Study finds no racial bias in hair testing drivers for drugs

A new study paid for by the Trucking Alliance has revealed no evidence that using hair to test drivers for drug use is biased against minority groups when compared to urine tests.

The study, “Assessing the Potential Disparate Impacts of Hair Drug Testing Among Commercial Motor Vehicle Drivers,” dated February 7, was conducted by researchers at the University of Central Arkansas (UCA). It analyzed urine and hair pre-employment drug screening results provided independently from three anonymous trucking companies (all members of the Trucking Alliance) for the years 2017-2019.

“Factors other than testing method seem to underlie ethnic group pass/fail rate differences,” concluded Dr. Douglas Voss, Professor of Logistics and Supply Chain Management at UCA, in the study’s cover letter to the Trucking Alliance. “Given these findings, we were unable to identify evidence of disparate impact among ethnic groups resulting from the use of hair tests.”

In summarizing the study, the Trucking Alliance, whose members include major truckload carriers J.B. Hunt [NASDAQ: JBHT] and U.S. Xpress [NYSE: USX], pointed out that opponents of hair testing have alleged potential racial bias as one reason for not requiring hair follicle tests to screen drivers.

The Owner-Operator Independent Drivers Association (OOIDA), which represents small-business trucking companies, emphasized to FreightWaves last year that it is concerned about “biases toward hair color and texture” in hair testing, as well as what it asserts is a lack of evidence connecting hair testing and crash reduction.

The Trucking Alliance also noted that the U.S. Department of Health and Human Services (HHS) has yet to issue guidelines for hair testing after Congress authorized the U.S. Department of Transportation in 2015 to make it an alternative to urine testing.

In December, U.S. Senator Roger Wicker (R-Mississippi) introduced legislation requiring HHS to submit a report to Congress explaining why HHS delayed in submitting scientific and technical guidelines for hair testing to the Office of Management and Budget.

Ethnic groups in each sample ranked by difference between hair and urine test failure rates.
Source: University of Central Arkansas (with Truckload Alliance data)

Voss confirmed to FreightWaves that the trucking companies supplying hair and urine drug screening data used in this year’s ethnicity analysis also supplied data used in a study conducted by UCA for the Trucking Alliance last year, which found that approximately 300,000 truck drivers who were currently on the road would fail or refuse a hair analysis. That level of driver test failure – if and when hair follicle testing is made a requirement in trucking – could cause a 10% capacity shakeout, trucking executives recently told FreightWaves.

In UCA’s ethnicity bias study, two trucking companies provided paired urine and hair test results from 2017, three provided results from 2018, and one provided results from 2019, from a total of 112,693 separate truck drivers. To provide anonymity to carriers and their drivers, the analysis focused on two different time periods: 2017-2019 combined and 2018 in isolation.

Researchers used two methods to test for disparate impact among eight ethnic categories: the “Four-Fifths Rule,” which uses an 80% passing rate threshold to assess disparity among minority groups, and “chi-square” difference testing.

“Results for each test in each sample met the required Four-Fifths Rule threshold, which indicates equal treatment of ethnic groups for both [hair and urine] testing methods,” the study concluded. In addition, chi-square difference tests “found significant differences across the ethnicity categories for hair and urine when judged separately,” Voss said.

“The theory that hair testing is biased due to biological hair type differences is not supported by our data, due to the fact that there is wide variety of hair types in the groups that failed the hair test more than average, and a wide variety of hair types that failed the hair test less than average.”