Urinalysis for detection of drug abuse is widely used in pre-employment screening and correctional administration. Substantial protocols have been developed in these applications of drug testing to minimize the impact of any given vendor on results. There is much less documentation for epidemiological drug testing programs such as the Arrestee Drug Abuse Monitoring program. The paper explores the impact on drug testing results of changes in laboratory vendors and data collection methods, and reports findings from a comparison study. More than 99 percent of the specimens tested the same at both laboratories. The study found minor differences in prevalence rates for some broad drug categories, such as cocaine, barbiturates and amphetamines. The article briefly discusses possible explanations for those differences; the effect of the small number of samples used in the study; and the overall low prevalence rates for some of the drugs tested. The article suggests that future research consider the effects of individual subject characteristics (e.g., body weight and health conditions) on metabolization speeds and degradation rates for drugs. Figure, tables, references