A method for studying controlled substance use: a preliminary investigation

This study examined relapse following treatment in a smoking reduction program. Twenty-two smokers with an average daily smoking rate of 25.7 cigarettes were provided a reduction-oriented smoking self-management program. Subjects were then classified on the basis of their reduction at posttreatment. A 50% or greater reduction from pretreatment qualified subjects as successful reducers. To provide further understanding of controlled smoking, subjects who successfully maintained at least a 50% reduction (at three month follow-up) were compared to those subjects who were unable to maintain that level of reduced smoking. A set of cognitive, smoking history, and reduction motivation variables were used to compare the two groups in a stepwise discriminant function analysis. Two variables, internal locus of control and self-label as a nonsmoker, produced an overall correct classification rate of 88.9%. Multiple regression analyses further demonstrated the importance of these two cognitive variables; self-label and internal locus of control accounted for 50.4% of the variance in smoking at three months follow-up. The implications of these findings for future study of controlled smoking are discussed.