OBJECTIVES: To examine whether smoking relapse after smoking abstinence was associated with the quitting trajectories and smoking cessation medications.

METHODS: We identified 2668 quitters from the 6-month follow-up phone interview during the period from March 2012 to January 2015. These quitters received monotherapy for smoking cessation covered by Taiwan Smoking Cessation Program. Quitting trajectories since the first time of receiving medications consists of two categories: delayed quitters (N=1290) and immediate quitters (N=1378). Among 2668 quitters, 1340 were users of Nicotine Replacement Therapy (NRT), 141 were bupropion users and 1187 were varenicline users. Smoking relapse was defined based on the abstinence status in the 12-month follow-up interview. Multivariable logistic regression modeling was performed to estimate the odds of relapse comparing the quitting trajectories and smoking cessation medications.

RESULTS: The overall relapse rate was 23.0%; delay quitters, 29.8% versus immediate quitters, 16.7%. Multivariable logistic regression showed that delayed quitters were more likely to relapse than immediate quitters (adjusted odds ratio [aOR] = 2.10, 95% confidence interval [CI]=1.71-2.55). The relationship between smoking cessation medications and relapse was insignificant (reference group: varenicline, NRT uses: aOR=1.01, 95%CI=0.82-1.25; bupropion uses: aOR=1.13, 95% CI=0.74-1.74).

CONCLUSIONS: Immediate quitters maintained smoking abstinence status better than delayed quitters. Smoking relapse was not associated with type of medication. To increase the long term cessation rates, health professionals should encourage and help smokers to quit smoking immediately and monitor their abstinence status through the process of smoking cessation treatment.