Objective: To determine smoking behaviour, acceptability, and toxin exposure when smokers switch to the potential reduced exposure product - Omni cigarette. Design: 12 week randomised, crossover study of Omni versus own cigarettes. Participants: 19 light/ultralight and 15 regular smokers. Outcomes: Cigarettes/day, smoking topography, craving, withdrawal symptoms, urinary cotinine plus its glucuronide (total cotinine), nicotine plus its glucuronide (total nicotine), and carcinogen metabolites (4-(methylnitrosamino)-1-(3- pyridyl)-1-butanol plus its glucuronides and 1-hydroxypyrene). Results: When switched to Omni, smokers smoked the same number of cigarettes/day, smoked Omni cigarettes less intensely (total puff volume = -11%) and had slightly lower total cotinine (-18%) levels than their own cigarettes, but had a slightly greater carbon monoxide boost/cig (+21%). Craving and withdrawal ratings were similar with Omni and own cigarettes. Carcinogen metabolite levels were somewhat but not significantly lower with Omni. About half of smokers rated Omni as better for their health and about two thirds stated it was weaker and worse tasting than their own cigarettes. Conclusions: Although Omni may be an adequate behavioural and pharmacological substitute for traditional cigarettes, it may not decrease carcinogen exposure and may increase carbon monoxide. Replications with larger sample sizes and longer follow up are needed. These results indicate the need for regulation of reduced exposure and reduced risk claims.