AIMS

Smoking cessation treatment trials often require that smokers quit on or before a protocol-defined date. The goals of this paper were to: (1) identify factors associated with adherence to a protocol-defined quit date and (2) determine whether such adherence predicts cessation outcome (relapse).

DESIGN

A quasi-experimental secondary analysis of data collected from a randomized placebo-controlled trial of fluoxetine (60 mg or 30 mg) versus placebo for smoking cessation.

SETTING AND PARTICIPANTS

Clinic-based smoking cessation treatment program comprising 989 non-depressed smokers.

INTERVENTION

Participants received cognitive behavioral therapy for smoking cessation and either study medication or placebo for 10 weeks. They were required to set a quit date within 2 weeks of their second study visit (by visit 4).

FINDINGS

Significant predictors of quit date adherence were low nicotine dependence and active drug treatment. High-dose fluoxetine (60 mg) and male gender were protective against relapse. Adherence to quit date was not an independent predictor of relapse; instead there was a significant interaction between quit date adherence and gender. Among non-adherers to the quit date, women were more than 2.5 times as likely as men to relapse; among adherers to the quit date, women were only 1.3 times as likely as men to relapse.

CONCLUSIONS

Although women were more likely than men to relapse regardless of quit date adherence, adherence was strongly protective against relapse for women.