The ASAM Principles of Addiction Medicine 5th Edition

41. Physician Health Programs and Addiction among Physicians

Paul H. Earley, MD, FASAM












The available research about addiction among physicians and physician health programs (PHPs) is extensive and has been well documented in several excellent overviews (16,127). Bissell and Haberman (7), Angres et al. (8), Nace (9), and Coombs (10) have written complete texts about addiction in physicians and other health professionals. Physicians are a convenient population to study; they are accessible both prior to and after treatment and are articulate about their disease. Research on physician addiction elucidates the natural course of addiction in a highly regulated and monitored population. At the same time, physicians differ from the general population in terms of education, income, and regulatory oversight; therefore, conclusions about the efficacy of addiction treatment among physician-patients cannot simply be generalized to the population at large. However, the highly structured and consistent treatment model developed for the care of this population does provide clues for treatment improvement. DuPont et al. (11) have suggested that a model that utilizes the PHP experience should be an integral part of the gold standard for effective treatment.


We have 20 years of debate about the actual and changing prevalence of addiction among physicians (12). Kessler et al. (13) reported that 3.8% of the general population at any given time has any substance use disorder and 1.3% meets criteria for alcohol dependence and 0.4% for drug dependence. Lifetime prevalence has been estimated at between 8% and 13% in the general population. Studies that attempt to determine the prevalence of addiction in physicians are based upon anonymous questionnaires (12,1421). Hughes et al. (19) reported a lifetime prevalence of alcohol abuse or dependence and drug abuse or dependence in physicians at 7.9%, somewhat less than the percentage reported in the general population by Kessler (13). However, methodologic differences may account for the observed differences. The Hughes study surveyed 9,600 physicians by mail with a lower response rate (59%) and relied on honest and denial-free reports by the physician self-report; the Kessler general population study utilized face-to-face interviews with trained interviewers.

In 1970, Vaillant et al. (22) reported on the types of substances physicians use. At that time, he noted that physicians were just as likely to smoke cigarettes and drink alcohol as the general population but more likely to take tranquilizers and sedatives. In a more comprehensive study 29 years later, Hughes et al. (17) noted that physicians were less likely to smoke cigarettes than nonphysicians and more likely to consume benzodiazepines and opioids. The change in cigarette use was presumably due to increasing medical data about the health risks and changes in physician attitude regarding tobacco. The decrease in smoking was also found by Mangus et al. in 1998 (23). Hughes et al. (19) stated that physicians drink more alcohol than the general population; the authors attributed this in part to their higher socioeconomic status. They also noted that 11.4% of physicians had used unsupervised benzodiazepines and 17.6% reported the unsupervised use of opioids. Vaillant (24), in his commentary on the Hughes study, rang an alarm bell by stating “physicians are five times as likely [than the general population] to take sedatives and minor tranquil-izers without medical supervision.” The use of opioids and minor tranquilizers commonly begins prior to or in medical school, since medical students use more of these drugs than age-matched cohorts (25). Clark examined substance abuse in medical students using a 4-year longitudinal study (26). Eighteen percent met the study’s criteria for alcohol abuse in the first 2 years of medical school. They reported that a family history of alcoholism was associated with alcohol abuse in the medical student.

Another view of physician abuse of alcohol and drugs can be derived from complaints reviewed by state medical boards. Morrison and Wickersham (27) noted that 14% of board disciplinary actions were alcohol or drug related, and another 11% were due to inappropriate prescribing practices—many of which are also addiction related. In 2003, Clay and Conatser (28) reported similar disciplinary rates, with 21% due to alcohol and drug issues and 10% due to inappropriate prescribing or drug possession.

Alcohol- and drug-related work impairment was the primary impetus for the formation of state PHPs in the United States and continues to account for the majority of physician impairment cases seen by most PHPs today (2). David Canavan, MD started the first PHP (New Jersey) in 1982. Since that time, “all but three of the 54 US medical societies of all states and jurisdictions had authorized or implemented impaired physician programs” (29). The most recent PHP (Georgia) opened its doors in 2012. In 2008, California moved against this trend by dissolving its PHP (30).

Ethnic variation in substance abuse in the general population is described in the National Epidemiological Survey on Alcohol and Related Conditions: Whites, Native Americans, and Hispanics have a higher prevalence of dependence than Asians; but no published data about physician addiction have been reported to date using ethnicity as an independent variable.

In summary, though the prevalence of addiction to all chemicals appears to be about the same as in the general population, currently physicians consume less tobacco and more opioids and sedatives. Research data suggest that physicians consume more alcohol than the general population.


Berry (31) has suggested a bimodal distribution of age at first presentation for treatment; physicians in training and early practice comprise the first wave, and physicians in mid- to late career comprise the second. Talbott et al. (32) reported a decrease in the age of presentation in treatment from 51 to 44 years between 1975 and 1985. In a 2008 analysis of more than 1,400 medical students, residents, and physicians at the same southeastern treatment program, Earley and Weaver (unpublished) noted an age range from 25.3 to 83.7 years, with a median age of 45.8, the ages distributed in a bell curve (Fig. 41-1).


FIGURE 41-1 Distribution of physician age at presentation to treatment.


Males account for the majority of treated physician addiction cases, with reported ratios approximately 7 to 1 (33). This contrasts with the 3-to-1 male-to-female ratio in the physician population at large (34). Although fewer females than males have drinking problems, female physicians are more likely to report problematic drinking by the end of medical school (5) and are more likely to have alcohol problems later in life than their nonmedical counterparts (35). At intake into one of four PHP programs, female physicians were more likely to be younger and to have medical and psychiatric comorbidity (36). Female physicians were more likely to have past or current suicidal ideation and were more likely to have attempted suicide regardless of whether or not they were under the influence at the time. Wunsch et al. (36) report that female physicians are more likely to abuse sedative/hypnotics than are men. Interestingly, woman physicians are the subject of more severe sanctions by medical boards than their male counterparts (27).


Bissell and Jones (35), writing in 1976 about 98 physicians, were among the first to systematically study this cohort. Using a follow-up questionnaire of physicians in Alcoholics Anonymous, she noted that psychiatrists and emergency medicine physicians were overrepresented in Alcoholics Anonymous (overrepresentation defined as a percentage of a cohort that is higher than predicted by the percentage of that cohort in the population of physicians at large). Hughes et al. (17) surveyed 5,426 physicians regarding substance use through an anonymous survey. A self-report of substance abuse or dependence to alcohol or other drugs was highest in psychiatrists and emergency medicine physicians and lowest in surgeons and pediatricians. This question did not break down the substance used or its legality.

A synopsis of the literature on addiction rates by spe-cialty appears in Table 41-1, which covers multiple authors and modes of analysis. The combined literature looks at the breakdown by specialty from multiple angles (treatment presentation, self-report, and medical board and PHP data); the data consistently validate that psychiatry and emergency medicine physicians have higher rates of substance abuse. Table 41-1 also suggests that family practice physicians might be overrepresented, and pediatricians and pathologists appear to have a lower prevalence of addiction.



PHP/MB, physician health program or medical board record study.

The problem of addiction in anesthesiologists continues to attract research and debate. Lutsky et al. (16) noted that anesthesiologists were heavier users of marijuana and psychedelics when compared with medicine and surgery physicians but suggested caution in the interpretation of these data owing to age differences between the medicine and surgery cohort and the anesthesiology cohort. Talbott et al. (32) note that anesthesiologists account for 5% of all physicians, yet they account for 13% of all physician-patients in a residential treatment program. In contrast, Hughes et al. (18) noted a low overall rate of substance use in anesthesiology, both in residency and after completing training (17). In their study, Lutsky et al. (16) found that the use of fentanyl (and its relatives) occurred only in anesthesiologists. Hughes et al. (17) noted a trend toward more frequent use of the major opioids, such as fentanyl, in anesthesiologists, but the finding did not reach statistical significance. Propofol use, although infrequent, appears to plague health care professionals who work in settings with high access.

Gold et al. (41) and McAuliffe et al. (42) have recently hypothesized that anesthesiologists may be sensitized to opioids and propofol through the inhalation of picograms of these potent agents in the operating room air. Assays of operating room air detected these agents, especially when taken near the expiration point of the anesthetized patient. This hypothesis rests on an uncertain foundation (the assumption that the quantities of these agents are sufficient to produce sensitization and that the resultant sensitization directly contributes to the etiology of addiction) but does introduce additional avenues of research.

With one significant exception in the data (17), anes-thesiologists appear to be frequent users of highly potent opioids and are strikingly overrepresented in treatment settings. Access to large quantities of these high-potency opioids (and other drugs) in the day-to-day practice of anesthesia is the most likely culprit for the prevalence of anesthesia personnel in treatment settings.



Two types of studies are used to assess the types of drugs abused by physicians: anonymous questionnaires (121521) and self-reports of drugs of choice of physicians as they appear in treatment or monitoring programs (32). Both types of research underscore that alcohol is, as expected, the most frequent primary substance of abuse by physicians, just as it is in the general population.


Tobacco dependence has been suggested as a risk factor for alcohol and other drug dependence in physicians (43) as in the general population (44). Tobacco use in physicians has decreased over time. Vaillant (22) reported that 39% of physicians reporting 10 or more cigarettes per day in 1953, decreasing to 25% in 1968 (22). Nelson et al. reported that smoking among physicians declined from 18.8% in 1976 to 3.3% in 1991 (45). In a 1996 study, Mangus reported 2% of medical school graduates were current smokers (23). From the earlier data, emergency medicine and surgery physicians are twice as likely to smoke as were other physicians (17). Preliminary data from Stuyt et al. (46) strongly correlate the continued use of tobacco with subsequent relapse into other drug or alcohol use, underscoring the relationship of tobacco use with addiction in physicians.


