Enhancing patient navigation with contingent financial incentives for substance use abatement in persons with HIV and substance use

Abstract

Substance use can interfere with HIV treatment. A previous multisite clinical trial (Metsch et al., 2016) tested 2 behavioral interventions designed to improve treatment engagement in people with comorbid HIV and drug or heavy alcohol use. Clinical trial participants were randomized to treatment as usual (N = 264), patient navigation (PN; N = 266), or PN with contingency management (PN + CM; N = 271) for 6 months. PN + CM patients could earn financial incentives both for entering substance use disorder (SUD) treatment and for submitting urine and breath samples negative for opioids, stimulants, and alcohol. This secondary analysis compared frequencies of treatment entry and sample submission in the PN versus PN + CM groups and examined associations with viral suppression (defined as ≤200 copies/mL). Incentives were associated with a higher percentage of patients entering SUD treatment (PN = 25.5%; PN + CM = 47.6%; p < .001), a higher percentage submitting samples for drug testing (PN median = 2, interquartile range [IQR] = 0.5; PN + CM median = 8, IQR = 5.1; p < .0001) and a higher percentage submitting samples negative for targeted drugs and alcohol (PN median = 1, IQR = 0.3; PN + CM median = 6, IQR = 2.9; p < .0001). Within the PN + CM group, up to 58% of those with high rates of engagement in activities were virally suppressed at 6 months versus 24-29% in subgroups with lowest engagement. In conclusion, CM was feasibly incorporated into PN for persons with HIV and SUD and was associated with higher rates of engagement in targeted substance use abatement activities. CM has the potential to improve health outcomes in this population. (PsycINFO Database Record (c) 2020 APA, all rights reserved).

Trial registration: ClinicalTrials.gov NCT01612169.

Figures

Fig. 1.

Individual differences in urine sample submission behavior. All participants were invited to submit urine samples for testing and take a breath alcohol test during sessions (maximum of 11 sessions) with their patient navigator. Percent of participants submitting 0–11 samples during the 6-month intervention is shown separately for PN (grey bars; N=266) and PN+CM (black bars; N=271). Group distributions (PN vs PN+CM) were significantly different (Wilcoxon test p

Fig. 2.

Urine testing outcomes over time. Percent of participants in each group (PN = 266; PN_CM = 271) submitting one or more samples during a given study month that tested negative for 6 drugs targeted in the abstinence incentive program plus negative BAL (triangle symbols) and for all 10 drugs tested plus negative BAL (circular symbols). The ten drugs tested were opiate [morphine], oxycodone, methadone, cocaine, amphetamine, methamphetamine, marijuana, benzodiazepines, ecstasy, and barbiturates. The six target drugs were opiate, oxycodone, methadone (without prescription), cocaine, amphetamine and methamphetamine GEE analysis revealed a significant group and time effect for both targeted and all drugs but no interaction.