Interrupting HIV and TB Stigma in the Household in Uganda

Fogarty International Center of the National Institute of Health, New York University, Makerere University, Walimu, National Institute of Mental Health (NIMH)

The investigators will carry out a prospective, household cluster-randomized, implementation trial evaluating a complex, multi-component, social and behavioral intervention designed to normalize the acceptance of HIV testing in the household and increase diagnosis of HIV.

An estimated 39 million people worldwide are living with HIV, but nearly half do not know their status because they are unaware that they are at risk, unable to access counseling and testing, or unwilling to accept testing because of stigma and fear. Layered on to the stigma of HIV is the risk of tuberculosis (TB), the leading cause of death among persons living with HIV (PLWH). Home testing is a promising approach to increase testing and linkage to care for HIV because it can reach individuals outside the health system, eliminate the costs of traveling for testing, offer testing in a familiar environment, and increase engagement with care among those found to be living with HIV. Nevertheless, many individuals offered home HIV testing in sub-Saharan Africa decline to test. The investigators developed a complex intervention to increase uptake of testing for HIV among household members by reducing perceptions of stigma associated with HIV and testing for HIV. The intervention involves lay health workers (LHWs) delivering a novel invitation strategy for HIV testing in households that includes 1) acceptance-optimized sequencing, 2) prosocial messaging, and 3) salivary HIV testing. The investigators will administer HIV and TB stigma scales before and after the invitation. The investigators will conduct a household randomized controlled trial to test the intervention's impact on household HIV stigma and uptake of testing among household members undergoing TB contact investigation. The investigators will measure our co-primary outcomes of HIV and TB stigma using standardized instruments before invitation and after completion of post-test counseling. The investigators will measure the proportion consenting to HIV testing, yield of HIV diagnoses, and the proportion of new PLWH linked to HIV care at 1 month and reassess household HIV and TB stigma at 3 months. This study will be conducted in Kampala, Uganda, and will involve TB index patients and their household contacts. Households will be recruited and enrolled through index patients initiating treatment for pulmonary tuberculosis (PTB) at Kampala Capital City Authority health facilities. The investigators will enroll index patients and their households from Kiswa, Kawaala, and Kisenyi Health Centers, each a public-sector, primary care facility. This study will enroll households (index patients + their household contacts). Only household contacts are eligible for participation; close contacts are not eligible. For the purposes of this study, household contacts are defined as those individuals "sleeping under the same roof" as the index patient for one or more nights within the past three months. Upon encountering a patient initiating treatment for TB or returning for 2-week follow-up, study CHWs will assess index patients for household-level eligibility for the study. All index patients whose households meet these criteria will be asked to provide verbal consent after reviewing an information sheet. If the index patient is a minor (age < 18 years), the guardian of the index patient will be asked to provide verbal consent. After study enrollment, a CHW will work with the index patient and, if available, treatment supporter to schedule the home visit. Households will be randomly assigned to one of the two CHW teams to receive the intervention or standard or care strategy. After the enrollment of the household through the index patient, household contact investigation will take place. All household contacts who meet these criteria will be asked to provide verbal consent after reviewing an information sheet. If the contact is a minor (aged 15-17), he or she will be asked to provide verbal assent with his/her parent/guardian providing consent. Enrolling in the study does not require eligibility for or consent to HIV testing. Those who choose to participate in the study and are eligible for HIV testing will be subsequently offered testing and may consent or decline. Those who choose to participate in the study and report conditions that render them ineligible for HIV testing, such as already known to be a PLWH, testing negative within the last three months, or currently in TB treatment, will not be offered HIV testing but will be eligible to participate in other study procedures. Variable block randomization will be done at the level of the household and will be performed at the time of household enrollment. Block sizes will have a minimum of 4 households, a maximum of 8 households. The investigators will utilize Study Randomizer, an online randomization tool with concealed allocation, to generate the allocation sequence. When a CHW determines that an index patient is eligible for the study, and after the index patient or guardian has provided verbal informed consent, the CHW will place a phone call to the study coordinator. The study coordinator will then enroll the household using the Study Randomizer tool and let the CHW know the study allocation. The CHW will then record the appropriate allocation in the survey software, along with the randomization ID, and contact the appropriate CHW team for the household visit. The index patient will be given the name of the community health worker who will be visiting the household, and immediately connected with that person by telephone to arrange the home visit. Households randomized to intervention group (Social Support Group) will be offered the social-behavioral intervention. Households randomized to the standard of care group (Standard of Care) will be offered oral testing without any social-behavioral intervention. CHWs will operate in teams that are always assigned to the same arm of the study. There will be three teams of CHWs in total: one intervention group, one standard of care group, and one clinic-based group that will always carry out initial enrollment of index patients and record any clinic follow-up by individuals in either arm of the study. Regardless of intervention or control arm, CHWs will deliver a short, 13-item TB and HIV stigma scale to all household contacts at the beginning (prior to TB evaluation) and end of the household visit (after all TB and HIV procedures). The scale was adapted from the Van Rie TB HIV stigma scale specifically for this context. The scales will be administered first at the beginning of the home visit, then again at its conclusion. The primary outcome is uptake of HIV testing, defined as the proportion of eligible individuals in the household who undergo testing after a test offer compared between the intervention arm and the standard of care arm..