Opioids are the second most frequently abused substance by physicians presenting for treatment (47). This finding has been remarkably stable over time, but the type of opioids used continues to change. Hughes et al. (17) differentiate opioid use into the major opioids (morphine, meperidine, fentanyl, and other injectable narcotics) and the minor opioids (hydrocodone, lower dose forms of oxy-codone, codeine, and other oral drugs). Distinguishing in this manner, they reported that family practice and obstetrics and gynecology specialists have a higher probability of abusing minor opioids. When compared with all physicians, the study reports that anesthesiologists were less likely to use minor opioids, with a trend toward an increased use of major opioids. If one assumes that use of major opioids results in a more aggressive manifestation and progression of addiction, this would partly account for the overrepresentation of anesthesiologists over other specialties in physician treatment programs (32). Several authors (17,19,48) posit that exposure to drugs of abuse in the workplace leads to higher abuse of those workplace drugs. This postulate is supported by the frequent abuse of major opioids by anesthesiologists. In a similar manner, family medicine and obstetrics and gynecology physicians are frequent prescribers and use more minor opioids than other specialties (17).


In one study, professions that use cocaine medicinally (ophthalmology, head and neck surgery, plastic surgery, and oto-laryngology) had a (not statistically significant) trend to higher cocaine use (17). Hyde and Wolfe (49) noted that when cocaine is abused by surgery residents, it often comes from hospital sources, but this study is from older literature and does not reflect the current pharmacy controls over this substance.


One hypothesis of substance misuse among physicians suggests that the physicians themselves might more commonly abuse drugs that are used and helpful in a physician’s line of work. Survey-based studies report that psychiatrists have a greater misuse of benzodiazepines; 26.3% report using unsupervised benzodiazepines in the past year, in comparison with 11.4% in other physician groups (17). This high rate of benzodiazepine misuse is reflected in the overrepre-sentation of psychiatrists in treatment.


Eighteen percent of anesthesia training programs report cases of propofol abuse (50). The prevalence of propofol abuse in such programs has increased fivefold in the past decade (50). Wischmeyer et al. identified 25 anesthesia personnel with propofol abuse; seven died as a direct result. This study described a positive correlation between hospitals with easy availability and subsequent propofol misuse. High availability was defined as little or no control over drug access within the training hospital. Although common among anesthesiologists in training, this propofol abuse pattern also occurs in physicians (and nurse anesthetists) in practice. In contrast, propofol use in nonmedical personnel is extremely rare; only one such case has been reported (51).

Propofol use has recently gained the national attention after the death of pop star Michael Jackson in 2009. Increasing reports of propofol abuse (52) and research about its addicting qualities have resulted in the Drug Enforcement Administration (DEA) placing fospropofol (53) under Schedule IV and a proposed Schedule IV for propofol as well (54). As of this writing, propofol remains unscheduled (55).

Illicit Drugs

The most common street drug of abuse among medical residents is marijuana (18). Physicians from all specialties abuse marijuana, with emergency medicine, anesthesiol-ogy, family practice, and psychiatry physicians displaying elevated odds of marijuana abuse over physicians as a whole (17). Cocaine use is more common in emergency medicine physicians, presumably from street sources. Several authors (48,56,57) have postulated that the personality style of these specialties attracts them to these drugs of abuse.

Other Drugs

Physicians are also found to abuse drugs that are not generally available or not recognized as having an abuse potential by the general public. Skipper (58) reported that tramadol was the third most frequent opioid mentioned by physicians (behind hydrocodone and meperidine) “although it was rarely the primary drug of choice” in a study of 595 physicians from two state PHPs over an 8-year period. Moore and Bostwick (59) described two cases of ketamine abuse in anesthesiologists; some professional treatment programs see several ketamine-dependent physicians per year.


The risk for addiction in physicians is an area rich in speculation and poor in research.


The strongest predictor of alcohol or drug problems in physicians is the same as in the general population: a family history of alcoholism or drug dependence (5). Of importance in this regard is the work of Moore (43) who observed several genetic and substance use factors in medical students that later correlated with alcohol abuse including non- Jewish ancestry (relative odds [RO] = 3.1), cigarette use of one pack or more per day (RO = 2.6), and regular use of alcohol (RO = 3.6).


All physician specialties are burdened with common stereotypes, and it has long been tempting to speculate about causal personality factors in the development of physician addiction. At the outset, it must be noted that in the general population, decades of research have failed to support an “addictive personality.” Observed physician personality dynamics may be a consequence or an epiphenomenon to the true etiology of the addictive process. With the preceding caveats, it is still interesting to review published speculations about physician personality types and addiction. Although personality issues may or may not be causative in addiction, they often play an important role in the progression, presentation, and treatment of addiction disorders and therefore are covered below.

McAuliffe et al. (60) noted “sensation seeking” as a personality factor that is correlated with recreational drug use among physicians in training. These authors speculate that such individuals gravitate to specialties such as emergency medicine. Emergency medicine physicians may self-select high-risk or illicit drugs owing to the same personality characteristics that draw them to their specialty. Emergency medicine physicians, as reported by Hughes et al. (17), were twice as likely to use marijuana as other specialties. Their data also suggested cocaine use was higher in this cohort. However, this hypothesis is not supported by data from other specialties also thought to attract sensation-seeking individuals, such as surgery, which is not overrepresented in treatment settings.

Bissell and Jones (35) suggest perfectionist behavior and a high-class ranking are risk factors for addiction. This is supported by the work of Higgins Roche (61), who noted that addicted anesthesiologists are often in the top 10% to 20% of their class. Udel (62) notes that compulsive personality disorder (or traits) is the most common personality diagnosis of physicians presenting for treatment. No data differentiating the occurrence of compulsive traits in addicted or nonaddicted physicians are available. Many consider compulsive traits beneficial in physician training and work. Zeldow (48) and Yufit et al. (56) speculate that the introverted and introspective qualities as well as a drive for an internal locus of control are partially responsible for the drug of choice selection in this population.

Drug Access

O’Connor and Spickard (1) described a subset of physicians who began abusing benzodiazepines and opioids only after receiving prescribing privileges. Drug access may also account for changing drugs of abuse within the opioid class over time. Green et al. (63) in 1976 and Talbott et al. (32) in 1987 reported the predominant opioid abused by physicians at the time was meperidine (Demerol). A more recent (2004) review of the Michigan and Alabama Physicians Health Programs reports hydrocodone as the number 1 opioid abused (40% of all opioid cases), meperidine dropping to 10% of cases (58). The most likely hypothesis for shifts in the drug of choice by physicians over time is the changing prescribing patterns and availability of these drugs in the marketplace.

Biologic Effect of the Drug of Choice

The neurobiologic effects of drugs used by those with an addiction color the characteristics of the addiction disorder itself. Drug-of-choice characteristics also skew the characteristics of the physician-patients arriving in treatment programs. For example, all opioids produce intense tolerance, resulting in histories of ever-increasing doses. Drug hunger drives the progressively tolerant physician to divert increasing quantities of opioids from work and, in doing so, increases the probability of detection. This partially explains why treatment-seeking or treatment-mandated physicians tend to present disproportionately with opioid abuse histories.

Major anesthetic opioids (such as fentanyl) when consumed parenterally produce a rapid downhill course owing to the development of remarkable levels of tolerance. The accelerated course of addiction from the most potent opi-oids can be postulated as contributing to deaths and the high percentage of anesthesiologists seen in physician treatment programs. Collins (3) has suggested that rapid onset (and the resolution of tolerance with brief periods of abstinence) and/or low therapeutic ratio may account for the high mortality rate in propofol-, fentanyl-, sufent-anil-, alfentanil-, and remifentanil-abusing anesthesiolo-gists. Increased awareness along with checks and balances to account for the remaindered volumes of fentanyl used in hospitals may detect diversion more rapidly and save lives of anesthesia personnel (16).


Thought and Mood Disorders

Physicians suffer from a spectrum of emotional and psychiatric problems similar to the general population. However, addicted physicians rarely have comorbid primary schizophrenia and related thought disorders. Although it is unclear whether physicians have higher or lower rates of unipolar depression, physicians who successfully complete suicide are more likely to have a drug abuse problem in their lives, self-prescribed psychoactive substances, a recent alcohol-related problem, a history of emotional problems prior to 18 years of age, and/or a family history of alcohol abuse and/ or mental illness (64). Substance dependence, self-criticism, and dependent personality characteristics are associated with depression in physicians (65). Bipolar disorder (types I and II) may contribute to the intensity of addictive disease in physicians, particularly for drinking during manic intervals (66).


PHPs are working with an increasing number of physicians with chronic pain and analgesic opioid use, many of whom have become physiologically dependent. In turn, an unknown percentage of those go on to become addicted. Eventual addiction is thought to be more common in patients with pain disorders (67) and, when combined with the 25% of physicians who self-prescribe (15), a perfect storm of high-risk factors emerges. Physicians who have significant pain and addiction disorders pose diagnostic, treatment, and management difficulties for assessors, treatment providers, and the PHPs. Regulatory issues cloud the treatment of addicted physicians with pain: Should a formerly addicted physician on opioid drugs be allowed to practice? Is it logical for state boards to prohibit methadone or buprenorphine maintenance for addiction treatment but permit potent opioids for pain management? These complex questions often result in ideologic or political decisions rather than evidence-based answers. Scientific data on the safety of physicians practicing on opioids, whether addicted or not, are sorely lacking. Insufficient data are available for a definitive decision, but appropriate concern remains (68).

Posttraumatic Stress Disorder

Posttraumatic stress disorder (PTSD) and alcoholism are closely intertwined (69), and PTSD increases the probability of addiction relapse in PTSD-related contexts (70). However, no studies about the prevalence of PTSD in physicians have been published. Physicians, like anyone else, are not immune from prior trauma histories. Several physician specialties, including emergency medicine physicians, trauma surgeons, and military psychiatrists, treat the immediate and chronic consequences of trauma. Although combat exposure is known to increase the statistical risks of addiction in veterans, no data exist to indicate whether such trauma increases the likelihood of substance abuse disorders in military physicians. Treating trauma can be, in itself, traumatizing to the caregiver.


