Health Behaviors, HIV/AIDS
Conditions
Brief summary
HIV antiretroviral therapy has the potential to dramatically decrease HIV transmission worldwide1; yet, a barrier to ending the AIDS epidemic in low-resource settings is the fact that healthcare is largely provided by traditional or spiritual healers rather than biomedical providers, and there are no strategies in place to identify HIV-infected patients among Traditional Healer patients and link them to HIV care. In order to reach the UNAIDS 90-90-90 benchmarks HIV services must reach marginalized populations in endemic regions, such as in southwestern Uganda. Uganda is one of seven sub-Saharan African (SSA) countries accounting for 90% of all new HIV infections in this region6. HIV prevalence is 7.3%, with \ 1.5 million people living with HIV/AIDS and 99,000 new infections in 2014. However, only 50% of sexually active Ugandans have ever tested for HIV8. In the project location of southwestern Uganda, like much of SSA, the majority of Ugandans utilize Traditional Healers (TH), but little is known about Traditional Healer practices or rates of HIV testing (or HIV infection) among their clients. Specific aims of this study are to: 1) identify key socio-structural factors that frame HIV testing behaviors among Ugandan adults who utilize Traditional Healers; 2) investigate acceptability of providing point-of-care HIV testing at Traditional Healer practice locations; and 3) develop and pilot a prospective HIV testing intervention among Traditional Healer patients to promote earlier diagnosis. Results will be used to implement subsequent, large-scale cluster-randomized HIV testing intervention at Traditional Healer practice locations. Findings from the proposed study include formative data on populations that utilize Traditional Healers in an HIV-endemic region of Uganda, and pilot testing of an HIV testing intervention at healer practice locations; these results could be applied towards expanding HIV testing in other low-resource, endemic settings.
Detailed description
This study has three specific aims: Specific Aim 1: Identify key socio-structural factors that frame HIV testing behaviors among clients of Ugandan TH, using an exploratory sequential, mixed-methods study design. First, qualitative interviews with 25-35 purposely-sampled TH clients will characterize the care-seeking trajectory, HIV testing behaviors and related stigma, healthcare costs, and HIV risks and exposures. Based on these results, a survey will be designed and administered to 300 Ugandan adults who utilize TH in Mbarara District. This cross-sectional study will identify 1) rates of HIV testing among TH clients and 2) variables that independently predict HIV testing within the prior 12 months. These data will inform a model of HIV testing behavior among TH clients based on the Andersen Model of Healthcare Utilization, and generate hypotheses to be tested in Specific Aim 2. Specific Aim 2: Characterize acceptability of receiving point-of-care (POC) HIV testing at TH locations, using an exploratory sequential, mixed-methods approach. Qualitative interviews with 15-25 purposely sampled TH and 15-25 purposely-sampled TH clients will explore attitudes and experiences with biomedicine, feasibility and acceptability of delivering POC HIV testing at TH locations, and investigate relevant themes emerging from Aim 1. Based on qualitative results, a survey will be developed and administered to 175 TH in Mbarara District. This cross-sectional study will characterize TH practices relevant to HIV testing, and identify characteristics of healers who favorably view the proposed intervention (Specific Aim 3). These data will be integrated with results from SA1 to develop an intervention that will overcome socio-structural barriers to HIV testing. Specific Aim 3: Develop and pilot an HIV testing intervention among TH clients using a cluster randomized study design. Integration of results from SA1 and SA2 will inform a pilot HIV testing intervention at 9 TH practice locations, to be compared with a control group receiving usual TH care at 8 practices. This pilot will offer HIV testing at TH practice sites to 250 TH clients at the intervention arm sites, and offer protocoled usual care to 250 clients at control arm sites. Primary outcome for this study will be rates of HIV testing among TH clients. Secondary outcomes will include i) number of new HIV diagnoses among TH clients, and ii) number of patients with +HIV POC test who successfully link to HIV care in 3 months.
Interventions
BEHAVIORALHIV testing at traditional healer practices
HIV 1/2 antigen-antibody POC test (Oraquick©) will be administered to those participants who agree to test. Pre-test counseling will be performed before results are delivered. Clients with positive tests will be provided with detailed contact information for the MUST ISS clinic, and given specific instructions to present to the clinic as soon as possible for confirmatory testing and linkage to care.
Sponsors
Mbarara University of Science and Technology
National Institute of Mental Health (NIMH)
Weill Medical College of Cornell University
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
HEALTH_SERVICES_RESEARCH
Masking
NONE
Intervention model description
This is a mixed methods study where Aims 1 and 2 have both qualitative and quantitative arms. Aim 3 is an unblinded intervention, with a control arm, to be delivered in cluster-randomized fashion (Sept 2019-June 2020). Healers will be treated as clusters and randomized to either intervention or control arm.
