Hair Testing’s Color Blind
The popularity of so-called hair testing to detect drug use is skyrocketing nationwide—a boon for the Cambridge company that is the nation’s largest provider of hair testing services. But with the increased popularity comes new controversy over the accuracy of hair testing, and its possible bias against people with dark hair.
Employers, including some of the nation’s most established corporations, favor hair testing over urine testing because it can reveal drug use months earlier, rather than just the previous few days. A snippet of an employee’s hair is sent to one of the companies that provides analysis—most often Psychemedics Corp. in Cambridge—and the strand’s composition is examined for evidence of drug residue.
General Motors, Anheuser Busch, BMW and Rubbermaid are among the 1,400 companies on Psychemedic’s growing business client list. The company is also conducting hair tests for more than three dozen schools, five Federal Reserve banks, and the police departments of of several major cities. The Boston Police Department just signed on. But hair testing is a matter of growing debate. High on the list of concerns: the test appears to give false positives disproportionately to African-Americans.
Doubts about its accuracy have been raised by several federal and private concerns—from the National Institute of Drug Abuse to the Society of Forensic Toxicologists—with the scientific concensus being that, until questions are resolved, the process is not sufficiently reliable for widespread use. Members of Congress are also entering the fray, calling for a closer look at the process. Sixteen members of the Congressional Black Caucus sent a letter to the secretary of the Army last month asking that the Army review its use of hair testing, in part because of a possible bias against non-whites.
Meanwhile, the American Civil Liberties Union has conducted its own research and concluded that the federal Food and Drug Administration—under its mandate to protect the public from flawed products—should remove it from the market. “I hope that companies will become more aware that the test is unrealiable and stop using it,” says Lewis Maltby, director of the ACLU’s workplace taskforce. “There’s too much at stake here. Why would any employer want to fire a good employee based on a test that isn’t accurate?”
But the companies that provide hair analysis say that what they do is more accurate than other forms of testing. Psychemedics Corp. of Cambridge says it’s own work has an especially good record. The company’s website points out that when CBS did a “blind test” by sending hair samples to Psychemedics and two other laboratories, Psychemedics was the only laboratory that correctly identified the hair of drug users and had negative findings for nonusers.
“Believe me,” says Raymond C. Kubacki Jr., Psychemedic’s chief executive officer, in an ABC interview, if the test was flawed, “the corporations we have would not use it.” Psychemedics’ “patented technology has withstood numerous court challenges,” the company’s literature says. And there is no doubt the company’s business is booming. It more than doubled sales of its hair test between 1993 and 1997, and in 1997 The Globe named Psychemedics one of the top 50 growth companies in the state. The company has tapped into a growing market. “The total annual market for drug testing in the United States has been estimated at between $500 and $600 million,” says Kubacki, “and is growing fast.”
Althea Jones, an African-American mother of two, says she is one of the victims of hair testing’s inaccuracies. Her lifelong dream was to be a police officer, but when she applied for admission to the Chicago Police Academy, it requested a sample of her hair, which it sent to Psychemedics—and the results came back positive for drug use.
“I was shocked. I couldn’t believe it,” says Jones. “I don’t even smoke or drink. I was heartbroken by this.” She was denied admission to the academy and is now a criminal justice major at Chicago State University. Adrian McClure also wanted a career with the Chicago police department. When she was a senior in college in 1997, she submitted a hair sample to the academy, which sent it to Psychemedics Corporation. Her test came back positive, too. She says she tried to explain that it was an error and requested a new test, but was rebuffed. “Everybody knows I don’t use drugs,” says McClure, who is also African-American. “They have shattered me.”
Jones and McClure, along with six other Chicagoans who say they received erroneous hair test results when applying for the Police Academy, have filed complaints of racial discrimination with the Equal Employment Opportunity Commission. The case is under investigation.
Concerns about hair testing’s accuracy are more than anecdotal. “The consensus of scientific opinion is that there are still too many unanswered questions for [hair analysis] to be used in employment situations,” Edward Cone, the National Institute of Drug Abuse’s leading researcher on the test, said last June. In a recent interview, Cone said that hair testing “is not ready for use yet, where people’s lives are at stake.”
The Society of Forensic Toxicologists stands by its 1990 report, which said: “The use of hair analysis for employees and pre-employment drug testing is premature and cannot be supported by the current information on hair analysis for drugs of abuse.”
D. Bruce Burlington, a doctor and director of the Food and Drug Administration’s Center for Devices and Radiological Health, testified on Capitol hill in July that “Many scientific questions remain […] about the effectiveness of hair testing for detecting drug use.” Not only does hair testing produce falsely positive results, he said, but test results appear to vary according to ethnicity. “Dark hair, blond hair, and dyed hair react differently [from each other], thus creating questions of equity among ethnic groups and genders,” said Burlington.