The natural history of addiction is, on the surface, similar in physicians to that of any other drug- or alcohol-dependent person. McAuliffe (57) reports that 27% of medical students and 22% of physicians had family histories of alcohol dependence. Lutsky et al. (16) and Domino et al. (71) put this figure at almost 75%. Moreover, the genetic research literature now supports inherited genetic vulnerabilities for all major classes of addictive drugs.

Clark et al. (21) reported that excessive alcohol consumption in medical students was positively associated with better grades in the first year and a strong tendency toward better scores on Part One of the National Board of Medical Examiners test. Alcohol abuse was found to have no discernible impact on clinical rotations in years 3 and 4 of medical school in this study. This led Clark to speculate that hard-drinking students may be prone to discount warnings and feel invulnerable to the effects of alcohol; their own internal experience does not match cautionary information provided to them during their medical education. This may exacerbate an emerging “us” (doctors) and “them” (patients) view of the world. These findings mirror extensive research by Schuckit, who consistently demonstrated that less intense, early-life responses to alcohol increase the risk for the later development of alcoholism (72,73).

Stress is often cited by the physician-patient as the primary agent that drives self-medication. Stress is an elusive concept; its exact correlation with substance use and addiction is unclear. Physicians report similar levels of stress as other health professionals (74). Physicians in treatment for chemical dependency report that the stress of medical training, when combined with social isolation, provides a fertile soil for the growth of drug consumption (3). Jex (75) suggests that the physician’s unhealthy response to stress is a more important determinant of addiction than the ubiquitous presence of stress itself.

No evidence supports a specific professional personality type as being determinant in addiction. Personality dynamics specific to physicians naturally must play a role in the course of the illness and its treatment (76,77). Vaillant et al. (78) have suggested that physicians commonly experience an emotionally barren childhood. Johnson and Connelly (79), who identified 72% of a 50-physician sample hospitalized for addiction as experiencing parental deprivation in their childhood, echo this postulate. Khantzian (80) eloquently depicts the physician’s efforts at caring for others as a partially successful transformation of the conflict about being cared for themselves and an attempt to correct the barren nature of their parental nurturance. Tillet (76) described this dynamic in helping professionals as a drive to “compulsively give to others what he (she) would like to have for himself (herself).” When this transformation fails, the addiction-prone physician, lacking other methods of self-care, has a propensity to turn to substance use.

Physicians in the act of saving human lives develop a varying degree of omnipotence (4). This omnipotence, when combined with knowledge of the drugs they prescribe, may produce feelings of invulnerability regarding drug or alcohol use. Vaillant (24) has speculated that self-prescribing (related to physician self-sufficiency and false omnipotence) plays a permissive role in the development of addiction in physicians. Physicians’ illusion of mastery over pharmaceuticals keeps them from distinguishing their lack of control over chemical use, opening the door to experimentation and, if continued, a progressive deterioration in their drug use.

Genetic vulnerability and the priming effects of the drug itself remain the most evidence-based etiologies of addiction. Childhood experiences, medical school training about pharmaceuticals, and the life-and-death nature of a physician’s work certainly modify the quality and progression of a nascent addiction problem. Physicians are taught in medical school and residency (and often in their childhood) to appear self-sufficient and in control. This façade of competence establishes the framework for a secretive and duplicitous personality and, once abusing drugs or alcohol, his or her secret garden provides a fertile soil for additional chemical use. Concealment and lying are not qualities that support a mature approach to marriage, life, and work. The illicit and secretive qualities of addiction promulgate additional personality regression.

The physician’s behavior deteriorates first at home, then with friends, and finally surfaces at the workplace. By the time a physician exhibits problems at work, significant familial discord (marital strife, divorce, difficulties with acting out in children) commonly exists. Rarely does the family “turn in” an addicted spouse or other family member (81). Often, hospital staff or a colleague becomes publi-cally worried first. The physician is then confronted at work when an undeniable incident occurs or a series of smaller incidents push colleagues and the hospital medical staff to confront the doctor. An active PHP, especially one that is supportive and confidential, can be very beneficial in reducing the threshold for reporting to punitive agencies and, thus, can promote early detection. Most physicians arrive in treatment with thin scraps of their façade remaining. They exhibit a demeanor of superiority and knowledge, deny any loss of control, and have a need to appear competent, in stark contrast to their crumbling lives.



Physicians present with a broad spectrum of symptom severity, from a physician self-identifying his alcoholism while in couples’ therapy, all the way to an apneic and asystolic physician on the floor of the operating room bathroom. In the past, denial, shame, and fear of reprisal tended to keep the physician from seeking proper help until significant external consequences coalesced (2). In more recent years, the emergence of clinically oriented, supportive, and confidential PHPs has stimulated earlier reporting, by either self- or colleague referral. Physician-patients have often had years of familial and social discord while struggling to maintain acceptable work performance, until this last refuge, too, collapses. Thus, disturbances of social or familial functioning may be more sensitive indicators of early substance dependence in the physician. Unfortunately, the family often protects the alcohol- or drug-dependent “bread winner” physician.

A variety of work-related behaviors can be clues to substance use. O’Connor and Spickard (1) describe conditions and warning signs that can help detect addiction (Table 41-2). Talbott and Wright (81) and Talbott and Benson (128) have independently reported a similar list of behavioral signs of addiction in the physician.



Adapted from O’Connor PG and Spickard A. Physician impairment by substance abuse. Med Clin North Am 1997;81(4):1037–1052.

If problems are not addressed early, the doctor’s work quality and attendance often suffer. In contrast, if a physician obtains drugs at work (e.g., samples from a drug closet or drugs diverted from the OR or ICU), he or she displays the opposite behavior—volunteering for additional shifts, arriving early for work, and signing up for more complex (i.e., easier drug access) cases.

Modes of Intervention

Several comprehensive guides to physician intervention have been published (2,82). In recent years, PHPs have become very skilled at directing the physician-patient into treatment without overly aggressive confrontation and ultimatums. Tension involved in the intervention process can be reduced by directing the physician suspected of addictive disease to undergo an evaluation rather than insisting that addiction exists and treatment is indicated. The physician in question is told about existing concerns (often without divulging the source of information) and the importance of resolving said concerns by undergoing a thorough and authoritative evaluation. Ultimately, the goal of intervention is early detection of whatever problem is causing concerns. The immediate goal is to get the physician in question into a “safe harbor,” to undergo appropriate evaluation.

If handled with tact, as is common with experienced PHPs, physicians can usually be “gently coerced” into an evaluation, given the alternative of possible medical board referral and possible legal action. However, some physicians, especially those who have in the past felt assaulted by a legal process or have undergone previous interventions, require additional external pressure to begin the evaluation and/or treatment process. Regardless of the level of encouragement needed to get physician-patients into evaluation, they often arrive with a thinly fabricated story depicting their entry into evaluation or treatment as self-motivated.

Most states have reporting laws (“snitch laws”) that require hospitals and colleagues to report to the state PHP or their state medical board a physician who is suspected of being impaired by alcohol or drugs. Treating physicians must have knowledge of the laws in their state before embarking down the road of caring for physician-patients. In 2001, the Joint Commission pressured hospital organizations to address the wellness of their medical staff through standard MS2.6 (83). This Joint Commission standard has helped formalize a physician health process in most hospitals and formalize the support and intervention network in hospitals. In most states, the PHP is willing to take on or assist the hospital in meeting this standard (84,85). Hospital wellness committees can be effective in early identification and referral of addicted physicians if the process maintains a balance of compassion with a firm directive hand. Wellness committees are distinct in their focus and agenda from a hospital credentialing or executive committee whose primary agenda is managing a hospital’s greater risk management strategy. Therefore, a firewall should be maintained between the wellness and credentialing/executive committees.

If substance dependence is not caught in its early stages, the possibility of impairment arises. Thus, the primary public health goal of PHPs is to diagnose and treat physicians early in the course of their illness. In a study of impairment of all types (not focused solely on substance-induced impairment), Igartua (86) reported that 7% of residents in her survey reported working with an impaired physician supervisor. Impaired supervisory physicians are no longer protected and enabled by their juniors. Reuben and Noble (87) reported that 72% of house officers would report an impaired attending physician.


Physicians vary on their need for assessment. Some are quickly identified and agree to cooperate with their treatment needs or at least with an outpatient evaluation. Physicians who are more entrenched in their addiction, who have more complex presentations, or who are frankly resistant need formal and more extensive assessment and a methodical, nonshaming confrontation of their denial complex. In these cases, timely and proper diagnosis is best made by a multidisciplinary evaluation using the guidelines established by the Federation of State PHPs (88). Assessment can be completed at the least intensive level of care that results in a comprehensive view of the patient and his or her family and social system. The examination process must prevent the assessed physician from hiding continued drug use and withdrawal as well as addiction-related interpersonal behaviors. Because of the complexity and comprehensive nature of these evaluations, many evaluators conduct them in a residential or partial hospitalization setting where the physician remains under continuous observation. A comprehensive evaluation is best performed by removing the doctor from his or her work role to a center with expertise and willingness to take on the sometimes laborious and difficult task of physician evaluation. Allowing physicians to self-select an evaluator commonly results in their choosing a friend or colleague or someone who lacks the necessary expertise in the nuances of a physician addiction evaluation. This results in an inadequate or limited evaluation and thus a missed chance at early diagnosis. Therefore, most PHPs have established criteria and maintain a list of competent evaluators.

The evaluation should include information from, but should not be carried out by, a current or past therapist, psychiatrist, or other caregiver (9). Many PHPs direct the evaluation to a multidisciplinary team composed of an addiction medicine physician and an addiction psychiatrist and include psychological and neuropsychological testing, family assessment, review of previous medical records, and the collection of collateral information from coworkers, hospital employees, friends, and PHPs themselves. A broad array of information from all available resources is critical to an accurate assessment. Table 41-3 outlines the purpose of each component of a comprehensive physician addiction evaluation.



aAll components of the evaluation contribute to determination of whether an addictive disease exists, the level of care needed, and treatment planning for eventual care, if any.

