Differential rates of relapse in subgroups of male and female smokers.

Subjects for this study were 265 participants of stop-smoking clinics (mean age = 42.6 years; average number of cigarettes smoked daily = 26.0) who were examined before and immediately after cessation and then followed for 1 year. The objective of this study was to identify subgroups of smokers with different rates of relapse using tree-structured survival analysis, a multivariate approach to classification. Five distinct subgroups that differed with respect to the rate of relapse were identified: (I) subjects (n = 15) with very low precessation cotinine levels (< or = 129 ng/ml), who had an exceptionally low rate of relapse (mean abstinence time = 270 days); (II) women 32 years old and younger (n = 24), who had a very high rate of relapse (mean abstinence time = 30.5 days); (III) women over 32 years old (n = 121), with the next highest rate of relapse (mean abstinence time = 98.9 days); (IV) men 36 years old and younger (n = 31), who had a mean abstinence time of 196.7 days; and (V) men over 36 years old (n = 74), who abstained an average of 130.2 days before relapsing. Relapse curves for all groups (except III vs V) differed significantly from each other, p < 0.05. Results indicate that this approach can identify interactions among individual differences that are variably associated with relapse rates. Identification of relapse subgroups may have important implications for both theories and treatment of smoking relapse.