"The hypothesis that patients who are appropriately matched to treatments will show better
outcomes than those who are unmatched or mismatched is well founded in medicine, behavioral
science, and alcoholism treatment," according to NIAAA Director Enoch Gordis, M.D. "These
findings challenge the notion that patient-treatment matching is necessary in alcoholism
About nine percent of U.S. adults meet diagnostic criteria for alcohol dependence (alcoholism)
and the less severe medical disorder alcohol abuse. Of these, more than 500,000 Americans were
treated in 1993 in more than 8,000 inpatient and outpatient alcohol treatment programs in the
United States. While some of these patients have experienced or eventually will experience
lasting remission, authoritative studies report relapse rates of more than 50 percent or more at
two to four years after treatment. A recent meta-analysis of past alcohol treatment outcome studies estimates
that more than 50 percent of treated patients relapse within the first 3 months after treatment.
Researchers during the 1980s saw promise for improving alcohol treatment outcomes in reports
from more than 30 small-scale studies indicating that patient characteristics interact with
behavioral treatments to affect results. At the end of the decade, when the National Academy of
Science's Institute of Medicine (IOM) recommended further research on patient-treatment
matching, potential benefits were seen not only as improved treatment outcomes but also as
increased cost-effectiveness and improved resource utilization.
NIAAA initiated "Matching Alcoholism Treatments to Client Heterogeneity" (Project MATCH)
in 1989 to provide a rigorous test of the most promising hypothetical matches.
Project MATCH investigators selected three behavioral treatments that differed markedly in
philosophy and practice: 12-step facilitation therapy (based on the principles of Alcoholics
Anonymous but an independent treatment designed to familiarize patients with the AA
philosophy and to encourage participation), cognitive-behavioral therapy (based on social
learning theory and designed to provide skills for avoiding relapse), and motivational
enhancement therapy (based on motivational psychology and designed to help patients mobilize
personal resources to effect change). The treatments were selected in part for their
distinctiveness and in part because each had demonstrated effectiveness, the potential to reveal
matching effects, and the potential to be incorporated into standard alcoholism treatment
programs. Many patients in the three treatments also participated in community meetings of
Alcoholics Anonymous, a mutual support fellowship rather than a formal treatment.
Patient characteristics, chosen on the basis of research and theory, included severity of alcohol
involvement, cognitive impairment, psychiatric severity, conceptual level, gender, meaning-seeking, motivational readiness to change, social support for drinking versus abstinence, sociopathy, and typology of alcoholism. The Project MATCH research group predicted 16
contrasts in patient responses to the treatments, to be measured as percentage of abstinent days
and average number of drinks per drinking day during the year following treatment.
The researchers recruited 1,726 patients for two parallel study arms--one with alcohol dependent
patients who received outpatient therapy and one with patients who received aftercare therapy
following inpatient or day hospital treatment--that involved identical matching hypotheses and
identical randomization, assessment, and followup procedures. Twenty-five therapists
administered the therapies over a 12-week period in individual counseling sessions that adhered
strictly to manuals developed for the trial.
The single confirmed match reported today is between patients with low psychiatric severity and
12-step facilitation therapy. Such patients had more abstinent days than those treated with
Overall, Project MATCH participants showed significant and sustained improvement in
increased percentage of abstinent days and decreased number of drinks per drinking days, with
few clinically significant outcome differences among the three treatments in either treatment arm.
For example, there was no difference in sustained abstinence among treatments in the aftercare
arm. However, outpatients who received 12-step facilitation were more likely to remain
completely abstinent in the year following treatment than outpatients who received the other
Patients who participated in Project MATCH also showed decreased use of other drugs,
depression, and alcohol-related problems as well as improved liver function--improvements that
were maintained throughout the 12 months following treatment, the period during which most
relapses typically occur.
The Project MATCH patients probably did well because the treatments were of high quality and
well delivered, according to Thomas F. Babor, Ph.D., Department of Psychiatry, University of
Connecticut Health Center and principal investigator for the Project MATCH Coordinating
Center: "The striking differences in drinking from pretreatment levels to all followup points
suggest that participation in any of the MATCH treatments would be associated with marked
Gerard Connors, Ph.D., chairperson of the Project MATCH Steering Committee and principal
investigator at the Research Institute on Addictions in Buffalo, New York, noted that today's
findings do not rule out the possibility that other patient-treatment matching effects may be
clinically important. "The MATCH data do not speak at all to matching patients types to
different treatment settings, therapists, psychotherapies other than those studied, or
pharmacological treatments," he said.
"A logical next step for alcoholism treatment research is to test our quite excellent behavioral
treatments in conjunction with promising pharmacological treatments for alcoholism," said
Richard K. Fuller, M.D., Director of NIAAA's Division of Clinical and Prevention Research,
which oversaw Project MATCH.
The Project MATCH findings may surprise but should not dismay those who foresaw a
revolution in alcohol treatment delivery based on patient-treatment matching, said Dr. Gordis.
"These findings are good news for treatment providers and for patients who can have confidence
that any one of these treatments, if well-delivered, represents the state of the art in behavioral
"Research attention now must focus full force on illuminating the complex biochemical
mechanisms of alcoholism, including the abnormal appetite for that leads to impaired control
over intake, tolerance--drinking ever greater amounts to achieve a desired effect--and the
discomfort of abstinence known as craving. Our best pharmacologic and behavioral treatments
await that understanding."
"Matching Alcoholism Treatments to Client Heterogeneity: Project MATCH Posttreatment
Drinking Outcomes" appears in the January 1997 issue of the Journal of Studies on
Alcohol , the oldest journal in the field of alcohol research.