Utility of self-report antiretroviral adherence for predicting HIV viral load among persons who inject drugs in Hai Phong Vietnam: assessing differences by methamphetamine use.

ABSTRACTIn resource-limited settings, alternatives to HIV viral load testing may be necessary to monitor the health of people living with HIV. We assessed the utility of self-report antiretroviral therapy (ART) to screen for HIV viral load among persons who inject drugs in Hai Phong Vietnam, and consider differences by recent methamphetamine use. From 2016 to 2018 we recruited PWID through cross sectional surveys and collected self-report ART adherence and HIV viral load to estimate sensitivity, specificity, positive and negative predictive values (PPV, NPV) and likelihood ratios (LR+, LR-) for self-reported ART adherence as a screening test for HIV viral load. We used three HIV viral load thresholds: < 1000, 500 and 250 copies/mL; laboratory-confirmed HIV viral load was the gold standard. Among 792 PWID recruited, PPV remained above 90% regardless of recent methamphetamine use with slightly higher PPV among those not reporting recent methamphetamine use. The results remained consistent across all three HIV viral load thresholds. Our findings suggest that when HIV viral load testing is not possible, self-reported ART adherence may inform decisions about how to prioritize HIV viral load testing among PWID. The high PPV values suggest self-reported high ART adherence indicates likely HIV viral suppression, irrespective of methamphetamine use.

Development and assessment of a community-based screening tool for mental health disorders among people who inject drugs.

INTRODUCTION: The prevalence of mental health disorders among people who use drugs is high and well documented. This hard-to-reach population faces a very low awareness and access to mental health care, especially in developing countries. The objectives of this study were to design and assess a quick screening tool (QST) that community-based organisations (CBO) could routinely apply to a Vietnamese population of people who inject drugs (PWID), in order to refer them appropriately to mental health specialists. METHODS: We devised a tool that included nine questions covering anxiety, depression, suicide risk and psychotic symptomatology. Its use required no specific background and 2 h training. Specificity and sensitivity of the QST were assessed in a population of 418 PWID recruited via respondent driven sampling, using the Mini International Neuropsychiatric Interview questionnaire plus clinical evaluation as a reference standard. Acceptability was assessed using a self-administered anonymous questionnaire submitted to all CBO members who used the QST. RESULTS: CBO members considered the QST easy to use, relevant and helpful to deal with mental health issues. Area under the curve for detection of any symptom using the QST was 0.770. The maximum sensitivity and specificity were reached with a cut-off of 2 [sensitivity was 71.1% (95% confidence interval 62.4, 78.8), specificity was 75.9% (70.5, 80.7)]. DISCUSSION AND CONCLUSIONS: The QST appeared to be both efficient and well accepted. Given the burden of mental health problems among hard-to-reach PWID in developing countries, community-based screenings such as this one could be a particularly appropriate response.

Associations between methamphetamine use and lack of viral suppression among a cohort of HIV-positive persons who inject drugs in Hai Phong, Vietnam.

OBJECTIVE: We assessed the association between methamphetamine use and lack of viral suppression among a cohort of HIV-seropositive persons who inject drugs (PWID) in Hai Phong, Vietnam. DESIGN: Cohort study with random effects logit modeling and mediation analysis for antiretroviral therapy (ART) adherence. METHODS: PWID were recruited from October 2016 to October 2017; HIV-seropositive PWID were enrolled in a cohort to assess HIV viral loads, changes in drug use, risk behaviors, and ART adherence during 24-month follow-up. Methamphetamine use in last 30 days was divided into three categories: 0 days (no use), 1-19 days (intermediate), and 20 or more days (heavy). Bivariate and a multivariable random effects logit models were used to assess the relationship between methamphetamine use and not being virally suppressed. We also assessed self-reported ART adherence as a mediating factor. RESULTS: A total of 645 HIV-seropositive PWID were included at baseline; 95% male, average age 40 (SD = 6.4). At baseline, methamphetamine use in last 30 days was 64% no use, 32% intermediate use, 4% heavy use. Approximately 74% of PWID reported high/complete adherence; 76% were at viral suppression. In random effects analysis, recent methamphetamine use was associated with not being virally suppressed during follow-up (adjusted odds ratio: 1.84, 95% confidence interval: 1.06, 3.17); the effect was not explained by a mediating effect of self-reported adherence to ART. CONCLUSION: Recent methamphetamine use is associated with not being virally suppressed among PWID. The results of this study indicate the need for targeted interventions for methamphetamine use with special focus on those with HIV infection.