Substance-Impaired Physicians: Treating Doctors and Protecting Patients

Medical boards have long been concerned with the problem of physicians who abuse alcohol or other drugs. Lately, however, the general public has become more aware of the issue. In 2004, for instance, newspapers reported Vice President Dick Cheney’s personal physician was battling an addiction to prescription drugs at the time he reassured the nation about the state of Cheney’s health.¹  In 2005, The Washington Post published a series of articles leveling harsh criticism at medical boards for their supposed laxity in protecting the public. The title of the first article sums up the tone: “Medical Boards Let Physicians Practice Despite Drug Abuse.”² 

It is little wonder that patients are alarmed. The irony is, in recent years, tremendous strides have been made in the identification, treatment and monitoring of physicians with substance use disorders. The quality and quantity of the treatment currently provided to such physicians is far superior to that of what the average person with an alcohol or drug problem receives. The post-treatment monitoring is much more rigorous, and the consequences for noncompliance are often severe.³  As a result, the outcomes for physicians who participate in treatment and monitoring programs tend to be exceptionally good. While published outcomes vary, some studies have reported extraordinary success rates of 90 percent or higher.^4,5  Unfortunately, these kinds of positive results rarely make the front page.

Yet despite the impressive gains, there remains room for improvement. It is always a tragedy when even one patient is harmed due to a physician’s abuse of alcohol or another drug — and it only takes one such incident to undermine public trust. “The confidence level of the public is very important to the board and to the profession,” says Tom Dilling, former executive director of the State Medical Board of Ohio. “You want to be able to show that the system does work, and when there are problems, that the system will step in and do what’s necessary to protect the public.”

For medical boards, the challenge is to balance the desire to help physicians who have a substance use disorder against the need to safeguard public welfare. “Certainly, I believe this is a treatable disease,” says Michel Sucher, M.D., medical director of the Monitored Aftercare Program run by the Arizona Medical Board. “However, the public also has a right to expect that physicians are unimpaired.” As it turns out, though, the same solution may address both problems. In fact, most experts argue the best way to protect the public is to get physicians into treatment before any harm has occurred and then support them throughout their recovery.

“Even though I’m known as a hardliner, I have complete commitment to this idea,” says Nancy Achin Audesse, executive director of the Massachusetts Board of Registration in Medicine. “We as a society should try to help people who are ill, and that includes physicians. It’s the right thing to treat illness, and it’s also the best thing for patients to get physicians back to safe, healthy practice. How much has society invested in educating and training these individuals? It’s in everyone’s best interest to get these physicians back to good health — not only in humanitarian terms, but also in terms of health care economics and patient access to care.”

The prevalence of dependence on alcohol or other drugs, excluding nicotine, is about 10–15 percent among physicians — similar to that in the general population. Certain special-ties appear to carry an increased liability. “The ones you hear the most about are anesthesiologists,” say Karen Domino, M.D., M.P.H., a professor of anesthesiology at the University of Washington School of Medicine. Anesthesiologists have direct access to potent addictive drugs, so it should perhaps come as no surprise they tend to be overrepresented in drug treatment programs.⁶ 

“Alcohol is the most commonly abused drug by doctors who are in trouble, followed by opioids,” says Gregory Skipper, M.D., medical director of the Alabama Physician Health Program (PHP). Skipper says, out of more than 500 Alabama physicians who received treatment and monitoring, 44 percent had a problem with alcohol and 33 percent with opioids. A similar pattern was found by the North Carolina PHP, where 50 percent of substance-impaired physicians abused alcohol, 25 percent abused opioids, 16 percent abused multiple substances, and smaller percentages abused barbiturates, benzodiazepines, cannabis, stimulants, or cocaine.⁷ 

As noted, outcomes for physicians who take part in treatment and monitoring programs are often excellent. In a recent study by the North Carolina PHP, for instance, 91 percent of substance-impaired physicians ultimately had a good outcome, compared to only 59 percent of physician assistants.⁸  One reason for this discrepancy may be physician assistants as a group don’t have the same ability to pay for long-term treatment. “They just don’t have as deep of pockets as the physicians when they get into trouble,” say Warren Pendergast, M.D., medical director of the North Carolina PHP and coauthor of the study. Another contributing factor may be that physicians have more to lose. “For a lot of physicians, if they’re not able to keep their license or get their license back, they don’t have a lot to fall back on,” says Pendergast. “Many of us don’t have other skills. Medicine is really all we’ve done.” The prospect of losing one’s livelihood and identity as a physician is a major motivator.

