Randomized Trial of Intensive Motivational Interviewing (IMI) to Improve Drinking Outcomes Among Women

The proposed study responds to the Program Announcement PA-10-100, "Alcohol Use Disorders: Treatment, Services Research, and Recovery" (R01) issued by the National Institute on Alcohol Abuse and Alcoholism (NIAAA). Our application addresses the priorities of this announcement by proposing a phase II randomized clinical trial comparing intensive motivational interviewing (IMI) to standard motivational interviewing techniques (SMI) for alcohol dependent women. The study will use mixed model quantitative and qualitative methods to respond to the program announcement's call for studies assessing mechanisms of change. Unlike many previous studies of SMI we will employ limited exclusion criteria and will enroll participants who present with co-existing drug and psychiatric disorders.

SMI was initially developed as a brief intervention (typically 1-3 sessions) for the treatment of problem drinkers (Miller \& Rollnick, 2002). The general strategy of SMI is to strategically combine supportive and directive interventions as a way to elicit ambivalence about substance use (i.e., pros and cons) and thereby increase motivation for change. Systematic reviews of the alcohol treatment literature have consistently shown SMI to be effective for treating alcohol use disorders (Hettema, Steele, \& Miller, 2005; Dunn, Deroo, Rivera, 2001). Studies have shown SMI to be effective as a stand-alone treatment for alcohol problems (Project MATCH Research Group, 1997; Miller, 1993; Sellman, Sullivan, Dore, Adamson, \& MacEwan, 2001) and as preparation for more intensive treatment (Burke, Arkowitz, \& Menchola, 2003). Perhaps the best-known and largest study of an SMI based was an assessment of "motivational enhancement therapy" (MET) as a stand-alone, 4 session condition conducted by the Project MATCH Research Group (Sellman, Sullivan, Dore, Adamson, \& MacEwan, 2001; Project MATCH Research Group, 1998). Participants were randomly assigned to 4 sessions of MET, 12 sessions of cognitive behavioral counseling (CBT), or 12 sessions of 12-step facilitation (TSF). Despite fewer sessions in MET, outcomes among the 3 conditions were comparable. A limitation of the study was that individuals with serious co-occurring drug or mental health problems were excluded as were those with housing instability.

Like Project MATCH, most studies of SMI for alcohol problems have excluded participants who have serious co-occurring drug and mental health problems. However, these are precisely the types of individuals that treatment practitioners are likely to encounter in publicly funded treatment programs. When SMI interventions have targeted illicit drug dependence the results have been mixed. While a number of reviews (Dunn, Deroo, Rivera, 2001) and meta-analyses (Burke, Arkowitz, \& Menchola, 2003; Hettema, Steele, \& Miller, 2005) have concluded standard low-dose SMI is effective as preparation for more intensive drug treatment, some studies contradict these findings (Downey, Rosengren, \& Donovan, 2001; Miller, Yahne, Tonigan, 2003; Winhusen et al., 2008; Mullins, Suerez, Ondersman, Page, 2004; Schneider, Casey, Kohn, 2000). When effects have been found for drug and co-occurring disorders they have most commonly been for retention, not substance use (Carroll et al., 2006).

Procedures for the proposed study draw from our current successful RCT assessing IMI for MA dependence. Successful aspects of the current study include achievement of recruitment goals, strong adherence to the treatment and research protocols, and excellent rates for follow-up interviews (>90%). The proposed study will take place at the same outpatient treatment program as the current study, New Leaf Treatment Center in Lafayette, California. Two hundred and twenty alcohol dependent women will be randomly assigned to IMI or SMI with an attention component for time equivalence (nutrition education). Participants in both groups will receive standard group treatment offered at the program. The primary outcomes will be measures of drinking, heavy drinking (4+ drinks), and severity of alcohol problems assessed at baseline and 2, 6, and 12 months. Secondary outcomes will include Addiction Severity Index scales, psychiatric problems, and symptoms of trauma. The study will include standard quantitative testing of potential mediators, including, the therapeutic alliance, self-efficacy, motivation, satisfaction, and use of outside services. However, the application also proposes an innovative use of qualitative procedures to identify unrecognized factors influencing outcome.

Aim 1: To compare drinking and alcohol problems among women receiving IMI and SMI.

Hypothesis 1.1: During the active phase of treatment (weeks 1-9) women in IMI will have fewer positive breathalyzer screens, drinking days, and heavy drinking days (4+ drinks) than women in SMI.

Hypothesis 1.2: Women in IMI will demonstrate fewer drinking days and heavy drinking days (4+ drinks) and lower scores on the Addiction Severity Index (ASI) Alcohol Scale at 2, 6, and 12 months.

Aim 2: To assess potential mediators of drinking and alcohol problems. Hypothesis 2.1: Measures of retention, therapeutic alliance, social support, and motivation will mediate the relationship between study condition and drinking outcome.

Aim 3: To Use qualitative interviews to identify additional factors that influence outcome. Content areas will include perceptions of the therapist, IMI \& SMI interventions, the CBT group, and external influences such as Alcoholics Anonymous (AA), intimate partners, friends, and family.

Exploratory Aims: 1) To compare IMI and SMI outcomes on ASI scales, psychiatric symptoms, AA involvement, and trauma symptoms.