behavioral risk for HIV, it is important that Black MSM be able to build interpersonal-level (peer-group/family) and intrapersonal-level (individual) capacities to challenge stigma, recover from stigmatizing experiences, and use strategies to protect against the internalization of stigma that may result from chronic exposures to stigma and social hardships [37, 38]. While limited data exist on assets among Black LGBT youth, available evidence indicates that LGBT youth generally have fewer assets than heterosexual youth. For example, in a population-based survey of assets among youth in grades 6–12, compared to non-LGBT youth, smaller proportions of LGBT youth had assets related to family support, positive family communication, safety, feeling valued in the community, family boundaries, adult role models, and youth programs [39]. The only asset with a higher prevalence among LGBT youth was engagement in creative activities. This intersects with the substantial existing evidence that LGBT youth of color have worse outcomes compared to the general population of LGBT youth [40–42]. To date, there are no known group-level, HIV behavioral risk-reduction interventions that also incorporate building family-based socioecological assets to optimize prevention and treatment outcomes for Black MSM. For this reason, 3MV is being adapted to be more responsive to the sociocultural realities of HBC life and to include family-based assets. The primary aims of this study are to: (1) conduct formative research to identify key modifications to the 3MV intervention manual and implementation protocol and (2) conduct a pilot cluster randomized controlled trial (CRCT) to determine the feasibility and acceptability of the modified intervention manual and trial protocol.
The description of the study protocol conforms to the specifications of the Standard Protocol Items Recommendations for Intervention Trials (SPIRIT) checklist (include as a S1 File). This study consists of a formative phase and a trial phase, as detailed in Fig 1 and depicted in Fig 2. We will use a mixed-methods approach to investigate our study aims. First, we will use qualitative (focus group discussions [FGDs] and interviews) methods to determine what HBC family-specific adaptations need to be made to the original intervention. We will then conduct a CRCT (quantitative) to determine the feasibility and acceptability of the trial protocol as well as to estimate the preliminary effect size of the OFOV intervention on HIV testing, PrEP use, engagement in HIV care and current use of antiretroviral therapy for HIV treatment. The ADAPT-ITT model is a systematic framework that outlines a step-by-step process for researchers and implementers to follow when translating existing HIV behavioral interventions for use with different populations [43, 44]. A modified ADAPT-ITT model will guide our approach to enhancing 3MV’s content to incorporate an asset-building framework that leverages the strengths of the families. The framework has been used in previous studies, including a study adapting 3MV for use with MSM in Ghana, West Africa [45]. Aim 1 will encompass ADPAT-ITT steps 1–7, leading to the combined multi-level intervention adapted to improve congruence with the realities of HBC family life. Although our previous programmatic experience already provided us with evidence to know what our selected intervention of focus will be, and that it requires adaptation for HBC (usually ADAPT-ITT Step 2), we will still leverage the ADAPT-ITT model to systematically guide our adaptation process. Aim 2 encompasses step 8, in which we will assess the feasibility and acceptability of conducting a CRCT of OFOV with a standard of care (SOC) wait-listed control condition. The study will be conducted in New York City (NYC), which is the geographic area with the highest concentration of individuals in the HBC [21].
The self-report measures for the CRCT phase of the study (Aim 2) are summarized in Table 2. Our primary outcomes will be HIV testing and PrEP use (for HIV-negative or HIV status unknown members) or “In HIV care” and “currently on HIV treatment” for people living with HIV (PLHIV) within 6 months of randomization. For HIV-negative MSM we will also record the frequency of HIV testing over the prior 3-months; however, we anticipate that few participants will take >1 test in this short period of time. Our secondary outcomes will be the number of family-based assets, resilience, number of sexual partners, and relative frequency of CAI. We will also assess sense of community [46] at all time points. In our previous research with Black MSM, we found that sense of community was associated with increased condom use for anal intercourse (OR = 1.26, 95%CI 1.05,1.52; p < .05) [47]. We will characterize the sample on intersectional stigma by creating a latent variable that is drawn from three scales that measures HIV stigma, same-sex stigma, and gender non-conforming stigma. For HIV negative MSM, the latent intersectional stigma variable will exclude the enacted and internalized subscales of the HIV stigma scale because those are designed for PLHIV. We will also assess implementation outcomes of feasibility and acceptability.