Daniel Angres, M.D., director of Rush Behavioral Health in Oak Park, Ill., and lead author of Healing the Healer: The Addicted Physician,⁹  believes temperament plays a role in promoting good outcomes for physicians as well. “The temperament we find elevated in almost all professionals is persistence, because that’s what it takes to become a physician or professional of any kind,” says Angres. “They have the tenacity to stay with something, and people who are tenacious enough to follow a treatment plan as prescribed are going to do better under any circumstances than those who aren’t as high in persistence.”

Of course, no matter how many positive factors a physician has in his or her favor, there’s no denying the difficulty of kicking an addiction and staying sober for the long haul. Domino headed up a 2005 JAMA study that looked at risk factors for relapse among health care professionals with substance use disorders who were enrolled in the Washington PHP. In this study, one-fourth of the 292 participants had at least one relapse. Having a family history of a substance use disorder increased the risk of relapse. So did the combination of using a major opioid (for example, fentanyl, sufentanil, morphine, meperidine) and having a coexisting psychiatric disorder. All three factors together led to a 13-fold increase in relapse risk. With each relapse, the likelihood of a subsequent one grew higher.¹⁰ 

Before the 1970s, the issue of substance abuse and dependence by physicians was treated strictly as a disciplinary matter — that is, when it was acknowledged at all. Over time, though, medicine came to recognize substance use disorders as legitimate illnesses. In 1975, the American Medical Association developed model legislation offering a therapeutic alternative to discipline for physicians with substance use disorders. In the three decades that followed, this approach became the prevailing model. PHPs were established to provide rehabilitation and monitoring to physicians with substance use disorders. Today, such programs have been developed in every state. Most PHPs are sponsored by state medical societies, but some are run by medical boards, private corporations, or other entities.¹¹ 

The conditions PHPs address have expanded over the years. A majority of programs now monitor at least one mental, physical, or behavioral condition in addition to substance use disorders. The populations served have broadened as well. Along with physicians, many programs now offer services to medical students and residents, other health care professionals, and family members. Funding for PHPs comes from a variety of sources, depending on the state. Common sources include state medical societies, medical boards, mal-practice insurance companies, participant fees, and contributions from hospitals and private donors.¹² 

The manner in which a PHP interacts with the medical board varies from state to state. In general, though, physicians can enter PHPs through two tracks. Some physicians are mandated to participate in the PHP by the medical board after a patient care complaint has surfaced or a legal problem has arisen. Other physicians enter the PHP voluntarily. Occasionally, these are true self-referrals by physicians who realize they’re in trouble and reach out for help. More typically, physicians contact the PHP under threat of being reported for disciplinary action if they don’t.¹³  Whatever the initial motivator, in most states, physicians who enroll voluntarily in a PHP can stay anonymous to the board so long as they meet certain criteria.¹⁴  Physicians can also avoid being reported to the National Practitioner Data Bank by a hospital if they take a voluntary leave of absence for treatment so long as no professional review action has been taken against them and they have not relinquished clinical privileges.¹⁵ 

The theory is confidentiality makes it more likely physicians will refer their colleagues or come forward themselves before any harm has been done. Lynn Hankes, M.D., medical director of the Washington PHP, says the best programs “are allowed the flexibility to provide assistance to physicians without reporting them to the board unless certain conditions apply. They get a first shot at trying to fix the problem. But if they can’t fix it, then they have a duty and an obligation to report those individuals to the board.” In Washington, the PHP is required to identify physicians to the medical board if they become dangerous to themselves or others, don’t respond to treatment, or fail to comply with their contract with the PHP.

It’s the medical board’s trust the PHP will hold up its end of the bargain that lies at the heart of an effective working relationship. Says Hankes, “If my medical board thinks I’m hiding clients — if I’m not reporting impairment on the job, if I’m allowing multiple relapses to occur, if I’m letting clients miss urine tests — then my program goes down the tube in two minutes. The most critical element that a physician health program has is trust and credibility with the board.”

Michael Gendel, M.D., medical director of the Colorado PHP, stresses trust must be founded on mutual respect for the distinct but complementary missions of the medical board and PHP. Says Gendel, “We understand the board’s job is to protect the public, and they might at times have to make decisions that are unpopular with some of our clients. They understand our job is to help remedy physician problems that may not be very amenable to discipline in the first place.” He notes, for physicians in the grips of an addiction, “just handing them a letter of admonition isn’t very likely to help them, and it’s not likely to protect the public.”

