Predictors of smoking reduction and cessation in a cohort of Danish moderate and heavy smokers

Background. The aim of this study was to examine the extent and gender distribution of unassisted tobacco reduction and cessation in a cohort of moderate and heavy smokers and to identify possible predictor variables associated with these changes in smoking behavior. Methods. This was a prospective population study of 3,791 moderate and heavy smokers, 15 g tobacco/day or more, who were enrolled in the Copenhagen City Heart Study in 1976-1978 and attended a reexamination 5 years later. Data on smoking behavior were collected at baseline and follow-up. Smoking reduction was defined as a decrease in mean daily tobacco consumption of 10 g or more. Using multivariate logistic regression, subjects who reported reduced smoking or who reported smoking cessation were compared with subjects who continued the habit unchanged. Results. After 5 years 13% of the men and 9% of the women had reduced their tobacco consumption, and 9 and 7%, respectively, had quit altogether. Smoking reduction was strongly associated with high tobacco consumption (25+ g/day) at baseline and also with severely impaired lung function (FEV1 < 50% predicted) and overweight (BMI > 25). Predictors of smoking cessation included impaired lung function and a tobacco consumption of 15-24 g/day. Additional determinants of smoking reduction and cessation such as inhalation habits and sociodemographic variables differed by gender. Conclusions. Several predictors of smoking reduction and cessation were identified, indicating that these subgroups of smokers differ substantially from continuing smokers. This should be taken into account when assessing potential health benefits from these changes in smoking behavior (C) 2001 American Health Foundation and Academic Press.