Review of: McDonell MG et al, Am J Psychiatry 2017;174(4):370–377. doi:10.1176/appi.ajp.2016.16050627.

Although studies have demonstrated the effectiveness of contingency management (CM) for illicit drug use, there’s less evidence for treatment of alcoholism—in part because a standard breathalyzer has a short detection window of 12 hours, meaning patients must only abstain from drinking since the previous night to pass the test.

Over the past few years, however, a more effective alcohol biomarker has been introduced. Ethyl glucuronide (EtG) is an alcohol metabolite that is present in the urine for at least 5 days after a patient’s last drink. It can therefore verify longer-term abstinence.

Researchers in Seattle recruited 79 patients who had both alcohol use disorder and serious mental illness such as schizophrenia, bipolar disorder, and recurrent depression. Roughly two-thirds were men, half were white, and the average age was in the mid-40s. All were in outpatient substance use disorder (SUD) treatment. Before being randomly grouped, participants had to complete a 4-week induction period designed to identify those who were most likely to stay in the actual 12-week study. Those who showed up during the induction phase were randomly assigned to a CM group (N = 40) or a noncontingent reinforcement (control) group (N = 39). Participants in both groups provided urine samples 3 times a week.

After the induction phase, participants in the CM group who submitted 3 consecutive urine samples negative for EtG earned “prize draws” from a container of tokens. Half of the tokens simply said “good job,” while the other half could be turned in for prizes ranging in value from $1 to $80. Participants also received $10 gift cards for attending SUD groups each week. Those in the control group received prize draws for each urine sample submitted, no matter what the result. Control participants also received gift cards regardless of whether they attended groups.

RESULTS
The CM group had significantly more EtG-negative urine samples (mean of 8.56) than the control group (mean 4.11). This translated to 1.5 weeks of additional continuous abstinence. Moreover, the CM group had significantly fewer drinking days and fewer days of drinking to intoxication throughout the study. These differences persisted into a 3-month follow-up period.

CATR’s Take
This well-designed study supports the effectiveness of contingency management for patients dually diagnosed with mental illness and alcohol use disorder. Point-of-care EtG costs money, but the benefits in terms of improved sobriety are likely worth it.