VOUCHER-BASED REINFORCEMENT FOR ALCOHOL ABSTINENCE USING THE ETHYL-GLUCURONIDE ALCOHOL BIOMARKER
Introduction
Study on the effects of contingency management (CM) intervention for alcohol consumption.
Conducted by Michael G. McDonell and colleagues.
Target population: 10 alcohol-dependent participants.
Key Details of the Study
Design: ABCA design was implemented.
Methodology: Vouchers were contingent upon ethyl glucuronide (EtG) urine tests (alcohol biomarker with a 2-day detection window) and alcohol breath tests.
Results: Percentage of negative urines over study phases:
First baseline phase: 35%
C phase: 69%
Return-to-baseline phase: 20%
Conclusion: EtG urine tests are a feasible method to deliver CM and promote alcohol abstinence.
Background on Contingency Management (CM)
Definition: CM is an evidence-based approach that provides tangible rewards to reinforce desired behaviors; widely supported for promoting abstinence from illicit drugs (Prendergast et al., 2006).
Drug Testing: Use of low-cost immunoassay urine drug tests that provide immediate results for drug use typically detected up to 2 days prior to the test.
Current Issues in Alcohol CM: Previous findings supporting CM for alcohol use were limited due to breath tests that only detect use up to 12 hours (Warner & Sharma, 2009).
EtG Testing Advantages:
Ethyl glucuronide (EtG) is a metabolite of alcohol; urine detection of EtG can last up to 3 days (Litten et al., 2010).
Increasingly utilized in forensic, employment, and treatment contexts.
Research Methodology
Participants
Sample Size: 15 adults (aged 21-65, M=29, SD=8.8).
Eligibility Criteria:
Must meet criteria for alcohol dependence.
At least four drinking episodes each week (3+ drinks per episode in the prior 30 days).
Demographics:
Majority male (87%, n=13).
Predominantly Caucasian (80%, n=12).
Unemployed (67%, n=10).
Exclusion Criteria:
History of alcohol withdrawal.
Current methamphetamine use.
Unsafe to participate in the study.
Measures
Testing Frequency: Urine and breath samples collected at the initial intake and bi-weekly during 9 weeks.
EtG Testing:
Conducted by American Drug Testing, Spokane, WA, using a Microgenics MGC-240 analyzer.
Cutoff level: 500 ng/ml (Bottcher et al., 2008).
Alcohol Breath Tests:
Positive if BAC > 0.001%.
Self-report: Utilized the Timeline Followback (TLFB) for drinking assessment (Sobell & Sobell, 2000).
Procedure
Approval: Institutional Review Board approved the study.
Recruitment: Online advertisements targeted individuals with problematic alcohol and methamphetamine use history.
Screening Process:
69 individuals responded, 45 screened over the phone, 29 eligible.
15 provided consent and participated in the study procedures at Washington State University.
Study Design
ABCA Design:
A phases: Baseline (2 weeks).
B phase: Warning of upcoming contingencies (1 week).
C phase: CM intervention (4 weeks).
Reinforcement Schedule: Participants received $5 for each visit.
Vouchers based on negative results from urine and breath tests.
Subsequent visits increased voucher value by $2 for consecutive negative results.
Positive samples or no-shows reset voucher value.
Maximum Earnings: Up to $96 for continuous abstinence.
Final Session Requirement: Participants received motivational interviewing and referrals upon completion, and a $30 gift card for completing all visits.
Statistical Analysis
Used Generalized Estimating Equations (GEE) to assess CM effects on alcohol abstinence.
Analyzed outcomes comparing phases B and C against baseline phase.
Missing EtG samples were coded as positive; breath test data not statistically analyzed due to no positives.
TLFB data also analyzed via GEE with alpha set at .05.
Validity Measures: Sensitivity and specificity calculated for EtG-I tests and compared against TLFB as the standard.
Results and Discussion
Submission of Negative Samples:
9/15 participants had alcohol-negative EtG-I samples at intake.
2 participants didn’t complete baseline phase; 3 were abstinent during baseline and sobriety warning phases.
Remaining participants showed significant changes in urine tests across study phases:
Baseline: 35%
B phase: 60%
C phase: 69%
Return-to-baseline: 20%
Odds Ratio Results:
CM phase (C) significantly increased likelihood of negative urine samples compared to baseline (Odds Ratio = 4.71, p < .05).
Self-reports via TLFB:
Abstinence rates: Baseline (37%), Sobriety Warning (65%), CM phase (63%), Return to Baseline (43%); notable increases during the sobriety warning and CM phases (p < .05).
Average Earnings: Participants averaged $77 (SD = $27) during CM phase.
Agreement between EtG-I and TLFB: 74% agreement on self-reported drinking at least 2 standard drinks over 2 days prior to urine collection.
Sensitivity and Specificity of EtG-I:
Sensitivity: 75% for drinking 1 day prior, 50% for 2 days prior.
Specificity: High at 97%, disproving prior concerns over false positives from incidental ethanol ingestion.
Limitations of the Study
Potential influence of breath test administration on participant behavior.
Need for larger studies for robust EtG-I accuracy evaluation.
Small sample size limits conclusions.
Future research should investigate on-site analyzers for accurate immediate feedback on EtG levels.
Conclusion
Recommendations for further investigations into the feasibility and effectiveness of CM interventions based on EtG-I testing for alcohol abstinence are necessary, especially regarding cost-effective on-site solutions.