Note: As I’ve promised, today I’m going to take a look at the article that ZBicyclist mentions in a recent post about a new urine test that is being hyped as a foolproof way to detect alcohol use. Due to copyright considerations, I can’t reproduce the article verbatim, but will quote important passages, paraphrase other sections and discuss what the author is getting at. Along the way, I’m sure I’ll add my own perspective, too. Those comments will appear in red.
Sometimes tests work too well, as Kevin Helliker discovered when he wrote “A Test for Alcohol — And Its Flaws” in the August 12, 2006 edition of The Wall Street Journal. In the article, he talks about a new urine test that “detects Sunday’s gin in Monday’s urine but … may be ensnaring some innocent people too.”
Helliker tells the story of a Harrisburg, Pa. nurse, Nancy Clark, whose nursing license has been suspended for flunking a new type of urine test. Clark is a recovering alcoholic who, the state says, violated an agreement to forego drinking.
The new analysis “is the gold standard of testing,” Shawn E. Smith, a lawyer for the state, argued during a June 7 hearing here on Ms. Clark’s fate.
Ms. Clark counters: “I didn’t drink.”
Helliker goes on:
Throughout history, few questions have prompted more lies than, “Have you been drinking?” For decades, the truth has been obtainable through urine tests and breathalyzers. But since alcohol dissipates from the system in a matter of hours, that truth always has been as fleeting as drunkenness itself. Whether a person is drunk this moment can be documented. But how about last weekend?
Turns out, there’s a new test, known as EtG, which is being used to monitor the urine of millions of Americans for the presence of metabolites of alcohol. This test is being used to keep tabs on people who aren’t supposed to drink — and must provide regular urine samples to prove it.
The test gets its name from ethyl glucuronide, or EtG, a metabolite of alcohol, which remains in a person’s system for up to 80 hours. About 10% of people who had been passing conventional urine tests are now testing positive for alcohol when subjected to the EtG test, according to test administrators.
This test should be of interest to Floyd Landis fans, as it could be used to determine whether traces of alcohol were still in his system after his night out for drinks before stage 17 (assuming there’s any leftover samples to test). Although Landis has said he doesn’t want to focus on this aspect too much, if alcohol can really decrease the amount of epitestosterone in an athlete’s system, the finding of EtG would certainly help bolster his case, at least a bit.
According to Helliker’s article:
About 20,000 urine samples a month are now being tested for EtG, which was introduced about two years ago, and that growth is continuing exponentially. At $25 — compared with about $7 for a standard drug screen — EtG represents an important new source of revenue for the urine-testing industry, whose largest players include Quest Diagnostics Inc. and National Medical Services.
LITTLE ADVERTISED, though, is that EtG can detect alcohol even in people who didn’t drink. Any trace of alcohol may register, even that ingested or inhaled through food, medicine, personal-care products or hand sanitizer.
The test “can’t distinguish between beer and Purell” hand sanitizer, says H. Westley Clark, director of the federal Substance Abuse and Mental Health Services Administration’s center for substance-abuse treatment. His office intends to study EtG and issue a statement on its use in the fall. “When you’re looking at loss of job, loss of child, loss of privileges, you want to make sure” the test is right, he says.
Doesn’t this lack of specificity strike you as problematic? It does me. If you can’t be certain it came from alcohol abuse, it’s not a very good test. Just as the WADA finding that diet can influence the amount of carbon13 in various hormones when conducting CIR/IRMS testing casts a certain amount of doubt on how specific a test it is for anti-doping screenings.
As Helliker notes, EtG is a molecule that any laboratory can identify, so nobody owns the test. Urine-testing companies are aggressively marketing it to a number of industries, boards or companies that license workers in health care, aviation, law and other professions. These groups routinely require urine monitoring of professionals who are recovering addicts. Some courts use the testing to monitor drunk driving offenders.
As long as the tests aren’t used to monitor federal employees and aren’t sold over the counter, these tests don’t require federal approval. The urine-testing companies leave it up to the courts and licensing boards to figure out how to use test results.
Helliker goes on:
“It’s a powerful tool,” says Doug Lewis, president of United States Drug Testing Laboratories Inc. near Chicago. “But it’s only a tool.”
Some in sobriety enforcement contend any alcohol, however ingested, could trigger a relapse in recovering addicts. “They must abstain from alcohol in any form,” says Kevin Knipe, manager of a Pennsylvania state program for monitoring physicians, nurses, pharmacists and others.
Yet critics worry that growing acceptance of the EtG test is punishing those who haven’t relapsed or aren’t problem drinkers. They argue it’s unfair to demand addicts produce urine free of any trace of alcohol because there is no comprehensive list of products that contain it. Mouthwash and cold medicine are sources of alcohol. It can also be found in pastries, perfume, salad dressing, insecticide, ripe fruit, lunch meat, vanilla extract, ice cream and automotive fuel.
Such critics have gained an unusual ally: the physician who pioneered EtG testing in America. “Use of this screen has gotten ahead of the science,” says Gregory Skipper, an Alabama addiction specialist and recovering addict. He says he has received about $10,000 in consulting fees, mostly from urine-testing firms, in connection with the test.
