Development of the motivational thought frequency scale for alcohol abstinence (MTF‐A)

Abstract

Background and aims For most treatment‐seeking patients with severe Alcohol Use Disorder (AUD), abstinence is the clinically indicated goal. Existing AUD motivation scales are non‐specific about treatment consumption goals, which limit their effectiveness. Desires and mental imagery are relevant in the motivation for AUD treatment engagement. The Motivational Thought Frequency Scale for an abstinence goal (MTF‐A) was adapted from the MTF for controlled drinking (MTF‐CD). This study psychometrically evaluated the MTF‐A in an alcohol‐dependent sample engaged in treatment with a goal of abstinence. To enhance the clinical utility of the scale, a secondary aim was to evaluate a psychometrically equivalent short version of the MTF‐A. Method A sample N of 329 treatment‐seeking patients with AUD (mean age of 44.44 years, SD = 11.89 years, 72% male) who were undertaking a cognitive behavioral treatment (CBT) program for abstinence completed the Motivational Thought Frequency Scale for Abstinence (MTF‐A) and the Severity of Alcohol Dependence Questionnaire (SADQ). The MTF‐A measured motivation for abstinence through four factors: intensity, self‐efficacy imagery, incentives imagery, and availability. Confirmatory factor analyses (CFAs) were conducted to examine factor structure and model fit. Cronbach's alpha assessed internal consistency. Predictive validity was determined by logistic regression predicting first‐session treatment non‐attendance and alcohol consumption between baseline assessment and commencement of treatment, controlling for potential confounds. Results A four‐factor structure provided the best fit for the MTF‐A, compared with one‐ and three‐factor models. A shortened 9‐item MTF‐A scale (S‐MTF‐A) provided better fit than the 13‐item MTF‐A scale. Both MTF‐A and S‐MTF‐A displayed good internal consistency. Although both MTF‐A and S‐MTF‐A successfully predicted first‐session treatment non‐attendance, neither predicted alcohol consumption between the baseline assessment and commencement of treatment. Conclusions The model fit of the four‐factor, 9‐item S‐MTF‐A was superior to the original 13‐item MTF‐A. Both scales were predictive of participation of AUD treatment. Desires and mental imagery play an important role in AUD treatment motivation.