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
Yes
Inclusion criteria
* 18 year of age or older * able to provide informed consent * not known to be HIV infected * willing to be contacted at 3 months following enrollment * willing to complete an exit survey after 3 months
Exclusion criteria
* being under the age of 18 years * incapable of giving informed consent * previously being diagnosed with HIV * being unwilling to receive HIV test results * unwilling to participate in the testing intervention
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Number of TH Clients Who Received HIV Testing | within 3 months following study visit | Primary study outcome was receipt of an HIV test within 90 days of enrollment. In the control arm, this was assessed via self-report at time of 90-day phone call follow up. For intervention arm, the participating healer recorded whether an HIV test was accepted by the participant at the time of the study visit. If the HIV test was accepted and delivered, the healer recorded the result of the test. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Number of New HIV Diagnoses Among TH Clients | within three months following enrollment. | For intervention arm participants, the healer recorded the result of the point-of-care HIV test for those participants who agreed to test during the study visit. In the control arm, this outcomes was assessed via self-report during the 90-day follow up phone call. |
| Number of Patients With +HIV Test Who Successfully Link to HIV Care | at three months following enrollment. | This outcome was assessed via self-report during the 90-day follow up phone call. For participants newly testing HIV-positive, questions were posed to determine if they had received confirmatory testing, and initiated ART medication. |
| Age in Years of Control Arm Subjects Who Received an HIV Test | 3 months following study enrollment | Participants in the control arm assessed for primary outcomes of receiving an HIV test within 3 months of study enrollment. Participant age in years at time of study enrollment, and association with primary outcome of receiving an HIV test |
| Gender of Control Arm Subjects Who Received an HIV Test | at 3 months following enrollment | Self-reported participant gender and association with primary study outcome of receiving an HIV test |
| Highest Level of Education for Control Arm Subjects Who Received an HIV Test | at 3 months following study enrollment | Participant self-reported highest level of education, and association with study primary outcome of receiving an HIV test |
Countries
Uganda
Participant flow
Recruitment details
25 traditional healers practicing within an eight-kilometer distance from Mbarara Township were screened for eligibility.
Pre-assignment details
Five healers were excluded due to self-reported weekly volume \<7 patients. Three healers declined to participate because of scheduling conflicts. On August 2, 2019 the remaining 17 healers were randomized into two study arms (9 to intervention, 8 to control. From August 8, 2019-February 7, 2020, 500 participants were enrolled at healer cluster sites
Participants by arm
| Arm | Count |
|---|---|
| Healer Control Arm Patients will undergo protcolized usual care at 8 healer practices, which include HIV education and a referral to receive VCT through local HIV testing clinics. Study staff will contact the client at 3 months following enrollment to assess for self-report of VCT. | 250 |
| Healer HIV Testing Intervention HIV testing at traditional healer practices: HIV 1/2 antigen-antibody POC test (Oraquick©) will be administered to those participants who agree to test. Pre-test counseling will be performed before results are delivered. | 250 |
| Total | 500 |
Baseline characteristics
| Characteristic | Healer HIV Testing Intervention | Total | Healer Control Arm |
|---|---|---|---|
| Age, Continuous | 28.5 years | 29 years | 30 years |
| Number of participants who are Christian | 237 Participants | 455 Participants | 218 Participants |
| Number of participants who are married | 142 Participants | 306 Participants | 164 Participants |
| Number of participants who completed primary education or lower | 143 Participants | 282 Participants | 139 Participants |
| Race and Ethnicity Not Collected | — | 0 Participants | — |
| Sex: Female, Male Female | 134 Participants | 279 Participants | 145 Participants |
| Sex: Female, Male Male | 116 Participants | 221 Participants | 105 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | 0 / 250 | 0 / 250 |
| other Total, other adverse events | 0 / 250 | 0 / 250 |
| serious Total, serious adverse events | 0 / 250 | 0 / 250 |
Outcome results
Primary
Number of TH Clients Who Received HIV Testing
Primary study outcome was receipt of an HIV test within 90 days of enrollment. In the control arm, this was assessed via self-report at time of 90-day phone call follow up. For intervention arm, the participating healer recorded whether an HIV test was accepted by the participant at the time of the study visit. If the HIV test was accepted and delivered, the healer recorded the result of the test.
