The impact of change in smoking status on 12-month substance abuse (SA) treatment outcomes was examined among an HMO population seeking SA treatment. Of the 749 participants who entered the study at baseline, 649 (86.9%) were retained at the 12-month follow-up. At treatment entry, 395 participants were smokers and 254 were nonsmokers. At 12-month follow-up, 13% of the 395 baseline smokers reported quitting smoking and 12% of the 254 baseline nonsmokers reported starting/relapsing to smoking. Those who quit smoking were less likely to be diagnosed as alcohol dependent compared to those that remained smokers. Those who started/resumed smoking were more likely to be diagnosed as both alcohol and drug dependent at treatment entry compared to all other groups. Total days abstinent from alcohol and illicit drugs was greatest for individuals who quit smoking (adjusted M=310.6) or who were nonsmokers (adjusted M=294.7) and lowest for those who started/resumed smoking (adjusted M=246.6) or remained smokers (adjusted M=258.2), even after controlling for demographic (i.e. age, income), psychosocial (ASI psychiatric severity), and other treatment characteristics (length of treatment stay, prescribed bupropion) that were associated with days abstinent at 12 months. Self-initiated smoking cessation does not appear to be detrimental to SA treatment outcomes, and may be beneficial. Starting/resuming smoking after entering SA treatment may be a clinical marker for individuals at greater risk of relapse. Future studies may want to measure the smoking status of all participants at all time points in order to include this higher-risk group of substance using smokers.