There are 3 teams of community health workers (CHWs) in total: one intervention, one standard of care, and one clinic-based that will always carry out initial enrollment and record clinic follow-up for both arms. Blinding of the CHWs is not feasible because it is a behavioral intervention that is easily recognized and CHWs must be trained to deliver either the standard of care or the intervention strategy. CHWs will collect data for the outcomes. CHWs will be separated into two teams and will only have electronic permissions to access training, procedures, forms, and contact information related to their assigned allocation within the electronic case record forms. They will be blinded to randomization procedures and outcomes. Participants will be blinded to their randomization assignment. Investigators and study staff will be blinded to study outcomes until the database is locked at the end of the trial. The onsite study coordinator who manages randomization will not be blinded.

CHWs will offer HIV testing to the individual nominated by the index patient as most likely to test. If this person is not present, CHWs will decide which contact should be offered testing first. CHWs will use a prosocial script for HIV testing: "Knowing your status sets a good example for your household." CHWs will follow an "opt-out" strategy: "This test kit is approved by the Ministry of Health and used in KCCA health facilities. I am going to offer you a free HIV test now, is that okay?". If the initial household contact who is offered HIV testing agrees to test, the CHW will ask if he/she is willing to share his/her decision to test with other members of the household: "Would you like to share your decision to test with the others? Sharing is completely optional. However, learning that someone else in their household decided to test sometimes gives people the strength to test themselves. Sharing your decision might help another person find the strength to test."

As a control for the socio-behavioral intervention, the control arm will lack the socio-behavioral intervention components. The order of testing invitation will be decided by the CHW; and CHWs will be trained at baseline to provide standard, opt-in framing of test offers, without any mention of asking contacts to share their testing decision with other household contacts. Oral HIV kits will also be used in control households.

The intervention to be evaluated in this study is a multidimensional, socio-behavioral "Norming" intervention. The various components of this intervention are designed to harness household dynamics and prosocial inclinations to encourage individuals to accept the HIV testing invitation. Each component is briefly described in the arm description.

To test the hypothesis that a norming intervention can increase test uptake, the investigators will compare testing uptake among the intervention households and control households using cluster-adjusted chi-squared tests of proportion and by fitting mixed effects logistic regression models with two levels (household, contact). The effect of the intervention will then be evaluated by comparing adjusted slope differences in the primary outcomes (test uptake) between the intervention and the control arms, using random intercept mixed logit models to account for clustering by household.

Change in HIV Stigma will be evaluated by measuring the change in HIV-TB stigma as measured by the Van Rie paired HIV-TB Stigma Scales (adapted for use in Uganda) using multilevel models. The investigators will use mixed effects Poisson regression models to evaluate changes in HIV stigma measured 3-months after the household visit among a two-stage,stratified random sample of participants.

Change in TB Stigma will be evaluated by measuring the change in perceived TB stigma as measured by the Van Rie paired HIV-TB Stigma Scales (adapted for use in Uganda) using multilevel models. The investigators will use mixed effects Poisson regression models to evaluate changes in TB stigma measured 3-months after the household visit among a two-stage,stratified random sample of participants.

The investigators will conduct a causal mediation analysis to determine the degree to which the effects of the intervention on HIV stigma explain the improvement in HIV test uptake using observed-variable structural equation modeling. HIV stigma will be measured using the Van Rie paired HIV-TB Stigma Scales (adapted for use in Uganda).

The investigators will conduct a causal mediation analysis to determine the degree to which the effects of the intervention on TB stigma explain the improvement in HIV test uptake using observed-variable structural equation modeling. TB stigma will be measured using the Van Rie paired HIV-TB Stigma Scales (adapted for use in Uganda).

The investigators will measure the proportion of testing preference predictions made by the index patient about their household contacts that were accurate, that is, the proportion of household members who accepted the HIV test among those nominated by the index patient as being most likely to accept the HIV test. The proportion of first testers who accepted the HIV test in the intervention arm will be compared to the proportion of first testers who took the HIV test in the control arm.

Relative risk of first testers' decision to disclose testing decision on subsequent household members decision to test for HIV

The investigators will estimate in the intervention arm the effect of first testers' decision to disclose on subsequent testers' decisions to test in the intervention arm using a logistic regression model.

Proportion of social influence nominations by the index patient that align with nominations by the household contacts

The investigators will assess the concordance of index patients' perceptions of their household contacts' influence on health decisions within the household with the perceptions of the rest of their household. To do so, the investigators will measure the proportion of social influence nominations by the index patient that align with nominations by the household contacts in the intervention arm.

Index Patient and Household Eligibility: Index patient is eligible for contact investigation (TB patient of any age identified in the facility's National TB and Leprosy Programme (NTLP) TB Treatment Register as initiating treatment or returning for 2-week follow-up) Index patient or legal guardian agrees to study procedures in addition to routine contact investigation Index patient has at least 2 self-reported household contacts age 15 or above Household is within the boundaries of the Kampala Capital City Authority, Uganda Does not have multi-drug resistant (MDR) TB Contact Participant Eligibility: Household contact age ≥15 years Agrees to study procedures in addition to routine contact investigation

De-identified will be available upon request after the publication of main trial results. Individuals interested in data must reach out to one of the principal investigators.

Data will be available after the publication of main trial results. Data will be held by the researchers for a minimum of 7 years after the end of the trial.

A research proposal must be submitted to one of the principal investigators outlining the type of data needed as well as intended use of data.

Armstrong-Hough M, Ggita J, Gupta AJ, Shelby T, Nangendo J, Ayen DO, Davis JL, Katamba A. Assessing a norming intervention to promote acceptance of HIV testing and reduce stigma during household tuberculosis contact investigation: protocol for a cluster-randomised trial. BMJ Open. 2022 May 25;12(5):e061508. doi: 10.1136/bmjopen-2022-061508.