Time frame: within 3 months following study visit
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Healer Control Arm | Number of TH Clients Who Received HIV Testing | 57 Participants |
| Healer HIV Testing Intervention | Number of TH Clients Who Received HIV Testing | 250 Participants |
Comparison: Given that the intervention arm showed 100% uptake, a multi-level regression model could not be created with a zero in the denominator in the intervention group. We therefore used Fisher's exact test to assess for significant differences in proportion of HIV testing between study arms.p-value: <0.00195% CI: [0.015, 0.363]Fisher Exact
Secondary
Age in Years of Control Arm Subjects Who Received an HIV Test
Participants in the control arm assessed for primary outcomes of receiving an HIV test within 3 months of study enrollment. Participant age in years at time of study enrollment, and association with primary outcome of receiving an HIV test
Time frame: 3 months following study enrollment
Population: Only control arm participants who received an HIV test were included in this analysis, as all intervention arm participants accepted an HIV test as part of the study and therefore, no other variates would be statistically associated with receiving a test more than being in the intervention study arm.
| Arm | Measure | Value (MEDIAN) |
|---|---|---|
| Healer Control Arm | Age in Years of Control Arm Subjects Who Received an HIV Test | 30 years |
Comparison: We considered variables that predicted the primary outcome of receiving an HIV test among control arm only, as uptake was 100% in the intervention. First we identified variables that were significant (two-sided p-value\<0.05) in univariate analysis and creating multivariate mixed-effects logistic regression model accounting for covariates and intercorrelation between participants from the same healer by including healers as random effects.p-value: <0.0195% CI: [1.02, 1.06]Regression, Linear
Secondary
Gender of Control Arm Subjects Who Received an HIV Test
Self-reported participant gender and association with primary study outcome of receiving an HIV test
Time frame: at 3 months following enrollment
Population: Participants in the control arm assessed for primary outcomes of receiving an HIV test within 3 months of study enrollment. Only control arm participants were included in this analysis, as all intervention arm participants accepted an HIV test as part of the study and therefore, no other variates would be statistically associated with receiving a test more than being in the intervention study arm.
| Arm | Measure | Category | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|---|
| Healer Control Arm | Gender of Control Arm Subjects Who Received an HIV Test | Males | 17 Participants |
| Healer Control Arm | Gender of Control Arm Subjects Who Received an HIV Test | Females | 40 Participants |
Comparison: We considered variables that predicted the primary outcome of receiving an HIV test among control arm only, as uptake was 100% in the intervention. First we identified variables that were significant (two-sided p-value\<0.05) in univariate analysis and creating multivariate mixed-effects logistic regression model accounting for covariates and intercorrelation between participants from the same healer by including healers as random effects.p-value: 0.0795% CI: [0.3, 1.04]Fisher Exact
Secondary
Highest Level of Education for Control Arm Subjects Who Received an HIV Test
Participant self-reported highest level of education, and association with study primary outcome of receiving an HIV test
Time frame: at 3 months following study enrollment
Population: We considered variables that predicted the primary outcome of receiving an HIV test among control arm only, as uptake was 100% in the intervention. First we identified variables that were significant (two-sided p-value\<0.05) in univariate analysis and creating multivariate mixed-effects logistic regression model accounting for covariates and intercorrelation between participants from the same healer by including healers as random effects.
| Arm | Measure | Category | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|---|
| Healer Control Arm | Highest Level of Education for Control Arm Subjects Who Received an HIV Test | Primary school education or lower | 30 Participants |
| Healer Control Arm | Highest Level of Education for Control Arm Subjects Who Received an HIV Test | Secondary school education or higher | 27 Participants |
Comparison: We considered variables that predicted the primary outcome of receiving an HIV test among control arm only, as uptake was 100% in the intervention. First we identified variables that were significant (two-sided p-value\<0.05) in univariate analysis and creating multivariate mixed-effects logistic regression model accounting for covariates and intercorrelation between participants from the same healer by including healers as random effects.p-value: 0.0995% CI: [0.16, 1.18]Fisher Exact
Secondary
Number of New HIV Diagnoses Among TH Clients
For intervention arm participants, the healer recorded the result of the point-of-care HIV test for those participants who agreed to test during the study visit. In the control arm, this outcomes was assessed via self-report during the 90-day follow up phone call.
Time frame: within three months following enrollment.
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Healer Control Arm | Number of New HIV Diagnoses Among TH Clients | 0 Participants |
| Healer HIV Testing Intervention | Number of New HIV Diagnoses Among TH Clients | 10 Participants |
p-value: 0.002Fisher Exact
Secondary
Number of Patients With +HIV Test Who Successfully Link to HIV Care
This outcome was assessed via self-report during the 90-day follow up phone call. For participants newly testing HIV-positive, questions were posed to determine if they had received confirmatory testing, and initiated ART medication.
Time frame: at three months following enrollment.
Population: Zero participants in the control arm were newly diagnosed with HIV infection.
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Healer Control Arm | Number of Patients With +HIV Test Who Successfully Link to HIV Care | 0 Participants |
| Healer HIV Testing Intervention | Number of Patients With +HIV Test Who Successfully Link to HIV Care | 7 Participants |
p-value: 0.015Fisher Exact