Exploring discontinuity in prediction of smoking cessation within the precontemplation stage of change

Background: A stage model of health behaviour change assumes homogeneity within the stages, in terms of how predictors relate to stage progression. It is not known whether this assumption is valid for the Transtheoretical Model (TTM) stages of change. Purpose: To test the homogeneity of the precontemplation stage of the TTM for smoking cessation. We hypothesized that if there was any discontinuity in prediction, it would be most likely to occur around a question which made a distinction between those who were "happy to smoke for the rest of their life" and those who "should quit, but are not currently thinking about it." Method: Repeated measures design with telephone surveys conducted two weeks apart. Participants were 370 randomly selected precontemplators aged 18-40 years. Results: Evidence of discontinuity in prediction was found within the precontemplation stage, but across a different split point than expected. More predictive power of a multivariate model was found among smokers who stated they were not interested in quitting, with this at least in part due to "wanting to smoke," than was found when the model was run on the overall sample. Conclusion: The study supported a possible discontinuity in prediction within the precontemplation stage, suggesting that the stage may not be uniform. Further research is required to confirm whether there is a discontinuity and, if so, to determine its precise location and identify the predictors for which there is a difference.