Police Drug Testing, Hair Analysis, and the Issue of Race Bias

Abstract

Poice agents have beg to use hair analysis as a drug-screening tool in both officer recruitment and officer monitoring. This has been controversial because of a putative racial bias associated with human hair specimens, especially focused on cocaine. However, little empirical evidence has been offered to substantiate this claim. To asses this issue, Hoffin (1999, p. 613) analyzed recruitment cohort data from a major metropolitan police department ( N - 1,80) for cocaine ad marijuana hair assays, contrasting recruits by race. Hoffman's conclusion, based on comparison of odds ratios and relative rates for positive assy outcomes, was that there was "no evidence that one group (i.e., race) was more adversely affected by hair testing compared to urine testing." This article reports on research seekig to replicate Hoffman's results concerning hair analysis by utilizing a similar analytic approach with a much larger data set ( N - 40,000) and examining a different major metropolitan police department. This research is disnct form the Hoffman study in two ways: The assessment was exclusively of cocaine prevalence, and the subjects were who job applicants but sworn police officers. The findings reported here support Hoffman's conclusion that there is no statically significant race bias attributable to hair analysis, at least as applied to cocaine.