Alcohol-Related Disorders, Drug Use, Human Immunodeficiency Virus
Conditions
Keywords
Immunologic Deficiency Syndromes, Substance Use, Acquired Immunodeficiency Syndrome, Immune System Diseases, Behavioral Symptoms, RNA Virus Infections, HIV Infections, Alcohol Use, Drug Use
Brief summary
The purpose of this study is to test the effectiveness and implementation of a brief, integrated behavioral intervention for HIV medication adherence and substance use in the HIV care setting in South Africa. The intervention is specifically designed to be implemented by non-specialist counselors using a task sharing model in local HIV clinics. The behavioral intervention will be compared to usual care, enhanced with referral to a local outpatient substance use treatment program (Enhanced Standard of Care - ESOC) on study endpoints (as described in study endpoint section below).
Detailed description
The HIV epidemic in South Africa (SA) is among the highest in the world. SA has a large antiretroviral therapy (ART) program, but some individuals exhibit poor ART adherence, which increases the likelihood of developing drug resistance and failing the only available first and second line ART regimens in SA. ART nonadherence contributes to greater morbidity, mortality, and higher likelihood of sexual HIV transmission when virus is detectable. At the same time, alcohol and other drug use is prevalent among HIV-infected individuals in SA and associated with worse ART adherence, lower rates of viral suppression, and HIV transmission risk behavior. Yet, despite the impact of untreated substance use on poor HIV treatment outcomes and continued HIV transmission, there is little if any integration of substance use and HIV care services in SA, which creates a fragmented and incomplete system of care. This study had three phases, first being formative, qualitative work which led to a systematic treatment adaptation phase. This third phase, the clinical trial, is based on this formative work and other empirical support using behavioral interventions to improve ART adherence and reduce substance use in resource-limited settings, including SA. This study is a Type 1 hybrid effectiveness-implementation trial of a lay counselor-delivered behavioral intervention for adherence and substance use integrated into the HIV primary care setting in SA. To ensure that those who need this intervention most will receive it, participants will be patients with HIV who are struggling with adherence (as defined in the investigator's inclusion criteria) and who have an elevated substance use risk.
Interventions
BEHAVIORALProject Khanya
This treatment involves integrating a behavioral intervention for substance use with a behavioral intervention for adherence.
Sponsors
University of Cape Town
National Institute on Drug Abuse (NIDA)
University of Maryland, College Park
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
TREATMENT
Masking
NONE
Eligibility
Sex/Gender
ALL
Age
18 Years to 65 Years
Healthy volunteers
No
Inclusion criteria
* HIV positive and on ART * 18-65 years of age * Elevated substance use risk (ASSIST score greater than or equal to 4 for drugs or greater than or equal to 11 for alcohol) * Have at least one of the following: 1. Not attained viral suppression from first line ART (VL\>400 copies/mL) 2. On second-line ART treatment 3. Reinitiated first-line treatment within the past three months 4. Had a pharmacy non-refill at least once in the past 3 months
Exclusion criteria
* Inability to provide informed consent or complete procedures in English or isiXhosa * Severe risk/likely dependence for opiates (ASSIST score \>26) because opiate substitution therapy may not be available * Severe alcohol dependence symptoms that may warrant medical management of potential withdrawal symptoms * Active, untreated, major mental illness (with untreated psychosis or mania) that would interfere with the paraprofessional adapted intervention
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Changes in Self-reported Substance Use | Assessed between baseline assessment and the acute outcome (approximately 12-weeks post-randomization/ post-intervention assessment) | World Health Organization Alcohol, Smoking, and Substance Involvement Screening Test (WHO-ASSIST). It is a measure used to assess substance use risk for alcohol, cannabis, cocaine, opiates, and amphetamines, hallucinogens, and other drugs. Standardized cutoff scores are used to categorize risk levels: low risk (0-3 for illicit drugs/0-10 for alcohol), moderate risk (4-26 for illicit drugs/11-26 for alcohol), or high risk (\> 26) for substance use-related problems. |