A US Navy study released by the National Institute of Drug Abuse in 1995 shows that the dark, coarse hair of African-Americans, Hispanics, and Asians is more likely to retain external contamination, such as drug residues absorbed from the environment, and thus is more likely to test positive, even if the person never used drugs. The issue of external contamination is particularly serious for police officers, who may be exposed to drugs during day-to-day law-enforcement operations.
A 1997 study by the National Institute of Drug Abuse reached the same conclusion, warning of “significant ethnic bias” in the results of tests for cocaine specifically. A year later, the institute’s scientists fnaly that melanin, a dark brown or black pigment in dark hair, is the most likely binding site for cocaine.
William Minot, director of marketing communications at Psychemedics Corp., says his company is “very conscious” of concerns that a person’s race may affect the outcome of the test. The current test, he says, is “fail-proof” because Psychemedics not only washes the hair samples thoroughly to remove external contaminants, but also extracts all the melanin before testing. Problem solved? Maybe not. The Journal of Analytical Toxicology reported in its March/April 1998 issue that “removal of melanin from hair digests by centrifugation does not eliminate hair color bias when interpreting cocaine concentrations.”
As scientists continue to study the accuracy of hair testing, the case of Sergeant Duane Adens has attracted congressional attention. Adens—an African-American father of five and a 14-year employee of the Pentagon—was within six years of retirement, and had received the highest possible performance rating in his last job evaluation, when in October of 1996 he was asked by Pentagon investigators to testify against an associate suspected of theft. When Adens declined, saying he knew nothing of the matter, he says he was warned that his refusal would prompt authorities to “play hardball” with him.
A request for a hair sample quickly followed. He refused, but did take a urine test, which came back negative. Three months later, two agents came to Adens’s home with a warrant “for the search and seizure of my body,” he recalls. He then consented to provide a hair sample. Test results from National Medical Services in Willowborough came back positive. Adens was stunned. He says he does not use drugs and had not been exposed to environmental contaminants. Indeed, seven urinalysis tests he had taken between October of 1996 and May of 1998—most of them random tests required by the military—all came back negative. Adens was brought before an Army court martial and, because of the hair test results, received a bad conduct discharge in July 1998. Although he has continued to work for the army, Adens was demoted and removed from the Pentagon.
Last week, he was informed that he will loose his job and his government-subsidized home in less than two months. Dr. George Jackson, a forensic toxicologist with National Medical Services, and the division head for its Criminalistics Department, says he is unfamiliar with the specifics of the Adens case, but is confident in his lab’s overall accuracy. “National Medical Services provides the highest quality of analytical lab services,” he said.
Is it possible that Aden’s results were a `false positive’? While such an outcome is hard to prove, Adens is not the only person who feels victimized by what he says are errors in the hair testing procedures.
Consider two New York cases. In the first, three police department applicants—all white—were told that the lab analysis found evidence of drug use in their hair samples. Outraged, two of the men sent hair samples off for a second test to a different laboratory, and were told their hair showed no drug use. In a second case, nine African-American police officers were dismissed in 1996 due to a positive hair test—even though all nine had passed a series of random urinalysis tests throughout their two-year probation period. Soon after hearing of the positive results, one officer sent another sample of her hair to a different lab. That test also came back negative. (When the officers appeared at a state Labor Department hearing to seek unemployment compensation, they found that `expert testimony’ on hair testing was provided by the director of quality assurance at Psychemedics.)
Even before members of the Congressional Black Caucus, in their May 14 letter to the Army secretary, raised questions about the Army’s use of hair testing in the Adens case, some members of Congress were expressing their discomfort with the procedure’s reliability. Rep. Cynthia McKinney, a Democrat of Georgia, informed Defense Secretary William Cohen in July that she was “exploring a possible legislative remedy to prohibit human hair testing for drugs in the military” given that hair testing has been proven by forensic toxicologists to be “racially biased.”
In October, she received notice from the Defense department saying that “DOD does not plan to use these tests in administrative drug testing programs until this matter is thoroughly studied and adequately addressed.” Meanwhile, Rep. John Conyers, Democrat of Michigan and a ranking member on the House Judiciary Committee, has asked his committee staff to look into the matter.
Whatever Congress does, it will probably not solve the problems of Duane Adens. “One of the things that really, really bothers me is that this is a federal conviction,” says Adens. “I will never be able to get a good job,” he says. “I lose my voting rights. Something I worked hard at for fourteen years is all going to be taken away from me—for no reason at all.”