Baseline factors associated with smoking cessation and relapse. MRFIT Research Group

Prev Med. 1991 Sep;20(5):590-601. doi: 10.1016/0091-7435(91)90057-b.


Background: Data on smoking cessation and relapse for 6 yers of the Multiple Risk Factor Intervention Trial were evaluated in univariate and multivariate analyses to determine the relationship between variables measured at the beginning of the trial and smoking cessation and relapse for special intervention and usual care participants.

Results: The variables positively associated with smoking cessation in both the SI and the UC groups included age, education, and past success in quitting; there was a negative association with the number of cigarettes smoked per day. The expectation of quitting was positively associated with cessation in the special intervention group only, while life events, alcohol, and the presence of a wife who smokes were significant predictors of reduced cessation for the usual care group. The special intervention program may have overcome obstacles which interfered with cessation among the usual care participants. Associations with relapse were generally stronger in the usual care group than in the special intervention group. For usual care participants, multivariate analyses showed that education, past success in quitting smoking, alcohol, and life events were associated with relapse rates. For special intervention participants, only alcohol emerged as a significant predictor. Conclusion. The data are relevant in terms of factors that govern smoking cessation and relapse for adult smokers who take part in formal intervention programs and for those who are left to modify their behavior on their own.

Publication types

  • Clinical Trial
  • Comparative Study

MeSH terms

  • Age Factors
  • Alcohol Drinking
  • Data Interpretation, Statistical
  • Educational Status
  • Female
  • Humans
  • Life Change Events
  • Male
  • Marriage
  • Program Evaluation*
  • Recurrence
  • Smoking Cessation*
  • Smoking Prevention*
  • Social Environment