To identify subgroups of smokers with different success rates, we appl
ied ''tree-structured survival analysis'' (TSSA) to data from a previo
usly published trial of transdermal nicotine. The subjects who receive
d active treatment (14 mg patch, n = 275, or the 21 mg patch, n = 262)
constituted the sample for this analysis. Using age, gender, the Fage
rstrom Tolerance Questionnaire (FTQ), motivation to quit, number of ci
garettes smoked at baseline, and body mass index (BMI) as classificati
on variables, TSSA identified two subgroups within the 14 mg patch gro
up and four subgroups of smokers within the 21 mg patch group. Among t
hose receiving the 14 mg patch, individuals with a BMI greater than 26
.4 kg/m(2) relapsed sooner than did those with a BMI less than or equa
l to this value. Within the 21 mg patch group, the survival curve for
males was significantly different from that observed in females, with
males experiencing a longer time to relapse after treatment than did f
emales. Among males, those who were less dependent relapsed significan
tly later than did those men who were more dependent. Among females, t
hose with a higher motivation to quit relapsed more slowly than did th
ose women with less motivation to quit. This information may be helpfu
l to clinicians seeking to match specific patients to specific treatme
nts with transdermal nicotine in order to maximize treatment outcomes.