The team involved in a multidisciplinary evaluation meets repeatedly during the course of the evaluation and, once again, when all data have been collected. Final diagnoses and recommendations are best produced by discussion (often lively discussion) by the evaluation team. The patient then meets with one or all members of the evaluation team to review the diagnosis and recommendations. The patient may elect to involve a family member. The evaluation team is best served by including the PHP or other referral source in the summation session; this action decreases confusion and splitting regarding the outcome. A comprehensive, integrated report is commonly sent to both the evaluatee and other relevant parties.


Approximately a dozen programs in the United States have experience and special expertise in the treatment of addicted physicians and other health professionals; some programs have more than 30 years of experience and have treated thousands of addicted physicians. However, some states are trending toward increased law enforcement actions against addicted physicians, as opposed to treatment. California, for example, decided to “sunset” the Physician Diversion Program in 2009, and it is far from clear what kind of structures will replace it. Strong political voices are recently heard to say that addicted physicians deserve no “strikes” and that they are, in essence, disposable in a competitive medical economy.

Clinical Considerations in Treating Addicted Physician-Patients

It has been alleged that physicians “make the worst patients.” (89) Physicians often deny symptoms of any disease, seek substandard care, and put off appropriate care for serious symptoms (90). As in any other medical situation, the physician-patient who enters addiction treatment has difficulty giving up the provider role and assuming the obligations of a patient (91). In treatment settings with an admixture of physician- and non–physician-patients, the treatment program must set firm limits, prohibiting the physician from providing medical advice or care to other patients. If a patient is the only physician in a given treatment setting, that patient will likely remain or lapse into his or her physician’s role the first moment another patient asks for medical advice or for stories from their career. This is a recipe for treatment failure. By contrast, when a physician falls into self-diagnosis, it is best to use this as grist for the therapeutic mill.

Physicians will attempt to fit the treatment into what they know: schooling and testing. Physician-patients have little trouble learning the didactic parts of treatment. Physicians early in treatment may arrive at a group therapy session with pen and paper in hand, hoping to glean one piece of information that will rocket them into recovery or, at the very least, accelerate their discharge. The transformation required of all patients in addiction treatment is an emotional, interpersonal, and, for many, a spiritual shift. Physicians have little experience in this area. They often become stuck trying to obtain an “A” in treatment and, in this way, miss the necessary wholesale changes that are needed to recover in earnest. When the staff attempt to correct the physician’s approach to treatment, they risk becoming ensnared in the physician’s tendency toward excess perfectionism. The resultant hostile projection produces negative transference and a thinly veiled contempt for “less educated” therapists and staff (91).

Physicians work and interact in an environment filled with physical and emotional pain. In order to succeed, they must at times distance themselves from the strife around them. When combined with an achievement-oriented childhood, the physician-patient defaults to intellectualization of his emotional experience or, on occasion, frank alexithymia (without words for feelings) (92). Treatment will necessarily reacquaint the physician with the subtle nuances of feeling states, often confused or conflated with craving or “stress.”

One particularly difficult emotional state is shame. Most addiction patients view their substance abuse and their lives through a lens of shame—and physicians seem to have a surfeit of shame. Fayne and Silvan (91) note that a key task in recovery is an honest appraisal of how the physician’s addiction has interfered with his ability to function as a physician. This requires a vigilant therapeutic group that models self-disclosure and self-examination. The physician, owing to childhood and training-induced drives for accomplishment and perfection, risks turning the task of self-examination into self-loathing. Treatment of such individuals mandates that the treatment staff and community encourage fearless self-examination without inadvertently pulling the hair trigger of the physician’s self-loathing. When in the state of shame, an additional defense of the physician-patient is to psychologically freeze. The precarious management of shame is further complicated by the patient’s transference and the therapist’s countertransference that arises when a bright physician-patient seems incapable (or willfully resistant) to the self-examination necessary for recovery.

Working with addicted physicians requires understanding of the dynamics of addiction and the distinct but highly interactive elements between addiction and the personality. Inexperienced or overly biased treatment providers tend to label the psychological effects of addiction as personality issues or, conversely, they view long-standing personality dynamics in the physician-patient as addictive thoughts and actions. A balanced understanding and therapeutic approach requires a healthy respect for both schools of thought. Addiction uses the specific personality dynamics of the physician-patient to serve its own ends, exaggerating and driving maladaptive forces to ensure its own survival. Conversely, the addictive process generates complicated internal and interpersonal pathology.

It is tempting to establish a cause-and-effect relationship between nonaddiction psychiatric disorders and the disease of addiction itself. Such a path often colludes with the patient’s denial system. A more powerful viewpoint is to envision a patient’s addiction and their Axis I and II psychiatric problems as distinct disorders that are independent but most certainly collaborative and mutually reinforcing.

Social and legal issues only further confound the type and course of treatment. Because of all the aforementioned issues, treatment is by its nature different in physicians. Medical boards, the general public, PHPs, and the physician him- or herself have low tolerance for the potential public harm that can occur when a physician becomes addicted; they are exquisitely intolerant of multiple relapses. This flies in the face of the nature of addiction: a disease char-acterized by remission and relapse. The societal pressure to “have a perfect recovery” creates a maladaptive alliance with the physician-patient’s own perfectionism (93).

Characteristics of the Treatment Setting

The treatment of physicians involves a prolonged continuum of care. When a physician leaves his or her initial treatment setting and returns to work, this is described by the unfortunate and inaccurate vernacular of having “completed treatment.” In fact, what physicians are asked to do in the second phase of treatment is in many ways more comprehensive care than what many patients receive during their primarytreatment (11). This “posttreatment” monitoring commonly involves weekly group therapy sessions, peer support groups, aftercare groups, individual and family therapy, self-help group attendance, drug testing, and work-site monitor reports for 5 years or more.

The confluence of known difficulties engaging physicians in treatment, the public demand for safety, and liability issues involved in allowing a physician to work while in outpatient addiction treatment have promoted physician-specific, long-term residential addiction treatment programs. A paucity of literature exists about the efficacy of less intensive treatment, but fair results have been reported by Dilts (94) and Reading (95). Smith and Smith (96) reported a small cohort of physicians treated in low- and high- intensity care, with substantively better results when longer-term residential care was employed. DuPont et al. (97), reviewing 16 state PHPs over 5 years, noted that 78% of physicians who required treatment went to residential treatment for 30 to 90 days, followed by less intensive outpatient treatment. The remaining 22% of treated physicians went directly to outpatient treatment. Hospitals, malpractice carriers, regulatory boards, health insurance companies, and family and friends have expectations of continuous abstinence. Most medical boards and, increasingly, malpractice insurance companies (who in many states have become a more powerful threat) penalize a physician if he or she relapses, even a single time. Owing to the research (albeit limited) on the effectiveness of residential treatment and the penalty placed upon relapsing physicians, most physician-patients are encouraged to attend longer treatment programs than nonphysician brethren (98).

Skipper (98) outlined the treatment of the impaired health professional. He reported that all physician-special-ized treatment programs use a 12-step philosophy as the core component of treatment. Such programs have proven effectiveness with physicians (11,97,99101). Studies show that if abstinence is the desired outcome point, consistent involvement with 12-step meetings produces the best results (102). All physician treatment programs reviewed by Skipper (98) utilize family therapy, and most offer a brief psychoeducational family program sometime in the physician’s treatment (1). Family participation also leads to a better outcome (103). Family members move through their own denial of the significance of the problem, anger at the physician-patient, and fear of loss of prestige and financial security. The initial goal of family treatment is to redirect the hostility away from the patient (as well as the treatment providers and PHP) toward the addictive illness itself, using this energy to build healthy and constructive family dynamics, focused on relapse prevention.

Physician-specific groups allow self-disclosure and sharing of alcohol- and drug-related behaviors that risked or, in some cases, caused patient harm. Such violations of the Hippocratic Oath generate shame. Once articulated, such lapses in physician responsibility are best linked to the addictive disease and away from the core self. Disclosures of the deepest violations of core values in profession specific groups can, if properly managed, provide relief and help the physician differentiate his or her actions while addicted from their self-concept. Physician-specific groups serve a different, more pragmatic, but equally important, purpose. Most physicians have work-related triggers (e.g., drug access at work, prescription pads, and locations in the office or hospital where use occurred). In these groups, participants explore work triggers and develop medically specific relapse prevention plans. On this practical level, physician-specific groups also address the myriad other issues physicians face when returning to practice, such as the difficulties of seeing their patients in AA, how to respond to questions from peers and other staff about their illness, Drug Enforcement Agency prescribing restrictions, and continued management of drugs and prescriptions in the office or hospital.

There is an increasing body of evidence for the safe and effective use of pharmacologic adjuncts in the treatment of substance dependence. Most programs that treat alcohol-dependent physicians utilize one or more medications including disulfiram, oral or injectable naltrexone, acamprosate, and/or topiramate for alcohol dependence. Medications are also useful in the treatment of opioid dependence among physicians. The opioid antagonist nal-trexone is prescribed for physicians who, upon return to practice, have continued easy access to opioids. It could be argued that monthly injectable naltrexone is especially desirable because the monitoring program is assured that the drug is continuously “on board.” Alternatives such as monitoring urine for the presence of naltrexone or observed administration of oral doses of naltrexone may also be used; however, observation quickly lapses replacing a safeguard with false security. Physician treatment programs and PHPs are currently conflicted about the use of buprenorphine or methadone in opioid-dependent physicians. This is covered below in the section: Controversies.

Ultimately, long-term randomized monitoring of physicians may be the most essential component of treatment and critical for sustained recovery. Monitoring and support groups are commonly provided by PHPs or occasionally by the treatment center itself, as discussed below.