| Biological Measure of Substance Use | Assessed at the acute outcome (approximately 12-weeks post-randomization/ post-intervention assessment) | Substance use measured with urinalysis. |
| Changes in HIV Medication Adherence Throughout Intervention Phase | Assessed between baseline assessment and the acute outcome (approximately 12-weeks post-randomization/ post-intervention assessment) | Percentage of prescribed antiviral therapy agent (medications) taken as measured by real time wireless motoring device |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Intervention Acceptability | Assessed at the acute outcome (approximately 12-weeks post-randomization/ post-intervention assessment) | 15-item acceptability subscale of a pragmatic, quantitative assessment based on RE-AIM developed by the Applied Mental Health Research group (AMHR) at Johns Hopkins University. Total scores are averaged across all items and range from 0 to 3. Higher scores indicate greater acceptability. Qualitative interviews will also be conducted with intervention participants at the end of the study to assess acceptability guided by RE-AIM and the Proctor model. |
| Intervention Fidelity | Assessed between randomization and the acute outcome (approximately 12-weeks post-randomization/ post-intervention assessment) | Independent fidelity ratings of a randomly selected subset (20%) of intervention sessions using a fidelity assessment developed for each session that includes 15-19 items that map onto each core intervention component, and factors unique to the peer delivery implementation strategy (i.e., appropriate self-disclosure, stigmatizing behaviors, common factors including warmth and non-judgment). |
| Intervention Uptake | Assessed between randomization and the acute outcome (approximately 12-weeks post-randomization/ post-intervention assessment) | Intervention participant attendance and retention (i.e., the mean number of intervention sessions attended by intervention participants) |
| Intervention Feasibility | Assessed at the acute outcome (approximately 12-weeks post-randomization/ post-intervention assessment) | 14-item feasibility subscale of a pragmatic, quantitative assessment based on RE-AIM developed by the Applied Mental Health Research group (AMHR) at Johns Hopkins University. Total scores are averaged across all items and range from 0 to 3. Higher scores indicate greater feasibility. Qualitative interviews will also be conducted with intervention participants at the end of the study to assess feasibility guided by RE-AIM and the Proctor model. |
| Biological Measure of Substance Use | Assessed at follow-up (approximately 24-weeks post-randomization/ 6-month follow-up assessment) | Substance use measured with urinalysis. |
| Changes in Self-reported Substance Use | Assessed between baseline assessment and follow-up (approximately 24-weeks post-randomization/ 6-month follow-up assessment) | World Health Organization Alcohol, Smoking, and Substance Involvement Screening Test (WHO-ASSIST). It is a measure used to assess substance use risk for alcohol, cannabis, cocaine, opiates, and amphetamines, hallucinogens, and other drugs. Standardized cutoff scores are used to categorize risk levels: low risk (0-3 for illicit drugs/0-10 for alcohol), moderate risk (4-26 for illicit drugs/11-26 for alcohol), or high risk (\> 26) for substance use-related problems. |
Other
| Measure | Time frame | Description |
|---|---|---|
| HIV Viral Load | Assessed at follow-up (approximately 24-weeks post-randomization/ 6-month follow-up assessment) | Percentage of patients with a suppressed viral load (\<400 copies/ml) |
| Changes in Self-reported Substance Use | Assessed between baseline assessment and follow-up (approximately 24-weeks post-randomization/ 6-month follow-up assessment) | Changes in percent days used any substance measured by timeline follow-back |
Countries
South Africa, United States
Participant flow
Participants by arm
| Arm | Count |
|---|---|
| Project Khanya Those assigned to Project Khanya (the behavioral intervention for substance use and adherence condition) will have approximately 6 sessions (including Life-Steps, behavioral activation, and relapse prevention) delivered by a peer interventionist plus standard of care, which is typically referral to a local outpatient substance use treatment clinic. They will also receive a Wisepill, a wireless, real-time adherence monitoring device.
