Evaluating three theory-based interventions to increase physicians' recommendations of smoking cessation services.

OBJECTIVE: To evaluate three theory-based interventions aimed at increasing the rate at which primary care physicians recommend smoking cessation services to smokers. DESIGN: Primary care physicians (n = 251) were randomized to one of four conditions: (a) information based on social cognitive theory (SCT) targeting outcome expectations, (b) information based on SCT + elaboration likelihood model, (c) feedback based on self-perception theory, or (d) control. MAIN OUTCOME MEASURES: Intentions to recommend and self-reported recommendations of smoking cessation services 1 week postintervention. RESULTS: Data were analyzed using covariance and mean structure analysis. Compared with the control group, only the SCT group reported more frequently recommending services (mean difference = 1.1 recommendations per week, Cohen's d = 0.46) and higher intentions. Mediation analysis was consistent with increased intentions resulting from changes in outcome expectations. There was no evidence that changes in intentions explained self-reported behavior change. CONCLUSION: The study provides preliminary "proof of principle" for further studies incorporating more robust outcome measures.