In these steps, we will receive feedback from the topical experts and summarize it in a report that will be distributed to the entire investigator team, including our key collaborators. We will discuss the feedback and provide follow-up clarifying questions to the topic experts, as necessary, before making final revisions to the adapted intervention manual. We will also produce a training manual to standardize training and permit future replication of the intervention in other settings. We will provide comprehensive training to facilitators whose roles are dedicated to delivering the intervention. We will also develop the training to be multi-faceted to facilitate comprehension and retention of training concepts, using strategies that the PIs have successfully deployed in other HIV prevention research projects [54].
We will use a rigorous intention-to-treat (ITT) analytic approach to evaluate the intervention outcomes of the pilot CRCT conducted under Aim 2. In an ITT approach participants are included in the analysis as originally assigned, regardless of whether they actually receive the intervention or SOC [55]. The study allocation sequence will be determined by the statistician. The randomization procedure will be carried out by the research coordinator. First, we will compare baseline data to see if randomization resulted in equivalent groups. If we determine non-equivalence, then the non-equivalent variables will be accounted for in the final analyses using a difference in differences analytic technique where randomization is used to minimize any perception of bias in selection of families by investigators. We will determine the proportions of HIV testing at 3- and 6-month of follow-up assessments among family members with 95% confidence intervals (CI). To test the primary study hypotheses that the intervention will increase uptake of HIV testing and PrEP (for HIV-negative members) and attendance at medical care appointment and ARV use (among PLHIV) within six months of randomization, we will use generalized estimating equation (GEE) with binomial distribution [56, 57], followed by effect size calculation with relative risk (RR). As odds ratios (OR) may overestimate RR for common events like those in our study, we will use a modified GEE with Poisson distribution [56, 57], which has robust error variances to estimate the RR and 95% CI of the OFOV intervention compared to SOC. Additionally, a modified Poisson regression approach will also be used to compute the RR and 95% CI for the binary outcomes (yes/no)–HIV testing, PrEP use, “engaged in HIV care” and “on ARV treatment”. These outcomes can be dependent on the health care facility (HCF) environment, such that participants who use services from the same HCF may be corrected or “clustered.” To account for the potential post-randomization clustering effect in this CRCT, a generalized linear mixed effects model with a logit link function will be fitted [58], and this model includes both HCF cluster-level and family-level factors.
Sections
"[{\"pmc\": \"PMC10490903\", \"pmid\": \"37683036\", \"reference_ids\": [\"pone.0289681.ref001\", \"pone.0289681.ref002\", \"pone.0289681.ref001\", \"pone.0289681.ref003\", \"pone.0289681.ref009\"], \"section\": \"Introduction\", \"text\": \"HIV continues to disproportionately affect Black communities in the United States (US). Current epidemiologic data indicate that Black individuals represent 42% of new HIV diagnoses, while representing only 13% of the US population [1]. Most new HIV diagnoses among Black individuals have been observed in men who have sex with men (MSM) [2]. Among Black males, male-to-male sexual contact accounted for 79% of new HIV diagnoses in 2019. It is estimated that intragroup transmission accounts for over half of the new HIV diagnoses via condomless anal intercourse (CAI) among Black males [1]; however, the likelihood of HIV risk exposure is influenced by a variety of socioecological factors [3\\u20139].\"}, {\"pmc\": \"PMC10490903\", \"pmid\": \"37683036\", \"reference_ids\": [\"pone.0289681.ref009\", \"pone.0289681.ref012\", \"pone.0289681.ref013\", \"pone.0289681.ref014\", \"pone.0289681.ref015\", \"pone.0289681.ref017\", \"pone.0289681.ref018\", \"pone.0289681.ref019\"], \"section\": \"Introduction\", \"text\": \"For many Black MSM, the increased exposure to HIV risk is driven by multiple socioecological factors [9\\u201312]. Some Black MSM seek refuge from environments that disaffirm and/or attempt to dismantle essential aspects of their personhood, including their gender expression(s) and sexualities [13, 14]. Nonetheless, exposure to disaffirming environments has been reported to have residual negative psychosocial impacts [15\\u201317]. In our previous work, we found that early-life (before age 18) exposure to environments that disaffirmed same-gender