Project Khanya: This treatment involves integrating a behavioral intervention for substance use with a behavioral intervention for adherence. | 30 |
| ESOC Those assigned to the ESOC (enhanced standard of care) condition will receive the standard of care, which is referral to a local substance use treatment clinic. The substance use clinics in the location that this study occurs follow the Matrix, and evidence-based 16-week outpatient program to treat substance use. We will enhance patients' normal referral to Matrix for ESOC participants by promoting facilitating and following up on the referral. Additionally, those in the control group will also receive a Wisepill, a wireless adherence monitoring device. | 31 |
| Pre-Randomization Participants enrolled in the study at the baseline assessment, but who were never randomized (approximately 2 weeks post-baseline) and did not continue with the study. | 5 |
| Total | 66 |
Withdrawals & dropouts
| Period | Reason | FG000 | FG001 | FG002 |
|---|---|---|---|---|
| Follow-up Assessment | Death | 1 | 0 | 0 |
| Follow-up Assessment | Withdrawal by Subject | 0 | 1 | 0 |
| Post-treatment Assessment | Death | 1 | 0 | 0 |
| Post-treatment Assessment | Lost to Follow-up | 1 | 1 | 0 |
| Post-treatment Assessment | Unable to contact for this assessment, but attended later assessments | 2 | 4 | 0 |
| Randomized | Excluded | 0 | 0 | 1 |
| Randomized | Lost to Follow-up | 0 | 0 | 4 |
Baseline characteristics
| Characteristic | Project Khanya | ESOC | Pre-Randomization | Total |
|---|---|---|---|---|
| Age, Continuous | 39.80 years STANDARD_DEVIATION 10.47 | 34.29 years STANDARD_DEVIATION 7.99 | 31.60 years STANDARD_DEVIATION 9.56 | 36.59 years STANDARD_DEVIATION 9.66 |
| Consumed any substance on timeline followback | 38.81 percentage of days STANDARD_DEVIATION 24.58 | 29.49 percentage of days STANDARD_DEVIATION 21.72 | 42.86 percentage of days STANDARD_DEVIATION 35.36 | 37.74 percentage of days STANDARD_DEVIATION 24.29 |
| HIV medication adherence | 53.6 percentage of days adherent STANDARD_DEVIATION 32.1 | 49.3 percentage of days adherent STANDARD_DEVIATION 29.8 | — | 46.60 percentage of days adherent STANDARD_DEVIATION 31.56 |
| Moderate or high-risk on Alcohol, Smoking, and Substance Involvement Screening Test (ASSIST) measure | 30 Participants | 31 Participants | 5 Participants | 66 Participants |
| Number of drinks on days drinking on timeline followback | 7.17 number of drinks STANDARD_DEVIATION 3.79 | 8.03 number of drinks STANDARD_DEVIATION 5.59 | 5.37 number of drinks STANDARD_DEVIATION 6.24 | 7.43 number of drinks STANDARD_DEVIATION 4.86 |
| Phosphatidylethanol (PEth) concentration | 686.0 ng/ML STANDARD_DEVIATION 639.9 | 456.1 ng/ML STANDARD_DEVIATION 530.8 | 370.8 ng/ML STANDARD_DEVIATION 418.4 | 554.1 ng/ML STANDARD_DEVIATION 582 |
| Positive alcohol or drug urine test result | 27 Participants | 29 Participants | 4 Participants | 60 Participants |
| Race/Ethnicity, Customized Black African | 29 Participants | 31 Participants | 5 Participants | 65 Participants |
| Race/Ethnicity, Customized Mixed race | 1 Participants | 0 Participants | 0 Participants | 1 Participants |
| Region of Enrollment South Africa | 30 participants | 31 participants | 5 participants | 66 participants |
| Sex: Female, Male Female | 13 Participants | 20 Participants | 3 Participants | 36 Participants |
| Sex: Female, Male Male | 17 Participants | 11 Participants | 2 Participants | 30 Participants |
| Suppressed viral load, <400 copies/ml | 15 Participants | 24 Participants | 3 Participants | 42 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk | EG002 affected / at risk |
|---|---|---|---|
| deaths Total, all-cause mortality | 2 / 30 | 0 / 31 | 0 / 5 |
| other Total, other adverse events | 0 / 30 | 0 / 31 | 0 / 5 |
| serious Total, serious adverse events | 8 / 30 | 0 / 31 | 1 / 5 |
Outcome results
Primary
Biological Measure of Substance Use
Substance use measured with urinalysis.
Time frame: Assessed at the acute outcome (approximately 12-weeks post-randomization/ post-intervention assessment)
| Arm | Measure | Category | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|---|
| Project Khanya | Biological Measure of Substance Use | Positive | 25 Participants |
| Project Khanya | Biological Measure of Substance Use | Negative | 1 Participants |
| ESOC | Biological Measure of Substance Use | Positive | 23 Participants |
| ESOC | Biological Measure of Substance Use | Negative | 3 Participants |
p-value: 0.2895% CI: [-1.61, 5.5]Mixed Models Analysis
Primary
Biological Measure of Substance Use
Substance use measured with phosphatidylethanol (PEth) concentration, which is an objective biomarker of alcohol use that can detect blood collected up to 21 days after alcohol consumption. Minimum detection value is 8 ng/mL. Higher PEth values indicate greater concentration of alcohol. Values of ≥ 50 ng/mL indicate unhealthy drinking.
