Two brief valid measures of therapeutic alliance in counseling for tobacco dependence

Abstract

Behavioral counseling is effective for smoking cessation and the psychotherapy literature indicates therapeutic alliance is key to counseling effectiveness. However, no tobacco-counseling specific measures of alliance exist that are suitable in most tobacco counseling contexts. This hinders assessment of counseling components in research and clinical practice. Based on the Working Alliance Inventory, and external expert review, we developed two alliance instruments: the 12-item and 3-item Working Alliance Inventory for Tobacco (WAIT-12 and WAIT-3). Two samples of 226 daily smokers via Amazon Mechanical Turk completed measures including demographics, tobacco characteristics, working alliance scales, and quit attempts. Both WAIT-12 and WAIT-3 had good to excellent internal consistency (0.92 and 0.88 for the WAIT-3 and 0.96 for the WAIT-12). The WAIT-12 1-factor model indicated poor fit (CFI=0.83, TLI=0.79, RMSEA=0.19, SRMR=0.09). The WAIT-12 3-factor model (CFI=0.94, TLI=0.93, RMSEA=0.11, SRMR=0.04) was indicative of acceptable fit. Both the WAIT-12 and the WAIT-3 were significantly associated with participants' self-reported cigarettes per day, quit attempts, and cessation. Initial validation of the WAIT-12 and WAIT-3 indicates they are psychometrically sound measures of tobacco dependence counseling alliance. The WAIT-3 provides brevity; it can be administered in under 1min. The WAIT-12 allows for assessment of specific components of therapeutic alliance. Overall, these instruments should allow for better measurement of alliance in clinical services and research.