Recently, Arizona lawmakers adopted tough deterrence-based legislation in response to heightened concern about the problem of alcohol-impaired driving. Interrupted time-series analysis of traffic data was used in measuring the impact of the new law. Results indicated that the legal intervention had no deterrent effect. Upon examination of possible competing events, it was observed that significant deterrence of drinking and driving was achieved prior to introduction of the new law. Initiation of this effect was attributed to pre-existing antidrunk-driving publicity. 4 tables, 12 figures, and 14 references. (Author abstract modified)