Time frame: Assessed at the acute outcome (approximately 12-weeks post-randomization/ post-intervention assessment)
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Project Khanya | Biological Measure of Substance Use | 484.2 ng/mL | Standard Deviation 398.7 |
| ESOC | Biological Measure of Substance Use | 414.7 ng/mL | Standard Deviation 389.6 |
p-value: 0.1695% CI: [-67, 382]Mixed Models Analysis
Primary
Changes in HIV Medication Adherence Throughout Intervention Phase
Percentage of prescribed antiviral therapy agent (medications) taken as measured by real time wireless motoring device
Time frame: Assessed between baseline assessment and the acute outcome (approximately 12-weeks post-randomization/ post-intervention assessment)
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Project Khanya | Changes in HIV Medication Adherence Throughout Intervention Phase | 60.0 percentage of days adherent to Wisepill | Standard Deviation 37.1 |
| ESOC | Changes in HIV Medication Adherence Throughout Intervention Phase | 28.2 percentage of days adherent to Wisepill | Standard Deviation 32.1 |
p-value: 0.0195% CI: [-0.507, -0.066]Mixed Models Analysis
Primary
Changes in Self-reported Substance Use
World Health Organization Alcohol, Smoking, and Substance Involvement Screening Test (WHO-ASSIST). It is a measure used to assess substance use risk for alcohol, cannabis, cocaine, opiates, and amphetamines, hallucinogens, and other drugs. Standardized cutoff scores are used to categorize risk levels: low risk (0-3 for illicit drugs/0-10 for alcohol), moderate risk (4-26 for illicit drugs/11-26 for alcohol), or high risk (\> 26) for substance use-related problems.
Time frame: Assessed between baseline assessment and the acute outcome (approximately 12-weeks post-randomization/ post-intervention assessment)
| Arm | Measure | Category | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|---|
| Project Khanya | Changes in Self-reported Substance Use | moderate or high risk | 23 Participants |
| Project Khanya | Changes in Self-reported Substance Use | low risk | 3 Participants |
| ESOC | Changes in Self-reported Substance Use | moderate or high risk | 25 Participants |
| ESOC | Changes in Self-reported Substance Use | low risk | 1 Participants |
p-value: 0.7795% CI: [-1.51, 2]Mixed Models Analysis
Secondary
Biological Measure of Substance Use
Substance use measured with phosphatidylethanol (PEth) concentration, which is an objective biomarker of alcohol use that can detect blood collected up to 21 days after alcohol consumption. Minimum detection value is 8 ng/mL. Higher PEth values indicate greater concentration of alcohol. Values of ≥ 50 ng/mL indicate unhealthy drinking.
Time frame: Assessed at follow-up (approximately 24-weeks post-randomization/ 6-month follow-up assessment)
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Project Khanya | Biological Measure of Substance Use | 538.4 ng/mL | Standard Deviation 554.4 |
| ESOC | Biological Measure of Substance Use | 386.1 ng/mL | Standard Deviation 392.6 |
p-value: 0.4495% CI: [-136, 304]Mixed Models Analysis
Secondary
Biological Measure of Substance Use
Substance use measured with urinalysis.
Time frame: Assessed at follow-up (approximately 24-weeks post-randomization/ 6-month follow-up assessment)
| Arm | Measure | Category | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|---|
| Project Khanya | Biological Measure of Substance Use | Positive | 20 Participants |
| Project Khanya | Biological Measure of Substance Use | Negative | 7 Participants |
| ESOC | Biological Measure of Substance Use | Positive | 24 Participants |
| ESOC | Biological Measure of Substance Use | Negative | 5 Participants |
p-value: 0.6395% CI: [-2.24, 3.67]Mixed Models Analysis
Secondary
Changes in Self-reported Substance Use
World Health Organization Alcohol, Smoking, and Substance Involvement Screening Test (WHO-ASSIST). It is a measure used to assess substance use risk for alcohol, cannabis, cocaine, opiates, and amphetamines, hallucinogens, and other drugs. Standardized cutoff scores are used to categorize risk levels: low risk (0-3 for illicit drugs/0-10 for alcohol), moderate risk (4-26 for illicit drugs/11-26 for alcohol), or high risk (\> 26) for substance use-related problems.
