Highlights of the American Psychological Association 114th Annual Convention

August 10-13, 2006; New Orleans, Louisiana

Joshua Fogel, PhD


November 30, 2006

In This Article

Alcoholism Treatment

Margaret E. Mattson, PhD, Health Scientist Administrator, National Institute on Alcohol Abuse and Alcoholism (NIAAA), Bethesda, Maryland, described the background and rationale for the Combined Pharmacotherapies and Behavioral Interventions for Alcohol Dependence (COMBINE) study.[1] The COMBINE study focused on understanding what combinations of behavioral and pharmacologic treatments are effective for treating those with alcohol dependence.[2] The medications studied were naltrexone and acamprosate. The behavioral treatments studied were less intensive medication management (MM) and moderately intensive combined behavioral therapy (CBI).

Stephanie O'Malley, PhD, Professor, Department of Psychiatry, Yale University School of Medicine, New Haven, Connecticut, discussed the methodology and results of COMBINE.[3] Men and women aged 21 years and older with current alcohol dependence were included. Severe mental illness (eg, schizophrenia) and drug dependence (except for nicotine or marijuana) were two of the exclusion criteria. There were 9 possible treatment options in total and 4 possible medication combinations:

  • Acamprosate and naltrexone;

  • Acamprosate only and naltrexone placebo;

  • Naltrexone only and acamprosate placebo; and

  • Double placebo.

In 4 of the groups, everyone received MM plus one of the 4 medication combinations. Another 4 groups received CBI in addition to MM, plus one of the 4 medication combinations. Those in the ninth group received only CBI, without MM or medication. The medications were provided over 16 weeks with a naltrexone dosage of 100 mg daily and an acamprosate dosage of 3000 mg daily. The behavioral treatments were manual-guided; the CBI intervention provided up to 20 individual sessions over a 4-month duration. There were 1383 individuals randomized to one of the 9 treatment options. Primary outcomes were percentage of days abstinent (PDA) and time to first heavy drinking day.

Overall, as expected, there were great reductions in drinking. PDA tripled from 25% to 76%, and total drinking decreased from 66 to 13 drinks per week (an 80% reduction). Unlike other studies showing benefits for acamprosate, the COMBINE study did not find this drug to be more effective than placebo for any of the primary outcomes. Naltrexone, on the other hand, significantly reduced craving and had increased PDA (80.6%) compared with placebo (75.1%) without CBI. Also, naltrexone had a hazard ratio of 0.72, indicating less risk for relapse compared with placebo. There was significant improvement for the primary outcomes for those in the group receiving CBI, MM, and placebo. However, combining treatments did not further improve outcomes. More specifically, treatment outcomes were not improved by combining (1) acamprosate with CBI, (2) acamprosate with naltrexone, or (3) naltrexone with CBI. Also, either naltrexone or CBI improved outcomes when used with MM.

Joseph LoCastro, PhD, Associate Professor, Department of Psychiatry, Boston University School of Medicine, Boston, Massachusetts, discussed the impact of including a placebo in the study.[4] He emphasized that trials of behavior therapy for alcoholism are typically not compared with medication treatments and are never compared with an additional medication placebo.

Dr. LoCastro reported that there were overall significant differences (P = .0002) for PDA after 16 weeks of treatment between MM and placebo (73%); CBI, MM, and placebo (80%); and CBI alone (67%). These significant differences remained in post hoc pairwise comparisons. Also, there was significantly more relapse to heavy drinking in the CBI-alone group compared with those receiving CBI, MM, and placebo (P = .05). There were no significant differences between the CBI-alone group and the MM and placebo group (P = .46). He discussed possible reasons for the observed placebo effects, including the psychological expectancy of taking medication, and that there was more concrete, extra clinical attention received as part of the MM meetings.

Dennis M. Donovan, PhD, Professor, Department of Psychiatry and Behavioral Sciences, University of Washington, Seattle, discussed the 1-year study follow-up.[5] He reported that PDA declined overall in all of the groups. There was no main effect for either naltrexone or acamprosate. There was a trend (P = .08) for those receiving CBI to have a higher PDA than those not receiving CBI, regardless of medication condition. Also, more individuals had at least 1 heavy drinking day. Only those taking naltrexone had less risk of returning to heavy drinking (P = .04). There was also a trend (P = .08) for those receiving CBI to have less risk for relapse to return to heavy drinking than those not receiving CBI.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.