sexual behaviors was associated with an increased number of sex partners and HIV diagnosis in adulthood [18]. In a longitudinal study [19], we also found that financial crisis was associated with reporting a greater number of male sexual partners in the past 6 months; a recent conviction was associated with a sexually transmitted infection (STI) diagnosis at 6-month follow-up; and unstable housing was associated with an STI diagnosis at 12-month follow-up. Thus, the absence of \\u201cbehavioral risk factors\\u201d does not confer protection against other economic and social hardships that still exert influence on HIV outcomes for Black MSM. It is therefore necessary to identify and strengthen assets to support resilience to hardships that can undermine the effects of HIV interventions.\"}, {\"pmc\": \"PMC10490903\", \"pmid\": \"37683036\", \"reference_ids\": [\"pone.0289681.ref014\", \"pone.0289681.ref020\", \"pone.0289681.ref021\", \"pone.0289681.ref020\", \"pone.0289681.ref021\", \"pone.0289681.ref022\", \"pone.0289681.ref022\", \"pone.0289681.ref023\", \"pone.0289681.ref023\"], \"section\": \"Introduction\", \"text\": \"The House Ball Community (HBC)\\u2014a national network of Black lesbian, gay, bisexual, and transgender persons (LGBT) kinship commitments (families)\\u2014provides caring, affirmation, and survival skills-building for its members [14]. The HBC is a thriving community within the larger Black and Latinx LGBT communities. The HBC is organized to meet the needs of its members for social solidarity and mentoring in a society largely hostile to diversity in sexual and gender expression [20]. The term \\u201chouses\\u201d refers to the actual familial networks, which are generally not localized in any one place or residence. The \\u201cballroom\\u201d or \\u201cballs\\u201d are the common terms for competitive dance and performance events that occur at regular intervals throughout the year and constitute an event circuit [21]. Each House structure has a \\u201cfather\\u201d and a \\u201cmother\\u201d who assume the traditional roles performed by parents of origin regardless of gender, sex or age, and act as surrogate parents to the House. Generally, parents provide a safe haven and source of support for their house members in situations where they have been rejected by their families of origin or other social institutions [20]. The HBC population size is estimated to be approximately 8,000 nationwide, with the largest concentration in New York City [21]. Despite the emerging visibility and celebration of the HBC\\u2019s contributions to culture in the Black LGBT communities, alarming disparities in HIV persist. Previous studies with HBCs identified HIV prevalence as high as 20% [22], as well as a high prevalence of stigma and history of stressful life events [22, 23]. Moreover, Black HBC members were 2.5 times more likely than non-Black Latinx members to be unaware of their HIV status [23].\"}, {\"pmc\": \"PMC10490903\", \"pmid\": \"37683036\", \"reference_ids\": [\"pone.0289681.t001\", \"pone.0289681.ref024\", \"pone.0289681.ref025\"], \"section\": \"Introduction\", \"text\": \"Many Men Many Voices (3MV) is a multi-session, group-level behavioral intervention developed to prevent HIV and STIs among Black MSM. A summary of the 3MV sessions is presented in Table 1. The intervention is designated as a best-evidence intervention by the U.S. Centers for Disease Control and Prevention for its demonstrated efficacy in improving rates of HIV testing and reducing the frequency of CAI [24, 25]. Although 3MV has strong evidence for reducing HIV/STI behavioral risk in the general population of Black MSM, it was not designed to specifically respond to the social and cultural dynamics of HBC family life that are realities for some Black MSM. To optimize the HIV prevention outcomes for Black MSM in HBC families, we sought to go beyond identifying and reducing risk factors towards simultaneously identifying, increasing, and supporting assets that are present in HBC family life.\"}, {\"pmc\": \"PMC10490903\", \"pmid\": \"37683036\", \"reference_ids\": [\"pone.0289681.ref026\", \"pone.0289681.ref032\", \"pone.0289681.ref033\", \"pone.0289681.ref035\", \"pone.0289681.ref035\", \"pone.0289681.ref036\", \"pone.0289681.ref037\", \"pone.0289681.ref038\", \"pone.0289681.ref039\", \"pone.0289681.ref040\", \"pone.0289681.ref042\"], \"section\": \"Introduction\", \"text\": \"Internal and external assets are key influencing factors in the health-seeking behaviors of MSM. Building internal social assets as part of family-based socioecological assets is critical to facilitating engagement in health-promoting behaviors [26\\u201332]. External assets are defined as relationships, support systems (e.g., family support), and opportunities such as positive peer influence and creative activities [33\\u201335]. Internal assets are skills and personal values such as self-esteem, a sense of purpose, positive identity, and planning/decision-making capacities [35, 36]. Although it is important to have interventions that focus on reducing behavioral risk for HIV, it is important that Black MSM be able to build interpersonal-level (peer-group/family) and intrapersonal-level (individual) capacities to challenge stigma, recover from stigmatizing experiences, and use strategies to protect against the internalization of stigma that may result from chronic exposures to stigma and social hardships [37, 38]. While limited data exist on assets among Black LGBT youth, available evidence indicates that LGBT youth generally have fewer assets than heterosexual youth. For example, in a population-based survey of assets among youth in grades 6\\u201312, compared to non-LGBT youth, smaller proportions of LGBT youth had assets related to family support, positive family communication, safety, feeling valued in the community, family boundaries, adult role models, and youth programs [39]. The only asset with a higher prevalence among LGBT youth was engagement in creative activities. This intersects with the substantial existing evidence that LGBT youth of color have worse outcomes compared to the general population of LGBT youth [40\\u201342]. To date, there are no known group-level, HIV behavioral risk-reduction interventions that also incorporate building family-based socioecological assets to optimize prevention and treatment outcomes for Black MSM. For this reason, 3MV is being adapted to be more responsive to the sociocultural realities of HBC life and to include family-based assets. The primary aims of this study are to: (1) conduct formative research to identify key modifications to the 3MV intervention manual and implementation protocol and (2) conduct a pilot cluster randomized controlled trial (CRCT) to determine the feasibility and acceptability of the modified intervention manual and trial protocol.\"}, {\"pmc\": \"PMC10490903\", \"pmid\": \"37683036\", \"reference_ids\": [\"pone.0289681.s001\", \"pone.0289681.g001\", \"pone.0289681.g002\", \"pone.0289681.ref043\", \"pone.0289681.ref044\", \"pone.0289681.ref045\", \"pone.0289681.ref021\"], \"section\": \"Study design and setting\", \"text\": \"The description of the study protocol conforms to the specifications of the Standard Protocol Items Recommendations for Intervention Trials (SPIRIT) checklist (include as a S1 File). This study consists of a formative phase and a trial phase, as detailed in Fig 1 and depicted in Fig 2. We will use a mixed-methods approach to investigate our study aims. First, we will use qualitative (focus group discussions [FGDs] and interviews) methods to determine what HBC family-specific adaptations need to be made to the original intervention. We will then conduct a CRCT (quantitative) to determine the feasibility and acceptability of the trial protocol as well as to estimate the preliminary effect size of the OFOV intervention on HIV testing, PrEP use, engagement in HIV care and current use of antiretroviral therapy for HIV treatment. The ADAPT-ITT model is a systematic framework that outlines a step-by-step process for researchers and implementers to follow when translating existing HIV behavioral interventions for use with different populations [43, 44]. A modified ADAPT-ITT model will guide our approach to enhancing 3MV\\u2019s content to incorporate an asset-building framework that leverages the strengths of the families. The framework has been used in previous studies, including a study adapting 3MV for use with MSM in Ghana, West Africa [45]. Aim 1 will encompass ADPAT-ITT steps 1\\u20137, leading to the combined multi-level intervention adapted to improve congruence with the realities of HBC family life. Although our previous programmatic experience already provided us with evidence to know what our selected intervention of focus will be, and that it requires adaptation for HBC (usually ADAPT-ITT Step 2), we will still leverage the ADAPT-ITT model to systematically guide our adaptation process. Aim 2 encompasses step 8, in which we will assess the feasibility and acceptability of conducting a CRCT of OFOV with a standard of care (SOC) wait-listed control condition. The study will be conducted in New York City (NYC), which is the geographic area with the highest concentration of individuals in the HBC [21].