A structured evidence-based literature review was completed to determine if there was epidemiological evidence of an association of opioid use and intoxicated driving, motor vehicle accidents (MVA) and MVA fatalities; to rate the quality of this research evidence according to Agency for Health Care Policy and Research (AHCPR) type of evidence and strength and consistency of the evidence rating scales; and according to this evidence determine whether patients taking opioids can drive safely. Relevant references were located from Medline, Psychological Abstracts, Science Citation Index and the National Library of Medicine Data Query databases by appropriate subject headings. A manual search was also performed utilizing the reference lists of the retrieved articles. All references relating to intoxicated driving, MVA and MVA fatalities were retrieved and reviewed in detail. Of these, references addressing opioid use were isolated and research information from these references was placed into tabular form under three major headings: Intoxicated driving and opioids; MVA and opioids; and MVA fatalities and opioids. Data were extracted from these references according to the following format: research question addressed, sample size, statistical analysis and results. The type of evidence each study represented was rated according to the AHCPR type of evidence rating scale. Each research area (intoxicated driving, MVA, MVA fatalities) represented by all the studies in each table was then rated utilizing the strength and consistency of the evidence AHCPR rating scale. Of the 6 reports addressing intoxicated driving, 5 were well designed non-experimental studies and one was a well designed experimental study. All studies in this group reported a prevalence for opioid use which was approximately 1/10 that of the point prevalence use for opioids in the general population. This evidence indicated that opioids probably are not associated with intoxicated driving. Overall, the evidence indicates that opioids are not associated with MVA. Of the 9 studies addressing MVA, 5 were well designed quasi-experimental studies and 4 were well designed experimental studies. All reports in this group except one indicated that opioids are not associated with MVA. Of the 10 studies addressing MVA fatalities, 10 represented Type IV studies. For the vast majority of the studies, the prevalence percentages for an opioid association with MVA fatalities was 1/5 that of the point prevalence percentage for opioid use reported in the general population. Only 1 study reported a possible association between opioid use and MVA fatalities. The evidence in this review indicates that opioids do not appear to be associated with intoxicated driving, MVA and MVA fatalities, and consistently indicated that opioids are not associated with MVA. Although the comparison of point prevalence rates to the point prevalence may be problematic, the results of this systematic review support the contention that patients taking opioids may be allowed to drive. As in all clinical decisions, this determination should be individualized according to clinical factors.