Two-way understanding is built on frequent communication through both formal and informal channels. “One of the things we do in Colorado that is very helpful is to have what we call a liaison committee,” says Gendel. The committee is composed of representatives from the medical board, PHP, administration, and attorney general’s office. “We share information, hammer out problems, ask questions — and sometimes they’re pointed questions,” Gendel says. “We try to reach some mutual understanding about tensions, which are naturally going to arise. I think tensions can be helpful as long as we respond to them in a respectful way.”

Hankes keeps abreast of medical board developments by attending meetings — not only hearings where he may need to testify for or against a client, but also business sessions. “I want to know what issues the board is facing that could potentially impact our relationship,” he says. When the board holds its annual workshop, Hankes is there as well to educate new members on the basics of PHPs and returning members on any emerging trends they should be watching.

In return, Hankes suggests that medical boards take these steps to facilitate a good working relationship with their state PHPs:

• Educate the PHP about your process. Medical boards operate under a different set of rules and regulations, and each board has its own policies and procedures. Make sure the PHP clearly understands every step of the process from the board’s point of view — from investigation, intervention, and disciplinary action if needed, to assessment, treatment, and follow-up care.

• Do not hesitate to ask the hard questions. “When the PHP submits its monthly, quarterly, or biannual report, the board should really look it over, not just file it away somewhere,” says Hankes. “If there’s something they don’t understand, they should ask about it. For example: ‘Your relapse rate this year was 12 percent. It was 10 percent last year. How do you account for that?’”

• Back up the PHP with noncompliant physicians. “If I report a noncompliant physician to the medical board and the board doesn’t do anything about it, my program is done,” says Hankes. “All of our clients know, while we get them in with the carrot, the stick is picking up the phone and calling the board. For the first few years, until our clients are in recovery and doing what they’re supposed to do because they want to do it, we have to force them.” Without the board’s clout behind it, a PHP loses its best leverage.

• Recognize the PHP’s clinical expertise. Hankes says, “Most medical boards don’t have experts in addiction medicine on them, and some boards try to make clinical judgment calls they’re not equipped to make.”

A decade ago, many decisions about treatment and aftercare for substance-impaired physicians were based largely on anecdotal reports and educated guesses. Today, the evidence base has grown substantially. “The data are there in a way they weren’t seven to 10 years ago,” says Angres. “We know who does well and under what circumstances, and we know what kind of monitoring works.”

With greater knowledge comes increased sophistication in treatment planning. The current trend is away from the one-size-fits-all approach that automatically ships off every physician with an alcohol or drug problem for several weeks of inpatient treatment. “We recognize that some people may not need inpatient care,” says Audesse of the Massachusetts board. “Everyone is required to undergo a full evaluation and comply with the resulting recommendations. If the intake evaluation indicates a need for an inpatient stay, then the physician has to comply with that in order to be accepted into the program. But this is an illness, and we want a medical determination of the appropriate treatment for that particular individual.”

In fact, addiction medicine specialists stress more isn’t necessarily better when it comes to the treatment of substance use disorders. “We use the ASAM [American Society of Addiction Medicine] patient placement criteria,” says Gendel of the Colorado PHP. These criteria are the most widely used medical guidelines on the placement, treatment, and discharge of patients with alcohol or other drug problems. Placement levels, from least to most restrictive, are early intervention, outpatient treatment, intensive outpatient/partial hospitalization, residential/inpatient treatment, and medically managed intensive inpatient. Within these five broad levels of service is a range of more specific levels of care that can be matched to an individual patient’s needs.¹⁶  Gendel says, “We believe physicians deserve the same kind of individualized approach to their care as does the general population. The argument is that, when people get either more or less treatment than they need according to the ASAM criteria, they tend to do more poorly.”

“From the day a doctor enters treatment, he or she is also a patient,” says Audesse. “We as a board don’t want to micromanage the care. We care about knowing that the treatment is working so that we don’t have anyone out there practicing medicine who is actively using alcohol or drugs. But how they get to that point of success we think is best left in the hands of the treatment providers.”

When it comes to aftercare, a five-year contract between the recovering physician and monitoring program is the norm. Typically, this contract covers random drug testing, self-help and support group attendance, work monitoring requirements, and continuing treatment or therapy as needed.¹⁷  Within those general parameters, however, there’s considerable variation from state to state.

