Drug testing after adverse events is a controversial idea
Some say it's time to test clinicians, but expect resistance
Mandatory drug and alcohol testing is routine for pilots and train conductors after a crash, so shouldn't healthcare providers also test physicians after an adverse event? This idea is being floated by some notable names in the industry who say that even if the idea seems radical, it could improve patient safety.
Post-event testing should be seen as part of an organization's overall approach to dealing with impaired physicians, says Peter Pronovost, MD, PhD, FCCM, a practicing anesthesiologist, critical care physician, professor, senior vice president of Johns Hopkins Medicine in Baltimore, MD, and senior vice president and director of the Armstrong Institute for Patient Safety and Quality at Johns Hopkins.
"There seems to be a problem, as with in all of society, with physicians who are impaired, and we need to have an effective strategy for both supporting those physicians and protecting patients," he says. "Using routine surveillance in conjunction with testing after adverse events would be an effective strategy."
Pronovost and colleagues proposed post-event testing, along with pre-employment and routine drug testing, in a commentary published online recently in The Journal of the American Medical Association. (See the story below for more on that commentary.)
Those who support the idea concede that there is no research showing a link between adverse events and drug or alcohol impairment. Pronovost argues that "given the prevalence of substance abuse in healthcare and the number of adverse events, it would be hard to imagine there is not an overlap somewhere in those two circles."
Another proponent of the idea is the highly regarded medical ethicist Arthur Caplan, PhD, director of the Division of Medical Ethics at New York University's NYU Langone Medical Center in New York City. He says testing is needed partly because no one knows if there is a causal connection between alcohol and drug use and adverse events.
"In other industries like transportation, they have enough evidence to know that drug and alcohol use plays a role in accidents, and I don't see any reason to think that would be any different for doctors and nurses," Caplan says. "We know that the availability of drugs makes abuse possible for many healthcare professionals, and a particular risk in some specialties. So it's not like we don't know that doctors sometimes abuse drugs."
Near misses also should prompt drug and alcohol testing, Caplan says. Physicians and other clinicians certainly will resist the idea of mandatory testing after an adverse event, Caplan says, but he sees nothing about the healthcare industry that should exempt it from the same standards used in other workplaces.
"A lot of people think we're making an accusation when you test, but it's really an inquiry," he says. "It may go better in the context of testing after an adverse event than if you implemented routine random testing, but any kind of drug and alcohol testing has a stigma. Doctors want to be trusted, but I go by the old Cold War adage that you trust, but verify."
It is believed that no healthcare providers currently require drug and alcohol testing after adverse events, but interest in post-event testing has grown since Pronovost and his colleagues first proposed the idea. Leaders at his own institution, Johns Hopkins, and other facilities have contacted him to discuss the logistics of setting up such a program.
"We are in discussion phases here, and several other places are considering it also. We're hearing from a lot of risk managers who say this just makes sense," he says. "Even routine testing is not done in healthcare right now, and that might be more easily implemented and would pick up a lot of problems even if you don't do testing after adverse events."
False positives could be devastating to a physician or nurse's career, but Pronovost says testing has become so reliable that false positives virtually can be eliminated by using more advanced techniques on samples that initially indicate impairment. Defining an adverse event is another challenge, but Pronovost says providers could start with preventable deaths or sentinel events. Those are clearly adverse events and could be the first threshold for triggering post-event testing. After working out the kinks with that level of testing, a provider might lower the threshold to include other adverse events, he suggests.
Implementing a testing program will require extensive education about the purpose, Caplan says. Simply announcing that you will begin testing after adverse events will result in huge pushback, he says. There also can be legal limitations to such testing and potential liability for the hospital. (See the story below for more on the legal issues.)
"Given the concern about quality of care and patient safety in healthcare institutions, I can't imagine that we wouldn't want to push in this direction," Caplan says. "I see it as part of a quality control push rather than a punitive measure or trying to respond to a drug epidemic."
Another prominent ethicist takes a dim view of the idea. John Banja, PhD, medical ethicist at the Center for Ethics at Emory University in Atlanta, says he opposes drug and alcohol testing after adverse events because there is no evidence to suggest impairment is linked to errors and bad outcomes.
"I don't like the proposal. It unfairly singles out doctors for drug and alcohol testing without any evidence there is a causal relationship," Banja says. "The literature clearly shows that when you have one of these nasty sentinel events go down at your hospital, it is almost always the result of multiple people making multiple mistakes."
Given that fact, a testing program should include dozens of people who were involved with the case, Banja says. That system would prove impractical and costly, he says. If it is done at all, the testing also should be extended to near misses in which errors were made but the patient was not harmed, he says, which would further expand the program.