Physician Health Programs


The importance of PHPs in supporting and promoting early detection and proper evaluation and treatment of physicians cannot be overstated. The heart of the physician’s health movement can be traced back to the founding of the International Doctors in Alcoholics Anonymous (IDAA) by Clarence Pearson, in 1949 (104). IDAA has grown from 24 physicians, meeting in Pearson’s garage in Cape Vincent, New York, to an international organization attracting thousands of physicians and other doctorate-level individuals in recovery from addiction. On the regulatory side, the Federation of State Medical Boards called for a model probation and rehabilitation process for addicted physicians in 1958. However, no meaningful change occurred until 1973 with the publication of the watershed JAMA article: “The sick physician. Impairment by psychiatric disorders, including alcoholism and drug dependence” (105). The American Medical Association (AMA) held its first conference on physician impairment in 1975. State medical societies orga-nized committees on physician impairment. The American and Canadian Medical Associations have jointly sponsored conferences on physician impairment every other year since 1975. Concern from medical organizations, governing bodies, and hospital regulatory boards resulted in the state-by-state emergence of PHPs over a period of 25 years. By 2007, almost every state in the United States has some type of PHP, ranging from one employee with a $20,000 budget to a 1.5 million dollar budget and 19 full-time employees (97). In 2007, PHP programs monitored more than 9,000 physicians across the United States (97).


PHPs have widely different organizational structures and lines of authority. More than half (54%) of PHPs are nonprofit foundations. Others are part of their respective state medical association (35%) or the licensing board itself (13%) (97). All PHPs have written agreements that guide their interaction with their state licensing boards. Most (59%) of PHPs evaluated in the DuPont et al. study from 2009 (97) have specific laws that sanction their actions and guide their operation. PHPs have evolved from two distinct sources. Some PHPs have descended from committees of medical board itself and have evolved, with varying degrees of autonomy, from that board. Other PHPs emerged from a state medical society or other concerned physician groups. The independent evolution of state PHPs coalesced into a federation in 1990. Many state medical boards continue to actively monitor some physicians while referring others to the state PHP. Interestingly enough, one comparison study of a state (Oregon) with both programs noted that “voluntary diversion program for appropriately selected physicians may enhance earlier referral and intervention” (37,106).

PHP Activities

Education and Referral

Most PHPs provide education about all types of physician impairment (including substance use disorders) and train local hospitals and physician organizations on techniques to help identify and report suspected impairment. Even more importantly, these educational programs offered by PHPs afford the PHP staff the chance to personally meet and network with medical leadership throughout their state. This public relations and training effort carried out by PHPs is important because it helps individuals understand and trust the supportive goal of the PHP, which in turn promotes early referral. Health care organizations have shown increased interest in these issues, thanks to the recent Joint Commission standard (currently MS 4.80), which mandates that “the medical staff implements a process to identify and manage matters of individual health for licensed independent practitioners. This identification process is separate from actions taken for disciplinary purposes” (83).

Addiction continues to be the most commonly identified problem addressed by many PHPs (28), but most PHPs address other psychiatric disorders, behaviorally disruptive physicians, and physicians who suffer from other compulsive disorders such as gambling and sexual misbehaviors. All PHPs offer consultation about a potential impairment, coordinate intake into treatment, and monitor physicians after treatment through statewide systems. Some PHPs offer initial assessment, triage, and ongoing therapy groups for the physicians in their state.

PHPs have become more professional, with credibility provided by their expertise, affiliation with the Federation of State Physician Health Programs, and other medical organizations, such as the AMA and the Federation of State Medical Boards. As professionalism has increased, so has their finesse and ability to carry out educational programs, expanding to a broader range of topics (stress and burnout, sexual misconduct, appropriate prescribing, etc.). The core concept of PHPs has become clear: “to detect problems that lead to impairment and to intervene and encourage physicians to obtain assistance prior to damaging their careers or harming patients” (Skipper GE. Personal communication, 2008.). Sophistication in dealing with addicted physicians has increased, in partnership with expert evaluators and treatment providers. Follow-up monitoring has become much more sophisticated with additional monitoring tools (hair testing, flexible variations in drug testing, new tests for alcohol, devices for monitoring alcohol, and so on). New software options are facilitating the aggregation and analysis of physician monitoring records, obtaining reports through online reporting and real-time oversight. Participant satisfaction with the PHP process, irrespective of whether they entered voluntarily through mandate, is quite satisfactory (107,108).

Abstinence Monitoring

All PHPs track the abstinence status of their recovering physicians. All programs use random witnessed body fluid analysis (most frequently through urine drug screens but often including hair and blood analysis) through an orga-nized monitoring program. Screens commonly taper in frequency over the course of monitoring, for a period of 5 or more years (47). Participation in PHP monitoring is contingent upon the physician “calling in” or checking a confidential Web site each day to see whether he or she has to provide a sample for assay. Urine screening in physician populations requires considerable expertise and accuracy, since addicted physicians can use their knowledge to evade detection (109). Most physician drug panels test for 20 to 25 drugs, including a wide variety of opioids (97). Specialty screens for fentanyl, alfentanil, and sufentanil are necessary in recovering physicians who have abused these drugs in the past and/or who have future access to such compounds. Hair testing can be important in this regard because fentanyl and its congeners have very brief half-lives but are readily detected in hair for weeks or months. Physicians also occasionally abuse more unusual drugs (ketamine, propofol, tramadol, and dextromethorphan); these physicians need assessment panels specifically designed to prevent a lapse in their abstinence to such substances. The screening is also broad as to the drug types. This breadth prevents switching from one substance to another, as commonly occurs during the natural course of the addiction disease.

More recently, PHPs began more sensitive testing for alcohol use by assaying for ethyl glucuronide (EtG) (110,111) and ethyl sulfate (EtS) (112), liver and lung tissue metabolites of ethyl alcohol. Newer testing for blood phosphatidylethanol (PEth) is more sensitive to even lower amounts of ethanol consumption (113). PEth testing has become more convenient with fingerstick blood sampling. False-positive test results for EtG, EtS, and PEth have been reported, owing to a combination of environmental exposure and the sensitivity of the tests (EtG, EtS, PEth) and the low-level production of EtG in by urine bacteria (EtG). The two most common culprits in false positives are incidental ingestion of ethanol-containing substances (e.g., mouthwash) and topical application of ethanol-based hand sanitizers (especially if inhaled). Physicians under monitoring are counseled to avoid these compounds. Alternatives to ethyl alcohol–based hand sanitizers (e.g., isopropyl alcohol hand sanitizers) are one important example.

Recovery Support

In addition to urine monitoring, most state PHPs provide some type of group experiences and behavioral monitoring (e.g., attendance records at support groups and therapy). The most common of these are caduceus groups, a vague moniker that varies from peer-led groups similar to 12-step meetings to large therapist-led groups that alternate between discussing a member’s issue or concern and process work in a large group setting. Unlike Alcoholics Anonymous meetings, direct feedback and discussion is encouraged in most caduceus groups. Newcomers may obtain recovery sponsors or guidance from physicians that are more senior in the network of PHP support groups.

All long-term sobriety studies of physicians underscore the importance of 12-step meetings (primarily AA and NA) as a central part of recovery (99,114). In a study of 100 physicians with an average of 33.4 months after treatment admission, Galanter et al. (99) noted, “A.A. was apparently perceived by respondents as the most potent element of their recovery.” Outcome studies in physicians show impressive abstinence rates, with studies extending to 21 years (114).


Significant consequences to the physician and the public can result from relapse. PHPs have developed models of assessing relapse severity. DuPont et al. (97) describe three categories of relapse derived from the earlier work of one of the authors (Skipper):

■  Level I relapse consisted of missing therapy meetings, support groups, dishonesty, or other behavioral infraction.

■  Level II relapse involved the reuse of drugs or alcohol but outside the context of medical practice.

■  Level III relapse involved drug or alcohol use within the context of practice.

This relapse system highlights the most frequent downhill slide for professionals in a monitoring system: Problems with behavioral adherence commonly precede substance use. It helps a PHP stage its interventions when a participant slides off the radar. The downside of this classification system is that a level I “relapse” obfuscates a commonly accepted term (relapse) and might be better described as a compliance failure. Thus, this relapse classification should be seen as unique to monitoring programs.

Hankes (reported in Domino, 71) has developed a more extensive relapse management decision tree for the Washington State PHP that classifies relapse and provides decision support for managing seven distinct categories of relapse. It is common for physicians in the first year after treatment to have a brief relapse or slip. If the slip is short-lived, the physician is often best placed in short-term relapse prevention programming, directed to reintegrate with 12-step meetings, and begin self-exploration in psychotherapy. Slips (and the resultant treatment), if managed quickly with appropriate psychotherapy, can deepen the physicians’ acceptance of their disease and solidify subsequent recovery. If managed properly, singular slips are most often helpful in the long run and are not indicators of failed treatment (116). Should a physician have a more extended relapse, he or she should engage in the following:

■  An evaluation of the physician’s ability to practice until he or she is more stable in recovery

■  A longer and tighter monitoring contract that includes behavioral monitoring, support group attendance, and more extensive toxicology testing

■  A reexamination of the patient’s psychiatric status, to determine whether an occult mood disorder, other addictive process, or past unaddressed trauma is present

■  A reassessment of the patient’s family dynamics and support system

■  An evaluation of the physician’s safety to practice

■  A determination of the need to repeat primary residential treatment (or to treat other elements of the addiction or other psychiatric disorder)

■  A reevaluation of the need for relapse prevention medications

Relapse is part of the disease of addiction. Physicians who have difficulty maintaining abstinence should be removed from the workforce until treatment providers with experience in physician recovery determine that the physician is safe to return. The point in time when a physician is safe to practice is best established by a joint decision of the physician’s treatment provider and the monitoring PHP. All stakeholders must be prudent and err on the side of caution when considering readiness to return to work in safety-sensitive occupations.

Reture to Work

Most PHPs insist on an initial removal from the workplace during the first phases of treatment and after any sustained relapse. The point in time when a physician is safe to prac-tice is best established by a joint decision of the physician’s treatment provider and the monitoring PHP. All stakeholders must be prudent about when to return physicians to their safety-sensitive occupations. Parameters to consider when returning a physician to his safety-sensitive occupation are reviewed in Table 41-4.



