BACKGROUND: It has been proposed that the expired-air carbon monoxide (CO) threshold for confirming smoking abstinence in clinical practice be reduced below 10 ppm. Optimal thresholds may vary across regions. Data are needed to assess the impact of such a change on claimed success. METHODS: A total of 253 smokers who attended the Tanglin quit smoking clinic in Malaysia were followed-up 1, 3 and 6 months after the target quit date. All participants received a standard behavioural support programme and were prescribed either varenicline or nicotine replacement therapy. Expired-air CO was measured at every visit. Respondents' smoking status was assessed using a range of different CO thresholds (3, 5 and 10 ppm) and the impact on quit rates was calculated. Predictors of success as defined using the different thresholds were assessed. RESULTS: The 6-month abstinence rates were: 1 month - 54.9% at 10 ppm, 54.9% at 5 ppm and 48.6% at 3 ppm; 3 months - 36.0% at 10 ppm, 35.2% at 5 ppm and 30.4% at 3 ppm; 6 months - 24.1% at 10 ppm, 24.1% at 5 ppm and 20.6% at 3 ppm. Older smokers were more likely to be recorded as abstinent at 6 months regardless of the threshold used. CONCLUSIONS: Reducing the threshold for expired-air carbon monoxide concentrations to verify claimed smoking abstinence from 10 ppm to 5 ppm makes minimal difference to documented success rates in Malaysian smoker's clinic patients. Reducing to 3 ppm decreases success rates slightly. Predictors of success at stopping appear to be unaffected by the threshold used.