"You can imagine how many drug tests you would be doing in a hospital every day," Banja says. "Think of a 500-bed hospital and how many errors and near misses occur, and how many people are involved in each of those. You'd be testing everyone all the time."
Banja also suspects that the testing program would find very few cases in which the adverse events could be blamed on an impaired physician or other clinician.
"People simply make human mistakes, and largely because they are attempting to satisfy productivity goals," Banja says. "A more productive strategy would be to encourage people to speak up when they see someone acting in an unsafe way. We know that people often will see something that makes them uncomfortable, but they don't speak up or they just keep it under the radar and talk among themselves."
Testing needed at least after sentinel events, doctors say
Hospitals should require mandatory alcohol-drug testing for clinicians involved with unexpected deaths or sentinel events and possibly more often than that, according to a commentary published online recently in The Journal of the American Medical Association (JAMA) by Peter Pronovost, MD, PhD, FCCM, a practicing anesthesiologist, critical care physician, professor, senior vice president of Johns Hopkins Medicine in Baltimore, MD, and senior vice president and director of the Armstrong Institute for Patient Safety and Quality at Johns Hopkins, and colleagues.
The authors recommend in their commentary that hospitals take several steps as a model to address what they consider an overlooked patient safety issue. They are:
The steps could be limited to hospitals and their affiliated physicians at this time, because hospitals have the infrastructure to conduct adverse event analysis and drug testing, note the authors. Hospitals also have the governing bylaws to guide physician conduct, and The Joint Commission can help with establishing standards, the authors add.
In cases in which a physician is found to be impaired, a hospital could "suspend or revoke privileges and, in some cases, report this to the state licensing board," the authors write. Impaired physicians would undergo treatment and routine monitoring as a condition for continued licensure and hospital privileges.
The full commentary is available online at http://tinyurl.com/kn65jmp. A JAMA subscription is required.
Many legal issues complicate drug and alcohol testing
Drug and alcohol testing after adverse events will come with a ton of potential legal problems. If you plan to implement such a program, think carefully about how you can construct it to reduce your liability risks and reduce friction with the medical staff.
It is possible to write physician bylaws to require drug and alcohol testing after an adverse event, but there will be many complicating factors that can threaten the effectiveness of any testing and expose the hospital to liability, says Michael R. Callahan, JD, a partner with the law firm of Katten Muchin Rosenman in Chicago.
"The only way to implement this is through the medical bylaws. If it is not in the bylaws, your physicians are going to say that they don't have to comply, and they will be right," Callahan says. "Your first real hurdle will be getting this requirement into the bylaws."
Another challenge will be defining what constitutes an adverse event and triggers the testing requirement. Only a strict, specific definition would have any chance of being entered into the bylaws and surviving a legal challenge, Callahan says.
"If you tried to say you're going to test after all adverse events, and you don't specify what it means and limit it to only the most justifiable conditions, you will have a rebellion among your medical staff," he says. "You have more leverage with your employed physicians, because you can impose more employment qualifications on them and set the bar higher."
Some attorneys see merit in the idea. Testing after an adverse event should be part of the post-event analysis, says Kathryn R. Coburn, JD, an attorney with the law firm of Kobrick and Wu in Los Angeles. She advises drawing up a contract in which the physician agrees to such testing before being hired or credentialed by the hospital, with carefully defined terms. The policy must be applied uniformly, and adhere to the triggers outlined in the contract.
"It also is important to be aware of the Americans with Disabilities Act [ADA]," Coburn says. "It does not protect current drug and alcohol users, but the definition of 'current' is a little fuzzy. You want to be very careful that people with a known problem are treated no differently. They cannot be treated more often or under different parameters."
The other big question is whether testing would help or hurt the provider in malpractice litigation, says Michael Thompson, JD, an attorney with the law firm of Reed Smith in Philadelphia. The results of the testing almost certainly would be available to the plaintiff, he says. State legislation could protect the test results as part of peer review, but there is no such protection now.
"If a hospital goes with this policy, you have to expect that those test results are going to be discoverable by the plaintiff," Thompson says. "And if you have a case where you have test results showing the doctor was impaired when he or she made a serious error and harmed the patient, that's going to be a very difficult case to defend."
- John Banja, PhD, Medical Ethicist, Center for Ethics, Emory University, Atlanta. Telephone: (404) 712-4804. E-mail: email@example.com.
- Arthur Caplan, PhD, Director of the Division of Medical Ethics, New York University NYU Langone Medical Center, New York City. Telephone: (646) 501-2739. Email: firstname.lastname@example.org.
- Peter Pronovost, MD, PhD, FCCM, Senior Vice President, Johns Hopkins Medicine, Baltimore, MD. Telephone: (410) 502-3231. Email: email@example.com.