PHPs and treatment providers have a wide variety of thoughts on how to structure the physician’s work and home life once a return to work date has been determined. Issues to be considered include workplace conditions, physician’s initial workload and whether shift change should be allowed, his or her safety to practice around addicting substances, whether solo or group practice should be considered, any restrictions on prescribing DEA scheduled drugs, and the need for remedial training. In an effort to increase consensus on this topic, an instrument called the Medical Professional Addiction Recovery Inventory has been developed to balance recovery status and the workplace environment (117).

Treatment Outcome Data

Physicians have been the subject of multiple outcome studies focused on the efficacy of extended, multimodal addiction treatment and monitoring. Most addiction treatment outcomes studies are plagued by high percentages of subjects being lost to follow-up. However, owing to the tight monitoring of PHPs, physician-based studies have excellent follow-up rates, approximating 90% in some studies (100). Physicians appear to have responded very well to their unique treatment and monitoring process. More sophisticated outcome analyses (40,47,71,100,101) attempt to define why physician treatment is so successful. The natural progression of this line of thought is to identify which components of the physician treatment process can be general-ized to the public at large (116).

Gallegos et al. (118) reported a 77% sustained abstinence rate in physicians followed for 5 years. In the North Carolina PHP, Ganley et al. (119) noted 65% of physicians had a good outcome (as defined by completing an aftercare contract), and another 26% had a good outcome with complications (e.g., relapsed but eventually completed a monitoring contract) in a 6-year study from 1995 to 2000, resulting in a 91% good outcome. In 2002, Lloyd (114) reported an impressive follow-up of alcohol-dependent physicians in the United Kingdom over 21 years, noting a mean sustained duration of abstinence of 17.6 years abstinence in 68 of 80 physicians reporting. He conservatively scored the 20% lost to followup as negative outcomes, and even with this, he noted that 73% of the physicians in his study of 80 physicians were in recovery. Knight et al. (40), reporting on the Massachusetts Medical society’s 10-year data between 1993 and 2003, cited 75% program completion (albeit some participants utilized multiple contract periods). Successful completion required continuous and complete abstinence for a contract period in this study. Eight percent relapsed, and 17% did not complete a contract for other reasons. They noted that the time to relapse was shorter for women. Medical board involvement in the oversight of the physician was associated with a positive outcome in this study.

Domino et al. (71) noted that 25% of physicians in the Washington State PHP (1991 to 2001) had at least one relapse. Family history, comorbid psychiatric disorder, and a previous relapse increased the probability of relapse. The use of major opioids increased the probability of relapse but only in the presence of a comorbid psychiatric disorder. McLellan et al. (47) evaluated the outcomes among 904 addicted physicians treated in 16 PHPs and found 78% were continuously abstinent throughout the 5- to 7-year period of evaluation; more than 90% of those physicians were still practicing medicine. Among those physicians who did relapse, 74% had only one episode of alcohol or drug use.


Conflicts between Privacy and Public Safety

Physician treatment with its mandated abstinence monitoring illustrates the conflict between the physician-patient’s need for privacy and the public’s need for safety. Added to this is a stigmatized view of addiction; the result is that the addicted physician has become the “whipping boy” of physician impairment. Many other problems among physicians can and do lead to mistakes and patient harm (e.g., sleep deprivation, overwork, poor communication with hospital staff, intemperate affairs), but they are not as directly addressed and do not receive a fraction of the public or regulatory board outcry or concern. Ironically, confidentiality for treatment of physician mental illness, including substance use disorders, actually increases patient safety by encouraging early referral and safe passage into treatment (37,106). During the first several years of implementing a state PHP, the new program commonly sees a flood of early participants who are identified by colleagues or family due to the privacy afforded by the PHP.

Conversely, many states have laws that mandate caregivers to report suspected physician impairment (a term that is not synonymous with addiction but is often confused as such. More accurately stated, impairment is a consequence of addiction if it is left untreated). Some states mandate that treatment providers report physicians to the medical board, regardless of whether impairment has been proven. In many cases, a default board action ensues. Although this may appear on superficial examination to protect patients, an excessively broad mandate for reporting actually decreases the probability that a physician will seek or accept a referral for assessment and treatment. If the perceived consequences of referral are sufficiently prejudicial, referral is delayed and ultimately only occurs when a major incident signals the transition from illness to impairment. In states with PHPs, regulatory boards allow PHP intercession, holding off disciplinary proceedings as long as the physician effectively addresses his disease in an appropriate, structured, and accountable manner. As soon as regulatory boards tilt toward law enforcement and away from treatment, physicians who become dependent, their colleagues, and care providers become reluctant to report. The addicted physician and his or her family delay or avoid treatment. An uninformed provider may hide behind the confidentiality of their profession and lose the benefit of the organized monitoring and peer support provided by a PHP.

The structure of PHPs, on the other hand, facilitates a proper balance between the privacy that is critical for treatment and the public’s need for safety. They hold the awkward middle ground between their medical board and treatment providers. PHPs provide confidentiality if the physician’s illness does not pose a threat to public safety but report to the medical board should a patient become uncooperative or a risk to the public at large. The promise of protected and effective treatment encourages all parties to refer to the PHP before the substance-abusing physician deteriorates to the point of a potential safety risk.

Is Monitored Recovery the Same as Self-Guided Recovery?

Pysicians frequently enter treatment to retain their medical license(s). One treatment goal is to shift the physician from this external driver to an internalized state of recovery as a lifelong journey. During treatment and subsequent monitoring, a finite number of physicians do not make this shift. Once the initial ravages of addiction remit, such individuals are held in a drug-free state by the oversight of drug screens and behavioral monitoring. In such cases, the internalization of recovery (an ongoing process of changing behaviors, attitudes, and beliefs) slows or stops; the transition into the self-motivated journey of recovery does not replace the holding cell provided by monitoring. The term disease stasis syndrome has been applied to this small subset of physicians. Such physicians have a high probability of returning to alcohol and other drug use, when and if monitoring is discontinued. In the disease stasis syndrome, the individual has made a commitment to abstinence only as a temporary means to an end. Such physicians are compliant and assume a false persona of acceptance to their treatment providers, monitors, and PHP personnel.

This false recovery is a by-product of external pressure and the intense treatment and monitoring that physicians undergo. Treatment providers should avoid pressuring patients to conform because physicians are, after all, good students who know how to give “correct” answers. Instead, providers should encourage patients to verbalize their resistance and dissatisfaction with treatment and to praise honest self-disclosure, especially if the patient is describing how he or she is stuck in the process of change. Physician-patients should be encouraged to disclose remnants of the central fallacy of the addicted mind: the fantasy that they may return to drinking or using drugs in a controlled and sociable manner once they are “strong enough” or have “learned enough about myself.” Open discussion regarding recovery ambivalence should be a recurrent theme in group therapy with this population. Psychodynamic psychotherapy may help such individuals integrate how past survival techniques of false compliance to authority figures are at play in their relationship with the current authority figures in their therapy, treatment centers, and PHPs (130). In the meantime, monitoring holds the physician behaviorally accountable and, if properly framed as appropriate supportive care, is not only justifiable but also a good medicine. In their treatment, individuals with the disease stasis syndrome should remain on random screens, until this syndrome improves. Some cases may need to remain on screens for an indefinite time.

Can Physicians Return to Practice on Opioid Agonist Treatment?

All addiction treatment programs in the United States that specialize in physicians consider complete abstinence from addicting drugs to be the end goal of treatment (97). Despite aggressive treatment, a small percentage of physicians are unable to maintain sobriety. In such cases, should physicians be treated using opioid agonist treatment with methadone or buprenorphine? Moreover, should such a physician receiving opioid agonist treatment be allowed to practice? Should the physician-patient be permitted to choose opioid agonist therapy as a treatment option?

In the general population, poor outcomes have been reported for opioid-dependent patients who are not receiving opioid agonist medication (120). Physicians, on the other hand, have very high sustained success rates using a treatment protocol without opioid agonist medications coupled with the additional assistance of well-designed continuing care and licensure leverage (40,47,71,100,101,118). With such high success rates, many PHPs feel the added problems of maintenance medications are unwarranted. Proponents of opioid agonist treatment argue that opioid agonist treatment is more humane, or at the very least, when other treatment has failed, opioid agonist maintenance therapy is not only justified but also safe.

A large study of 904 physicians from 16 PHPs (47) found only one addicted physician was reported to be receiving opioid agonist treatment. More recently, Skipper (Skipper GE. Personal communication, 2008) surveyed PHPs and reported that 14 of the 36 PHP respondents indicated they were following up at least one physician receiving opioid agonist treatment. When contacted, each state indicated that the number of cases they had were few and usually involved complex pain issues.

Medical-legal issues seem to appear when addicted physicians are receiving opioid medications. Gray (Gray R. Personal communication, 2007) states, “At least one major statewide malpractice carrier has indicated that they will not insure an addicted physician if he is on opioid maintenance therapy, due to the difficulties in defending such a physician in a malpractice case.” Opioid agonist treatment does occur in the treatment of physicians, but there is no clear consensus on which cases need opioid agonist treatment. Research assessing physicians who work while receiving opioid ago-nist treatment does not exist. The widespread availability of buprenorphine therapy for the general public has increased the intensity of this controversy.

At the heart of the issue is the degree, if any, of cognitive impairment that occurs with opioid agonist treatment. Hamza and Bryson (68) reviewed the current (and limited) research as to the cognitive effects of opioid agonist treatments, including buprenorphine, and conclude that cognitive changes do occur. Opioids are indeed potent drugs with primary targets in the central nervous system; it should come as no surprise that they alter brain functioning (129).

This issue becomes more complex when one considers cases of opioid-addicted physicians who suffer from chronic, nonmalignant pain. The opioids may be necessary to maintain the quality of life in such an individual. However, that same individual may have a history of inappropriate opioid use or even diversion of opioids from patients. In this case, the PHP and treatment providers are balancing the physician’s need for pain control with the safety of the public and the fear of reprisal by an uninformed public. The decision about a physician’s ability to practice in such situations should be approached with caution and a complete knowledge of the research and clinical knowledge in this area. Although the use of chronic opioids may be necessary in such cases, loss of control from prescribed doses does occur. The resolution of this conundrum should rest upon the effect the medication has on the brain and behavior of physicians who take such medications, not upon the disease for which they are prescribed. It must be stated, however, that the public and regulatory agencies are more tolerant of opioids for analgesia than they are of opioid maintenance.