\"}, {\"pmc\": \"PMC10490903\", \"pmid\": \"37683036\", \"reference_ids\": [\"pone.0289681.t002\", \"pone.0289681.ref046\", \"pone.0289681.ref047\"], \"section\": \"Self-report measures\", \"text\": \"The self-report measures for the CRCT phase of the study (Aim 2) are summarized in Table 2. Our primary outcomes will be HIV testing and PrEP use (for HIV-negative or HIV status unknown members) or \\u201cIn HIV care\\u201d and \\u201ccurrently on HIV treatment\\u201d for people living with HIV (PLHIV) within 6 months of randomization. For HIV-negative MSM we will also record the frequency of HIV testing over the prior 3-months; however, we anticipate that few participants will take >1 test in this short period of time. Our secondary outcomes will be the number of family-based assets, resilience, number of sexual partners, and relative frequency of CAI. We will also assess sense of community [46] at all time points. In our previous research with Black MSM, we found that sense of community was associated with increased condom use for anal intercourse (OR = 1.26, 95%CI 1.05,1.52; p < .05) [47]. We will characterize the sample on intersectional stigma by creating a latent variable that is drawn from three scales that measures HIV stigma, same-sex stigma, and gender non-conforming stigma. For HIV negative MSM, the latent intersectional stigma variable will exclude the enacted and internalized subscales of the HIV stigma scale because those are designed for PLHIV. We will also assess implementation outcomes of feasibility and acceptability.\"}, {\"pmc\": \"PMC10490903\", \"pmid\": \"37683036\", \"reference_ids\": [\"pone.0289681.ref054\"], \"section\": \"Steps 6 (integrate) and 7 (train)\", \"text\": \"In these steps, we will receive feedback from the topical experts and summarize it in a report that will be distributed to the entire investigator team, including our key collaborators. We will discuss the feedback and provide follow-up clarifying questions to the topic experts, as necessary, before making final revisions to the adapted intervention manual. We will also produce a training manual to standardize training and permit future replication of the intervention in other settings. We will provide comprehensive training to facilitators whose roles are dedicated to delivering the intervention. We will also develop the training to be multi-faceted to facilitate comprehension and retention of training concepts, using strategies that the PIs have successfully deployed in other HIV prevention research projects [54].\"}, {\"pmc\": \"PMC10490903\", \"pmid\": \"37683036\", \"reference_ids\": [\"pone.0289681.ref055\", \"pone.0289681.ref056\", \"pone.0289681.ref057\", \"pone.0289681.ref056\", \"pone.0289681.ref057\", \"pone.0289681.ref058\"], \"section\": \"Statistical analysis plan\", \"text\": \"We will use a rigorous intention-to-treat (ITT) analytic approach to evaluate the intervention outcomes of the pilot CRCT conducted under Aim 2. In an ITT approach participants are included in the analysis as originally assigned, regardless of whether they actually receive the intervention or SOC [55]. The study allocation sequence will be determined by the statistician. The randomization procedure will be carried out by the research coordinator. First, we will compare baseline data to see if randomization resulted in equivalent groups. If we determine non-equivalence, then the non-equivalent variables will be accounted for in the final analyses using a difference in differences analytic technique where randomization is used to minimize any perception of bias in selection of families by investigators. We will determine the proportions of HIV testing at 3- and 6-month of follow-up assessments among family members with 95% confidence intervals (CI). To test the primary study hypotheses that the intervention will increase uptake of HIV testing and PrEP (for HIV-negative members) and attendance at medical care appointment and ARV use (among PLHIV) within six months of randomization, we will use generalized estimating equation (GEE) with binomial distribution [56, 57], followed by effect size calculation with relative risk (RR). As odds ratios (OR) may overestimate RR for common events like those in our study, we will use a modified GEE with Poisson distribution [56, 57], which has robust error variances to estimate the RR and 95% CI of the OFOV intervention compared to SOC. Additionally, a modified Poisson regression approach will also be used to compute the RR and 95% CI for the binary outcomes (yes/no)\\u2013HIV testing, PrEP use, \\u201cengaged in HIV care\\u201d and \\u201con ARV treatment\\u201d. These outcomes can be dependent on the health care facility (HCF) environment, such that participants who use services from the same HCF may be corrected or \\u201cclustered.\\u201d To account for the potential post-randomization clustering effect in this CRCT, a generalized linear mixed effects model with a logit link function will be fitted [58], and this model includes both HCF cluster-level and family-level factors.\"}]"
Metadata
"{\"Data Availability\": \"No datasets were generated or analysed during the current study. All relevant data from this study will be made available upon study completion.\"}"