Nicotine dependence criteria of the DIS and DSM-III-R: a factor analysis.

This paper reports a factor analysis of the symptoms of nicotine dependence that were determined in an assessment of 821 current cigarette-smoking research volunteers, according to the Diagnostic and Statistical Manual of Mental Disorders, 3rd edition, revised (DSM-III-R) of the American Psychiatric Association as well as an analysis of a subset who unsuccessfully attempted to quit (n=636). In the total sample, two factors with eigenvalues greater than 1 accounted for 62.7% of the variance. When the factor analysis was repeated with the subset of research volunteers who unsuccessfully attempted to quit, only one DSM-III-R nicotine dependence symptom loaded on the second factor. This finding suggests that the two-factor structure found in this and a previous factor analysis study of the nicotine dependence segment of the DSM-III-R may be an artifact of the skipout pattern of the DSM-III-R, which assumes that smokers who have not attempted to quit have not experienced withdrawal symptoms or used tobacco to avoid these symptoms. Goodness-of-fit measures suggested that the two-factor structure is a better fit than the one-factor structure for both the total population and the subset who unsuccessfully attempted to quit or cut down. Our sample of current smokers who had not attempted to quit (n=185) was too small to permit factor analyses. Further work with other large samples from the general population of current smokers who have unsuccessfully attempted to quit as well as those who have not attempted to quit will enhance our understanding of the factor structure of the nicotine dependence segment of the DSM-III-R and clarify the effect of the skipout pattern on its factor structure.