Different PHPs approach this controversy from different angles. A few approve the use of opioid agonist treatment in practicing physicians as a treatment option. In such states, the PHP assists care providers in deciding who is a proper candidate. Other states strongly oppose the use of opioids because of the success with abstinence-based treatment in this cohort. Still, other states see opioid agonist treatment as a last resort and follow such cases carefully and/or limit that physicians scope of practice to mitigate (the real or perceived) danger. No clear consensus will emerge until additional research validates the approach of PHPs in these difficult cases.

Are Opioid-dependent Anesthesiologists Safe to Return to Their Profession?

Multiple conflicting publications debate the advisability of opioid-dependent anesthesiologists returning to the operating room. Menk et al. (121) reported a successful reentry rate of only 34% for parenteral opioid–using anesthesiol-ogists versus 70% for nonopiate abusers. This oft-quoted 1990 study promulgated a pessimistic view of anesthesiolo-gists returning to work but has been criticized because it was essentially a retrospective survey of anesthesia training directors, subject to recall bias. Of the 159 anesthesia training programs surveyed, 113 responded, providing 180 case reports, with most programs providing only a single case report of a resident having been addicted. Critics contend that if most programs reported only a single case, it is likely such reports were skewed toward disasters. Collins et al. (122) also surveyed anesthesiology residencies in 2001, noting that 50% of treated anesthesiologists remained in anesthesiology after treatment, with 91% completing training and 9% dying of relapse-related incidences.

Paris and Canavan (123) compared 32 anesthesiologists with 36 physician controls for an average of 7.5 years; they showed no difference in the relapse rates between these two groups. When stratified by residents versus attending physicians, no significant difference was found. Domino et al. (71) examined the risk of relapse over 11 years and 256 participants in a Washington State PHP, including 32 anesthesiologists. The relapse rate for anesthesiologists was not statistically significantly different from other physicians. Additionally, there was not a single episode of patient harm or death from overdose by any anesthesiologist in this study. A similar report from Pelton (38) involving 255 physicians who had participated in the California Diversion Program over 10 years showed no difference in relapse rates for anesthesiologists.

Domino et al. (71), evaluating physicians in the Washington State PHP, noted that physicians who had abused fentanyl had a slightly lower incidence of relapse than those who had abused other major opioids. Anesthesiologists who returned to the practice of anesthesiology did have an increased risk of relapse when compared to those who did not return, although they caution that their numbers are small and the significance uncertain. Major opioid users had a higher risk of relapse as did physicians with an existing comorbid psychiatric disorder or a family history of addiction. They conclude that anesthesiologists who use major opioids and do not have other risk factors (family history, comorbid psychiatric disorder, and history of relapse) are good candidates to return to the practice of anesthesiology. A more recent study by Skipper (101) reviewed data from PHPs in 16 states, culling information about anesthesia providers. They noted that anesthesiologists had excellent outcomes similar to other physicians, with no higher mortality, relapse rate, or disciplinary rate and no evidence in their records of patient harm. These authors postulated that the type of treatment and monitoring that these physicians received from the 16 state PHPs account for the differences from earlier reports.

Several studies point to the importance of opioid antagonists in the long-term management of the addicted anes-thesiologist. Merlo et al. (124) described impressive data (but preliminary, due to the number of cases) comparing anesthesiologists treated with naltrexone to those without the protection of this opioid antagonist. They conclude that mandatory naltrexone is an appropriate treatment option.

Oreskovich also noted the wide disparity in outcomes when one compares anesthesiologists who are aggressively treated and have well-designed, long-term monitoring with those who do not have such care (125). Studies that followed anesthesiologists under close monitoring in PHPs or regulatory boards (Donovan (71) in Washington State (71); Paris and Canavan (123), New Jersey; Pelton (38), California; Skipper (101) and 16 different states with active PHPs) describe outcomes for anesthesiologists that are similar to other physicians, whereas studies that are based upon a survey of the memories of anesthesiology program directors (where patients had uncertain treatment and monitoring) describe poor, and at times life terminating, outcomes. The number of studies where PHPs were central in the management of addiction in anesthesiologists underscores their critical importance in maintaining recovery for their participants and safety for the public.

No study has correlated relapse rate with the type or length of treatment and the intensity of PHP supervision and monitoring. Additional studies that include the use of opioid antagonists are needed. The controversy on this matter has, however, driven some residencies to retrain an anesthesia resident in an alternative field once addiction to major opioids occurs. This decision trades the loss of years of training for the potential loss of life. The decision for a long-careered anesthesiologist proves more difficult.

What Happens When a Physician Relapses?

Periods of recovery alternating with relapse characterize the disease of addiction. In contrast, the expectation by medical boards and the public is that physicians should never relapse, placing another burden of perfectionism upon a cohort of already perfectionist and harshly self-judging individuals. For some physicians, the experience of recovery feels more like a jail of perfectionism instead of a journey where one learns to accept imperfections. The consequence of relapse for any person with addiction entails a loss of self-efficacy. For the physician, it may involve a loss of livelihood and facing possible board or legal sanctions.

Physicians early in their recovery often experience a brief “discovery” relapse (a return to drug use where the individual’s relapse experience validates and internalizes a heretofore poorly accepted diagnosis) (71,115). PHPs are familiar with such occurrences; medical boards and the public at large are not. Research into, and standardization of, interventions in the event of an early recovery relapse should improve outcomes and at the same time increase public trust.

Repeated relapses that run the risk of public harm should be managed by removing such an individual from his or her practice. Multiply relapsing physicians may need a sustained period of remission prior to a return to practice. Involvement in support systems and length of remission predict the best prognosis moving forward (126).


Physicians were the first professional group to address addiction within their profession; this leadership continues today. The disease of addiction in physicians follows a similar course as in the public at large, with several notable exceptions. The access to potent drugs is one of the most important of these exceptions. The identification, evaluation, and treatment of addiction in this population in many ways set the gold standard for the care of addiction, with many possible applications for the treatment of addiction in the public at large (116). Despite the exemplary care and self-monitoring by the physician community, physicians are punished for being ill.

The treatment of physicians is different (especially in the United States), partly driven by public outcry for complete and sustained remission in a disease that is chronic and relapsing by nature. PHPs are integral and imperative elements in the comprehensive disease management of physician addiction; they increase the long-term prognosis for physicians who suffer from addictive disease. Controversies in the management of addiction in physicians abound and call for further research in this interesting and complex population.


1.O’Connor PG, Spickard A. Physician impairment by substance abuse. Med Clin North Am 1997;81(4):1037–1052.

2.Centrella M. Physician addiction and impairment—current thinking: a review. J Addict Dis 1994;13(1):91–105.

3.Collins GB. Drug and alcohol use and addiction among physicians. In: Miller, NS., ed. Comprehensive handbook of drug and alcohol addiction. New York: Marcel Dekker, 1991:947–966.

4.Bissell L, Hankes L. Health professionals. In: Lowinson J, Ruiz P, Millman RB, eds. Substance abuse: a comprehensive textbook, 3rd ed. Baltimore, MD: Lippincott Williams & Wilkins, 1997:897–908.

5.Flaherty JA, Richman JA. Substance use and addiction among medical students, residents and physicians. Psychiatr Clin North Am 1993;16(1):189–197.

6.Berge KH, Seppala MD, Schipper AM. Chemical dependency and the physician. Mayo Clin Proc 2009;84(7):625–631.

7.Bissell L, Haberman PW. Alcoholism in the professions. New York: Oxford Press, 1984.

8.Angres DH, Talbott GD, Bettinardi-Angres K. Healing the healer: the addicted physician. Madison, CT: Psychosocial Press, 1998.

9.Nace EP. Achievement and addiction: a guide to the treatment of professionals. New York: Brunner/Mazel, 1995.

10.Coombs RH. Drug impaired professionals. Cambridge, MA: Harvard University Press, 1997.

11.DuPont RL, McLellan AT White WL, Merlo LJ, et al. Setting the standard for recovery: Physicians’ Health Programs. J Subst Abuse Treat 2009;36(2):159–171.

12.McAuliffe WE, Rohman M, Breer P, et al. Alcohol use and abuse in random samples of physicians and medical students. Am J Public Health 1991;81(2):177–181.

13.Kessler RC, Berglund, P, Demler O, et al. Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the national comorbidity survey replication. Arch Gen Psychiatry2005;62:593–602.

14.Niven RG, Hurt RD, Morse RM, et al. Alcoholism in physicians. Mayo Clin Proc 1984;59:12–16.

15.McAuliffe WE, Rohman M, Santangelo S, et al. Psychoactive drug use among practicing physicians and medical students. N Engl J Med 1986;315:805–810.

16.Lutsky I, Hopwood M, Abram SE, et al. Use of psychoactive substances in three medical specialties: anesthesia, medicine, and surgery. Can J Anaesth 1994;41(7):561–567.

17.Hughes PH, Storr CL, Brandenburg NA, et al. Physician substance use by medical specialty. J Addict Dis 1999;18(2):23–37.

18.Hughes PH, Baldwin DC, Sheehan, DV, et al. Resident physician substance use, by specialty. Am J Psychiatry 1992;149(10): 1348–1354.

19.Hughes PH, Brandenburg N, Baldwin, DC, et al. Prevalence of substance abuse among U.S. physicians. JAMA 1992;267:2333–2339.

20.Hughes PH, Conard SE, Baldwin DC, et al. Resident physician substance use in the United States. JAMA 1991;265(16):2069–2073.

21.Clark DC, Daugherty SR, Baldwin DC, et al. Assessment of drug involvement: applications to a sample of physicians in training. Addiction 1992;87(12):1649–1662.

