Premium
A comparison of two measures of stage of change for smoking cessation
Author(s) -
Etter JeanFrancois,
Perneger Thomas V.
Publication year - 1999
Publication title -
addiction
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.424
H-Index - 193
eISSN - 1360-0443
pISSN - 0965-2140
DOI - 10.1046/j.1360-0443.1999.9412188113.x
Subject(s) - smoking cessation , psychology , transtheoretical model , computer assisted web interviewing , stage (stratigraphy) , clinical psychology , medicine , social psychology , behavior change , paleontology , pathology , biology , marketing , business
Aims. To compare two questionnaires used to identify the stages of change of current and former smokers: a conventional five‐item questionnaire and an alternative one‐item questionnaire. Design and setting. Mail surveys of 1167 ever smokers in Geneva (Switzerland), conducted in 1997. Participants were classified into five stages: precontemplation, contemplation, preparation, action and maintenance. Other questions covered smoking‐related behaviours, attitudes and self‐efficacy. Findings. Only 62% of participants were classified as being in the same stage by the two questionnaires (weighted kappa = 0.69). The five‐item questionnaire produced more missing data (8%) than the one‐item questionnaire (2%, p < 0.001). Using the conventional questionnaire, the precontemplation stage included a group of smokers who had absolutely no intention of stopping smoking and a group more prone to change, and the preparation stage included only 43% of people who had made a "firm decision" to quit smoking in the next 30 days. Using the alternative questionnaire, the contemplation stage was also quite heterogeneous. The action stage included over 35% of people who were still smoking occasionally, whichever questionnaire was used. Conclusions. The single‐item questionnaire was better at avoiding missing responses. However, both staging questionnaires classified smokers in heterogeneous groups, and both misclassified many occasional smokers and ex‐smokers, which suggests that a discrete five‐stage model does not fit reality well. This may reflect underlying conceptual issues, notably that the classic definition of stages incorporates separate dimensions, albeit incompletely (current behaviour, quit attempts, intention to change, time). Both theoretical and methodological developments are needed to overcome these problems.