This study examined the application of a Markov chain analysis for modeling
transitions in the process of smoking cessation from the perspective of th
e Transtheoretical "Stages of Change" model. A convenience sample of 308 in
dividuals, who were followed at 6-month intervals for a 30-month period, an
swered questionnaires regarding their past and current smoking behavior, an
d their quitting attempts and intentions. A first order, single transition,
Markov model was found to fit the data well, indicating that stage informa
tion from a single previous 6 month time point was sufficient to predict cu
rrent stage membership. A composite transition probability matrix indicated
that significant stability occurred in Precontemplation, Contemplation, an
d Maintenance stages with greater transitions among stages for smokers in P
reparation and Action stages. In addition, more people tended to progress i
n stage status toward successfully maintained cessation than to either regr
ess or make no change in stage membership over the 2 1/2 years of the study
. These results support the utility of stage based models of change in unde
rstanding the process of smoking cessation. Smoking cessation is marked by
significant periods of stasis as well as recycling through the stages. For
a significant number of this cohort of participants then was progress and m
ovement toward successful change that seemed to support the spiral movement
through the stages of change proposed by the Transtheoretical "stages of c
hange" model.