Time frame: Assessed between baseline assessment and follow-up (approximately 24-weeks post-randomization/ 6-month follow-up assessment)
| Arm | Measure | Category | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|---|
| Project Khanya | Changes in Self-reported Substance Use | moderate or high risk | 24 Participants |
| Project Khanya | Changes in Self-reported Substance Use | low risk | 3 Participants |
| ESOC | Changes in Self-reported Substance Use | moderate or high risk | 22 Participants |
| ESOC | Changes in Self-reported Substance Use | low risk | 7 Participants |
p-value: 0.8795% CI: [-1.85, 1.58]Mixed Models Analysis
Secondary
Intervention Acceptability
15-item acceptability subscale of a pragmatic, quantitative assessment based on RE-AIM developed by the Applied Mental Health Research group (AMHR) at Johns Hopkins University. Total scores are averaged across all items and range from 0 to 3. Higher scores indicate greater acceptability. Qualitative interviews will also be conducted with intervention participants at the end of the study to assess acceptability guided by RE-AIM and the Proctor model.
Time frame: Assessed at the acute outcome (approximately 12-weeks post-randomization/ post-intervention assessment)
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Project Khanya | Intervention Acceptability | 2.98 average score on a scale | Standard Deviation 0.04 |
Secondary
Intervention Feasibility
14-item feasibility subscale of a pragmatic, quantitative assessment based on RE-AIM developed by the Applied Mental Health Research group (AMHR) at Johns Hopkins University. Total scores are averaged across all items and range from 0 to 3. Higher scores indicate greater feasibility. Qualitative interviews will also be conducted with intervention participants at the end of the study to assess feasibility guided by RE-AIM and the Proctor model.
Time frame: Assessed at the acute outcome (approximately 12-weeks post-randomization/ post-intervention assessment)
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Project Khanya | Intervention Feasibility | 2.98 average score on a scale | Standard Deviation 0.18 |
Secondary
Intervention Fidelity
Independent fidelity ratings of a randomly selected subset (20%) of intervention sessions using a fidelity assessment developed for each session that includes 15-19 items that map onto each core intervention component, and factors unique to the peer delivery implementation strategy (i.e., appropriate self-disclosure, stigmatizing behaviors, common factors including warmth and non-judgment).
Time frame: Assessed between randomization and the acute outcome (approximately 12-weeks post-randomization/ post-intervention assessment)
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Project Khanya | Intervention Fidelity | 91.7 Percentage fidelity to treatment | Standard Deviation 13.3 |
Secondary
Intervention Uptake
Intervention participant attendance and retention (i.e., the mean number of intervention sessions attended by intervention participants)
Time frame: Assessed between randomization and the acute outcome (approximately 12-weeks post-randomization/ post-intervention assessment)
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Project Khanya | Intervention Uptake | 4.77 Sessions attended | Standard Deviation 1.96 |
Other Pre-specified
Changes in Self-reported Substance Use
Changes in percent days used any substance measured by timeline follow-back
Time frame: Assessed between baseline assessment and follow-up (approximately 24-weeks post-randomization/ 6-month follow-up assessment)
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Project Khanya | Changes in Self-reported Substance Use | 30.7 percentage of days used any substance | Standard Deviation 25.9 |
| ESOC | Changes in Self-reported Substance Use | 25.1 percentage of days used any substance | Standard Deviation 23.2 |
p-value: 0.6395% CI: [-10.5, 17.1]Mixed Models Analysis
Other Pre-specified
Changes in Self-reported Substance Use
Changes in number of drinks measured by timeline follow-back
Time frame: Assessed between baseline assessment and follow-up (approximately 24-weeks post-randomization/ 6-month follow-up assessment)
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Project Khanya | Changes in Self-reported Substance Use | 4.61 number of drinks on days drinking | Standard Deviation 3.29 |
| ESOC | Changes in Self-reported Substance Use | 4.96 number of drinks on days drinking | Standard Deviation 4.25 |
p-value: 0.8595% CI: [-0.34, 0.28]Mixed Models Analysis
Other Pre-specified
HIV Viral Load
Percentage of patients with a suppressed viral load (\<400 copies/ml)
Time frame: Assessed at follow-up (approximately 24-weeks post-randomization/ 6-month follow-up assessment)
| Arm | Measure | Category | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|---|
| Project Khanya | HIV Viral Load | Suppressed | 16 Participants |
| Project Khanya | HIV Viral Load | Unsuppressed | 11 Participants |
| ESOC | HIV Viral Load | Suppressed | 22 Participants |
| ESOC | HIV Viral Load | Unsuppressed | 7 Participants |
p-value: 0.6595% CI: [-2.49, 1.55]Mixed Models Analysis