A Randomized Controlled Trial of Ethyl Glucuronide-Based Contingency Management for Outpatients With Co-Occurring Alcohol Use Disorders and Serious Mental Illness

Abstract

Objective To determine if a contingency management intervention using the ethyl glucuronide (EtG) alcohol biomarker resulted in increased alcohol abstinence in outpatients with co-occurring serious mental illnesses. Secondary objectives were to determine if contingency management was associated with changes in heavy drinking, treatment attendance, drug use, cigarette smoking, psychiatric symptoms, and HIV-risk behavior. Method Seventy-nine (37% female, 44% non-white) outpatients with serious mental illness and alcohol dependence receiving treatment as usual completed a 4-week observation period and were randomized to 12-weeks of contingency management for EtG-negative urine samples and addiction treatment attendance, or reinforcement only for study participation. Contingency management included the variable magnitude of reinforcement “prize draw” procedure contingent on EtG-negative samples (<150 ng/mL) three times a week and weekly gift cards for outpatient treatment attendance. Urine EtG, drug test, and self-report outcomes were assessed during the 12-week intervention and 3-month follow-up periods. Results Contingency management participants were 3.1 times (95% CI: 2.2, 4.5) more likely to submit an EtG-negative urine test during the 12-week intervention period, attaining nearly 1.5 weeks of additional abstinence relative to controls. Contingency management participants had significantly lower mean EtG levels, reported less drinking and fewer heavy drinking episodes, and were more likely to submit stimulant-negative urine and smoking-negative breath samples, relative to controls. Differences in self-reported alcohol use were maintained at the 3-month follow-up. Conclusions This is the first randomized trial utilizing an accurate and validated biomarker (EtG) to demonstrate the efficacy of contingency management for alcohol dependence in outpatients with serious mental illness.

Using a randomized controlled trial to test whether modifications to contingency management improve outcomes for heavy drinkers with serious mental illness

“…1). Although our published research demonstrates that uEtG > 499 ng/mL is associated with poor treatment response [20], unpublished analyses suggested that a lower cut-off of uEtG > 349 ng/mL predicts poor treatment response similar to a uEtG cut-off > 499 ng/ mL. The use of a lower cut-off of uEtG > 349 ng/mL cut-off will allow increase the number of potential participants randomized into the 3 CM conditions during the treatment phase.…”

“…However, variables that differ between groups at baseline and during the induction phase will be used as covariates in subsequent analyses, as will baseline measures that co-vary with the outcome measures. This is a strategy of preliminary analysis that we and several other teams in the area of CM and addiction treatment have utilized for several decades [20,21,23,50–53]. …”

“…In a previous study of CM as a treatment for AUD in adults with SMI, we used the alcohol biomarker, urine ethyl glucuronide urine (uEtG) to assess abstinence. uEtG can detect use during the previous 2 days and heavy drinking up to 5 days after drinking [20,22,23]. CM participants were 3.1 times more likely to submit negative uEtG samples, relative to those receiving treatment-as-usual (TAU) and reinforcers for participation only [20].…”

“…uEtG can detect use during the previous 2 days and heavy drinking up to 5 days after drinking [20,22,23]. CM participants were 3.1 times more likely to submit negative uEtG samples, relative to those receiving treatment-as-usual (TAU) and reinforcers for participation only [20]. However, participants with a pre-treatment uEtG > 499 ng/mL (i.e., daily heavy drinking) did not respond to CM [20].…”

The American Psychiatric Association Practice Guideline for the Pharmacological Treatment of Patients With Alcohol Use Disorder

“…In addition, some (Harasymiw and Bean 2007;Wetterling et al 2014) but not all (Bertholet et al 2014;Liangpunsakul et al 2010) studies suggest that physiological biomarkers can supplement patient self-report in identifying alcohol use in community samples, primary care, and other medical settings. Research also suggests that physiological biomarkers can be used to identify relapse to drinking (Mundle et al 1999) and to promote abstinence (McDonell et al 2017) or to demonstrate risk for alcohol-related behaviors such as driving while intoxicated (Maenhout et al 2014;Marques et al 2010) or health complications after liver transplant (Kollmann et al 2016;Piano et al 2014;Staufer et al 2011). Additional information on the rationale for using physiological biomarkers in the management of individuals with AUD can be found in the advisory from the Substance Abuse and Mental Health Services Administration (2012).…”