As Helliker writes, “An unfair monitoring test could dissuade addicts from entering monitoring programs voluntarily. ” The same may well be true for athletes who might not compete due to the current state of the anti-doping regime. He continues with more about Nancy Clark:
After Ms. Clark, the Pennsylvania nurse, confessed her addiction to her supervisor, she entered a state- monitored program, requiring attendance at two 12-step meetings a week. On average, she has attended three a week, her sponsor says. She also has to phone a toll-free number each weekday morning to find out whether a urine sample is required that day.
In five years, she has never failed to call or submit a specimen, according to court testimony. She pays the $120-a-month cost of monitoring. According to testimony, after getting sober, Ms. Clark joined a church, entered its ministerial program, started a program to entertain patients and raised three children.
In 2004, the state of Pennsylvania introduced the EtG test. After flunking it, she says she read the label of every imaginable product — edible and inedible — only to flunk it again. “I don’t know what else I could have done,” says Ms. Clark, 49, who has practiced nursing since 1978. She recently passed a polygraph test asking whether she has drunk alcohol.
In January, the state suspended her license, costing Ms. Clark her job of eight years as an assessor of patient care in a hospital. She is spending thousands of dollars to fight her suspension. Her witnesses include her boss at the hospital, who describes Ms. Clark as indispensable, and Dr. Skipper, who doesn’t believe her positive EtG scores represent proof of drinking.
So an innocent person is out of work due to a test that is not specific enough to determine what the testing companies say it does. Not terribly fair in my book.
As Helliker writes, the purpose of urine testing has always been to prevent — or provide early warning of — relapse. And preliminary evidence, he adds, suggests monitoring programs may increase success: Addicted physicians subject to testing relapse at a rate of 25% over five years, compared with a rate among the general population of 75% in a single year. He goes on:
Despite innocent positives, some courts and licensing boards are digging in their heels. They are arguing that a participant “must produce a negative urine” sample, says Mr. Lewis, the drug-testing company president. “Trying to argue that you’re an innocent victim — good luck.”
Aint that the truth! Trying to prove you’re an innocent victim is difficult in any anti-drug or anti-doping testing regime. Helliker continues:
Indeed, the state of Pennsylvania isn’t saying that Nancy Clark drank, only that she failed to produce clean urine. “This case is not about relapse,” said Mr. Smith, the lawyer for the state, in the June argument against her appeal.
Before the EtG test came along, it was possible for a crafty individual to deceive monitors. Alcohol dissipates from the system in a matter of hours. So a recovering alcoholic could drink on the weekend or even during the work week if he or she stopped early enough. By the time the person’s urine would be collected, any trace of booze would be out of his or her system.
Catching such drinkers wasn’t the primary interest of Dr. Skipper when he set about searching for a better screen for alcohol. As director of Alabama’s monitoring-and-assistance program for addicted physicians, he wanted to find more convincing proof of sobriety.
Dr. Skipper, a 56-year-old internist, had a personal interest in such a test: He is a recovering narcotics addict. After entering a chemical-dependency program in 1981, he underwent additional training, received a certification in addiction medicine, and began caring for others. He had a relapse in 1990, and confessed to stealing painkillers from the medical facility where he worked. His license was suspended for a year.
Dr. Skipper says he has been clean and sober since Nov. 16, 1990. He continues submitting voluntary urine samples, and advises other recovering physicians to do the same.
In 2001, Dr. Skipper, along with Swiss psychiatrist Friedrich Wurst, conducted research that showed a strong connection between alcohol consumption and the presence of EtG in a person’s urine. This research created a great deal of interest among U.S. urine testing laboratories, as well as their clients.
Due to very limited funds, Dr. Skipper and his colleagues conducted limited research, mostly focused on whether drinkers could somehow avoid producing EtG. And with the exception of some rare cases, it turns out that drinkers can’t avoid producing EtG. What they didn’t focus on was whether EtG could show up from contact with other potential sources, like mouthwash, food, or hand sanitizers.
After examining the results for about 1500 specimens, Dr. Skipper believed that any level of EtG above 100 ng/ml of urine proved that the subject had consumed alcohol. Some urine-testing companies went further, saying any EtG in the urine is proof positive of drinking. A news release for Quest Diagnostic’s Northwest Toxicology unit boldly stated, “EtG is not detectable in urine unless an alcoholic beverage has been consumed.” Helliker state that according to a Quest spokeswoman, before introducing the test, the company had “evaluated it on some people here at the lab.”
So Dr. Skipper set a very low threshold, believing that it would be a definitive indication of alcohol use. Unfortunately, as we’ll see in a bit, he was wrong. But now that his “standard” is out there, others have latched onto it and won’t let go. And others have taken this to an extreme, saying that any amount is a positive test. This begins to sound like the “strict liability” standard for the anti-doping process. If it’s there, you’re guilty. Even if you’re not.