Consider the seemingly straightforward issue of how frequently to conduct random drug tests and which drugs to include. A recent study published in this journal found the frequency of drug testing initially required by state monitoring programs ranged to 15 per month to fewer than one per month, with a mean of four monthly tests. Most, but not all, programs varied this frequency over the course of a client’s contract depending on factors such as continued compliance or suspected relapse. The number of tested substances ranged from five to 300. Most programs reported always testing for the client’s drug of choice, but a few did not.¹⁸ 

“We thought some of the states really did not test frequently enough, nor did they test for a wide enough range of drugs for the population they were monitoring,” says Sucher, one of the study’s coauthors. The problem with infrequent testing is that drugs vary in how long they can be detected after ingestion. While most drugs remain in the body long enough to be detected in once weekly or twice monthly tests, some don’t. Alcohol, the most commonly abused drug, can only be detected in the urine by traditional methods for six to 12 hours after use. Physicians are savvy enough to know which drugs have short detection windows and calculate the odds of getting caught if they use a given drug at a particular time. Yet despite the limitations of current programs, Sucher adds “you’re still looking at a group that is better monitored than any other group in the country, with the possible exception of pilots.”

In terms of the breadth of drug testing, Sucher’s program in Arizona is the most inclusive in the nation, routinely testing for 300 drugs. Yet even Arizona doesn’t routinely include certain drugs, such as fentanyl or sufentanil, that require more complicated, expensive tests. Nevertheless, Sucher says, “my personal belief is you always test for a person’s drug of choice, no matter what it was.” This type of additional testing can add hundreds of dollars to the cost, but Sucher believes it’s justified. In most cases, the recovering physician picks up the tab.

The short retention time of alcohol, coupled with its popularity as a relapse drug, makes detection of alcohol use a particular challenge. As a result, the new ethylglucuronide (EtG) test, which checks for a minor metabolite of alcohol, has garnered considerable attention. EtG can be detected in the urine for three to five days after the consumption of alcohol, providing a much longer detection window than traditional tests for alcohol itself. Research indicates the EtG test has excellent sensitivity and specificity.¹⁹  Skipper, who helped bring EtG testing to the United States in 2003, believes as many as 80 percent of state monitoring programs may now be using the test at least occasionally for specific clients.

The mere availability of EtG testing may dissuade some recovering physicians from trying to beat the system. Nevertheless, one major issue that still needs to be resolved is establishing appropriate cutoff levels for the test. “Certain alcohol-containing disinfectants used in operating rooms and dental offices can be inhaled and produce a low level of EtG,” says Sucher. Other nondrinking sources of alcohol exposure include foods, over-the-counter medications, hairspray, aftershave, and mouthwash. If the cutoff level is set too low, you may pick up such incidental exposures as well as actual drinking. But the higher the cutoff, the shorter the detection window. If the cutoff level is set too high, the window of detection shrinks to the point where the test loses its advantage over traditional alcohol tests.

As with treatment planning, the trend today is toward greater individualization in aftercare planning based on a growing understanding of factors that may influence the risk of relapse. “We’re learning more about those individuals who have a difficult time maintaining sobriety even after being treated,” says Martin Doot, M.D., medical director of the Illinois Professionals Health Program. With this knowledge comes the opportunity to intervene proactively with high-risk physicians, who might otherwise have great difficulty stabilizing in recovery.

Gendel says, “I think the most important issue facing us nationally is better follow-up for individuals with concurrent psychiatric and medical illnesses,” which are recognized as important risk factors for recovering physicians. In addition to the stress such conditions can cause, they may also lead to exposure to addictive drugs being used for therapeutic purposes. For PHPs, then, better tailoring the care provided to recovering physicians who have coexisting illnesses and other relapse risk factors is apt to be a top priority for many years to come.

For medical boards, the perennial question is determining how many chances at recovery a physician should be given before his or her chances run out. Arizona, for one, has adopted a tough stance on multiple relapses. “The policy here is basically three strikes and you’re out,” says Roger Downey, public information officer at the Arizona Medical Board. “The first strike is the first disciplinary action. You’re put into the monitored aftercare program [by the board, rather than by self-referral] and placed on probation. If you’re still in the aftercare program and you relapse, you’re put on probation for a longer period of time, and you sign an agreement that basically says, ‘If I do it again, I must surrender my license.’ Then, if you do relapse again, you either surrender your license or have it revoked.”

Ohio, in contrast, has adopted a case-by-case approach. “You could be out on strike one, or you could potentially make it through three strikes, depending on things like the severity of the relapses and the length of time in between,” says Dilling. Another factor that the Ohio board takes into consideration is a physician’s response to his or her own relapse. If the physician is willing to recognize the problem and get back into treatment, the board sees that as a positive sign.

Physicians who relapse have much to lose, yet some inevitably do so, because addiction is such a powerfully self-destructive disease. This is an issue boards around the country undoubtedly will continue to confront. “It’s a balancing act,” acknowledges Downey. “You have to protect the public while at the same time allowing the doctor an opportunity to become well again.”