Predictors of Smoking Cessation in Outpatients

Kang, Yune Sik; Kim, Jang Rak; Jang, Joung Soon; Hwang, Young Sil; Hong, Dae Yong

Abstract

OBJECTIVE
This study was conducted in order to investigate predictors of smoking cessation in outpatients. METHOD: Subjects were 401 adult smoking patients who saw their doctors in the outpatient setting at a university hospital, regardless of their willingness of otherwise in smoking cessation. Physicians delivered a brief, stop smoking prompt to all patients who smoked one or more cigarettes a day. Then they referred to on-site counselors who provided a brief, nurse assisted intervention with a survey to a randomly assigned intervention group (200 smoking patients), whom the counselors telephoned later to prevent relapse or promote the motivation to quit, or gave only a survey to a control group (201 smoking patients). After at least 5 months, self-reported current smoking cessation was confirmed later using cut-off values of 7 ppm or less in expired alveolar air after breath holding portable CO analyzer. RESULTS: After 5 months, subjects in the intervention group were 1.56 times (95% C.I. 0.89-2.73) more likely to quit smoking than those in the non-intervention group (14.0% vs. 9.0%). Willingness to quit smoking in a month, scheduled admission in a month, self efficacy score and FTND (Fagerstrom Test for Nicotine Dependence) score were all significantly related with smoking cessation. In stepwise multiple logistic regression, previous attempts to quit smoking were significant instead of self efficacy score. In the intervention group who had willingness to quit smoking in a month (132 smoking patients), FTND score, whether quit date was today, and whether quit promise paper was submitting were all significantly related with smoking cessation. In stepwise multiple logistic regression, scheduled admission in a month and whether quit date was today were significant predictor variables. Smoking cessation treatment should be tailored to individual smoking patients considering these predictors.