Helliker talks about a Lorie Garlick, a California pharmacist who is in recovery from narcotics addiction. He notes:
To comply with an agreement with the California Board of Pharmacy, [the pharmacist] says she attended 12-step meetings and avoided drugs and alcohol. When her urine tested positive — twice — for EtG in 2005, she went online and discovered the new screen was touted as definitive proof of drinking.”The first thing that went through my head was that there must have been a mix-up at the lab — my urine got swapped with somebody else’s,” says Ms. Garlick, who says she never drank. Her license has been suspended and she hasn’t worked in more than a year. The California Board of Pharmacy didn’t return phone calls.
She found a Web site with a chat room for addicts claiming to have been victimized by EtG. Its founder: Dr. Skipper. “I’d been hearing from people saying they were innocent, and I wanted to research that,” he says.
Dr. Skipper started to think that incidental exposure to other products could lead to higher levels of EtG than he had expected. One product that he suspected could cause elevated levels of EtG was sweeping through hospitals nationwide: hand sanitizers. Call it the Purell effect.
Although the federal Centers for Disease Control and Prevention recommends the use of alcohol-based sanitizers in hospitals, schools and day-care centers, the issue of whether alcohol in these sanitizers could be absorbed into a person’s system hasn’t been deeply studied, Helliker quotes physician John Boyce, who served as chairman of the CDC’s hand-hygiene task force, as saying. And Pfizer Inc., owner of Purell, the dominant brand, also hasn’t studied the issue.
Helliker’s article goes on:
A small study of 24 people that Dr. Skipper helped perform found that use of Purell could result in EtG showing up in urine. It concluded alcohol in the sanitizer can enter the body through inhalation, rather than through the skin. That study, presented in May at the scientific conference of the American Society of Addiction Medicine, hasn’t been published or peer-reviewed.
Interesting. So Dr. Skipper found that there were possible causes of false positives. I wonder how much study of false positives has been done on CIR/IRMS? No, wait, USADA is looking for people to study CIR/IRMS and validate the procedure (as “the king” noted in a comment on another post at this site). And WADA has research that shows a 3+ percent false positive rate on the tests just due to diet.
Dr. Skipper conducted an experiment, with the pharmacist as the subject. At his suggestion, she entered a treatment center and stayed two days under the supervision of counselors instructed to search her and her belongings for any products containing alcohol. Ater washing her hands repeatedly with Purell during the day, a urine sample she provided gave an EtG score of 770 (each morning, a sample returned a minimal or non-existant value for EtG). The result is more than seven times Dr. Skipper’s original cutoff value that he believed represented proof of alcohol consumption. A single alcoholic drink could produce a peak EtG level of perhaps 6,000 nanograms, 60 times Dr. Skipper’s original cutoff.
So clearly the original threshold, 100 nanograms, is way too low.
Dr. Skipper believes that this doesn’t affect the test’s value. A negative EtG score would still provide definitive proof of sobriety. But, Dr. Skipper says he has always believed a low-level score should be considered a warning sign, not proof of guilt. Helliker goes on to say, “In [Dr. Skipper’s] experience, most secret drinkers offered no defense when confronted with EtG scores.”
Dr. Skipper now believes that incidental exposure to alcohol could result in EtG levels as high as 1,500 nanograms, much more than the threshold value that was costing many addicts their professional licenses.
In August 2005, Dr. Skipper wrote an open letter to state boards that monitor health- care workers, calling on them “to refrain from taking action against an employee or licensee based on urine EtG testing alone.”
Unfortunately, Dr. Skipper’s warnings have been largely ignored. Licensing boards and other organization still punish recovering addicts for producing positive EtG scores — even very low level scores.
Most laboratories are now recommending cutoff levels of 250 or 500 nanograms. But they say it is up to their clients to decide whether a level above that amount represents proof of drinking or some other exposure to alcohol.”The industry needs to do some population studies to get a sense of what a reasonable cutoff is,” says Dr. Clark of the federal substance- abuse office.
Establishing an EtG cutoff high enough to spare the innocent may involve allowing the guilty occasionally to slip through, concedes Dr. Clark. But the only ethical option is “to err on the side of due process,” he says, noting that a true addict will get caught soon enough anyway.
Helliker ends the story talking about a home-health nurse in Wichita, Kan., who ran afoul of the Kansas state disciplinary board when she the mistake of drinking a martini at a going-away party for a colleague, even though she had a patient to visit after the party.
She was caught, confessed and put into a program that included monitoring of EtG in her urine. One day she received a notice of a positive along with the following note, Helliker reports:
“Please be aware that [over-the-counter] medications that contain alcohol can cause a positive alcohol screen,” wrote Mary Carder, executive director of the Kansas Nurses Assistance Program, which oversees monitored nurses in Kansas. “You should also be cautious of any foods, sauces, pastries etc. that may have alcohol.”
The whole experience, including a positive EtG test when she hadn’t been drinking and the warning she received that another positive test and her license would be suspended has left her so worried that she’s going back to school to pursue another career.
As you can see, this test works too well, and it’s often being used without any critical thinking about what the results actually mean and what may have caused those results. It’s an awfully good thing nothing like that happens in cycling, isn’t it? 😉