22.Vaillant GE, Brighton JR, McArthur C. Physicians’ use of mood altering drugs: a 20 year follow-up report. N Engl J Med 1970;282:365–370.

23.Mangus RS. Hawkons CE, Miller MJ. Tobacco and alcohol use among 1996 medical school graduates. JAMA 1998;280(13):1192–1193.

24.Vaillant GE. Physician, cherish thyself: the hazards of self prescribing. JAMA 1992;267:2373–2374.

25.Baldwin DC, Hughes PH, Conard SE, et al. Substance use among senior medical students: a survey of 23 medical schools. JAMA 1991;265:2074–2078.

26.Clark DC, Eckenfels EJ, Daughterty SR, et al. Alcohol-use patterns through medical school: a longitudinal study of one class. JAMA 1987;257:2921–2926.

27.Morrison J, Wickersham P. Physicians disciplined by a state medical board. JAMA 1998;279(23):1889–1893.

28.Clay SW, Conatser RR. Characteristics of physicians disciplined by the State Medical Board of Ohio. J Am Osteopath Assoc 2003;103(2): 81–88.

29.Federation of State Physician Health Programs – History. Accessed December 10, 2012.

30.About CPPPH – History. December 10, 2012.

31.Berry AJ and the Task Force on Chemical Dependence. Model curriculum on drug abuse and addiction for residents in anesthesiology. The American Society of Anesthesiologists,, 1998.

32.Talbott G, Gallegos K, Wilson P, et al. The Medical Association of Georgia’s Impaired Physicians’ Program: Review of the First 1,000 physicians—analysis of specialty. JAMA1987;257:2927–2930.

33.McGovern MP, Angres DH, Uziel-Miller ND, et al. Female physicians and substance abuse, comparisons with male physicians presenting for assessment. J Subst Abuse Treat1998;15(6):525–533.

34.Smart D. Physician characteristics and distribution in the U.S. Chicago: American Medical Association Press, 2006.

35.Bissell L, Jones RW. The alcoholic physician: a survey. Am J Psychiatry 1976;133:1142–1146.

36.Wunsch MJ, Knisely JS, Cropsey KL. Women physicians and addiction. J Addict Dis 2007;26(2):35–43.

37.Shore J. The Oregon experience with impaired physicians on probation—an eight year follow-up. JAMA 1987;257:2931–2934.

38.Pelton C, Ikeda RM. The California Physicians Diversion Program’s experience with recovering anesthesiologists. J Psychoactive Drugs 1991;23(4):427–431.

39.Myers T, Weiss E. Substance use by interns and residents: an analysis of personal, social, and professional differences. Br J Addict 1987;82: 1091–1099.

40.Knight JR, Sanchez LT, Sherritt L, et al. Outcomes of a monitoring program for physicians with mental and behavioral health problems. J Psychiatr Pract 2007;13(1):25–32.

41.Gold MS, Melker RJ, Dennis DM, et al. Fentanyl abuse and dependence further evidence for second hand exposure hypothesis. J Addict Dis 2006;25(1):15–21.

42.McAuliffe P, Gold MS, Bajpai M. Second-hand exposure to aerosolized intravenous anesthetics propofol and fentanyl may cause sensitization and subsequent opiate addiction among anesthesiologists and surgeons. Med Hypotheses 2006;66(5):874–882.

43.Moore RD. Youthful precursors of alcohol abuse in physicians. Am J Med 1990;88:332–336.

44.True WR, Xian H, Scherrer JF, et al. Common genetic vulnerability for nicotine and alcohol dependence in men. Arch Gen Psychiatry 1999;56:655–661.

45.Nelson D, Giovino GA, Emont SL, et al. Trends in cigarette smoking among us physicians and nurses. JAMA 1994;271(16):1273–1275.

46.Stuyt EB, Gundersen D, Shore J, et al. Tobacco use by physicians in a physician health program, implications for treatment and monitoring. Am J Addict 2009;18(2):103–108.

47.McLellan AT, Skipper GS, Campbell M, et al. Five year outcomes in a cohort study of physicians treated for substance use disorders in the United States. Br Med J 2008;337:1154–1156.

48.Zeldow P, Daughtery S. Personality profiles and specialty choices of students from two medical school classes. Acad Med 1991;66(5):283–287.

49.Hyde GL, Wolf J: Alcohol and drug use by surgery residents. J Am Coll Surg 1995;181:1–5.

50.Wischmeyer E, Johnson BR, Wilson JE, et al. A survey of propofol abuse in academic anesthesia programs. Anesth Analg 2007;105:1066–1071.

51.Fritz GA, Niemczyk WE. Propofol dependency in a lay person. Anesthesiology 2002;96(2):505–506.

52.Bryson EO, Frost EA. Propofol abuse. Int Anesthesiol Clin 2011;49:173–180.

53.Drug Enforcement Administration Department of Justice. Schedule of controlled substances; placement of fospropofol into schedule IV. Final rule. Fed Regist 2009;74(192):51234–51236.

54.Drug Enforcement Administration Department of Justice. Placement of Propofol Into Schedule IV. Proposed rule. Fed Regist 2010;75:66196–66199.

55.U.S. Department of Justice, Drug Enforcement Administration, Department of Diversion Control. List of Controlled Substances, September 2012. html Accessed December 27, 2012.

56.Yufit R, Pollock G, Wasserman E. Medical specialty choice and personality. Arch Gen Psychiatry 1969;20:89–99.

57.McAuliffe W. Risk factors in drug impairment in random samples of physicians and medical students. Int J Addict 1987;22(9):825–841.

58.Skipper GS. Tramadol abuse and dependence among physicians [letter]. JAMA 2004;292(15):1818–1819.

59.Moore NN, Bostwick JM. Ketamine dependence in anesthesia providers. Psychosomatics 1999;40:356–359.

60.McAuliffe W, Rohman M, Wechsler H. Alcohol, substance use, and other risk-factors of impairment in a sample of physicians-in-training. Adv Alcohol Subst Abuse 1984;4(2):67–87.

61.Higgins Roche BT. Substance abuse policies for anesthesia. Winston-Salem, NC: All Anasthesia, 2007.

62.Udel MM. Chemical abuse/dependence: physicians’ occupational hazard. J Med Assoc Ga 1984;73:775–778.

63.Green RC, Carroll GJ, Buxton WD. Drug addiction among physicians. The Virginia experience. JAMA 1976;236:1372–1375.

64.American Medical Association Council on Scientific Affairs. Results and implications of the American Medical Association-American Psychiatric Association Physician Mortality Project. JAMA1987;257:2949–2953.

65.Brewin CR, Firth-Cozens J. Dependency and self-criticism as predictors of depression in young physicians. J Occup Health Psychol 1997;2(3):242–246.

66.Angres DA, McGovern MP, Shaw MF. Psychiatric comorbidity and physicians with substance use disorders: a comparison between the 1980s and 1990s. J Addict Dis 2003;22(3):79–87.

67.Savage, SR. Addiction and pain: assessment and treatment issues. Clin J Pain 2002;18(Suppl 4):S28–S38.

68.Hamza H, Bryson EO. Buprenorphine maintenance therapy in opioid-addicted health care professionals returning to clinical practice: a hidden controversy. Mayo Clinic 2012;87(3):260–267.

69.Volpicelli J, Balaraman G, Hahn J, et al. The role of uncontrollable trauma in the development of PTSD and addiction. Alcohol Res Health 1999;23(4):256–262.

70.Norman SB, Tate SR, Anderson KG, et al. Do trauma and PTSD symptoms influence addiction relapse context? Drug Alcohol Depend 2007;90:89–96.

71.Domino, KB, Hornbein TF, Polissar NL, et al. Risk factors for relapse in health care professionals with substance use disorders. JAMA 2005;293(12):1453–1460.

72.Schuckit MA, Gold EO. A simultaneous evaluation of multiple markers of ethanol/placebo challenges in sons of alcoholics and controls. Arch Gen Psychiatry 1988;45(3):211–216.

73.Schuckit MA. A longitudinal study of children of alcoholics. Recent developments in alcoholism : an official publication of the American Medical Society on Alcoholism, the Research Society on Alcoholism, and the National Council on Alcoholism. 1991;9:5–19.

74.Nace EP. Achievement and addiction: a guide to the treatment of professionals. New York: Brunner/Mazel, 1995:66–69.

75.Jex SM. Relations among stressors, strainers, and substance use in physicians. Int J Addict 1992;27(8):979–994.

76.Tillett R. The patient within — psychopathology in the helping professions. Adv Psychiatr Treat 2003;9:272–279.

77.Merlo LJ, Gold, MS. Prescription opioid abuse and dependence among physicians: hypotheses and treatment. Harv Rev Psychiatry 2008;16(3)181–194.

78.Vaillant GE, Sobowale NC, McArthur C. Some psychological vulnerabilities of physicians. N Engl J Med 1972;287:372–372.

79.Johnson RP, Connelly JC. Addicted physicians: a closer look. JAMA 1981;245(3):253–257.

80.Khantzian EJ. The injured self, addiction, and our call to medicine: understanding and managing addicted physicians. JAMA 1985;254(2): 249–252.

81.Talbott GD, Wright C. Chemical dependency in health care professionals. Occup Med 1987;2(3):581–591.

82.Fleming MF. Physician impairment: options for intervention. Am Fam Physician 1994;50(1):41–44.

83.Joint Commission on the Accreditation of Healthcare Organizations. Comprehensive accreditation manual for hospitals. 2006. Comprehensive accreditation manual for hospitals (CAMH): The Official Handbook. Chicago: Joint Commission, 2008

84.Physician Health Services of the Massachusetts Medical Society corporation. JCAHO Requirement—MS.4.80 LIP HEALTH

85.Model Wellness Committee Bylaw published by the Tennessee Medical Foundation. Accessed December 18, 2012.

86.Igartua KJ. The impact of impaired supervisors on residents. Acad Psychiatry 2000;24(4):199–194.

87.Reuben DB, Noble S. House officer responses to impaired physicians. JAMA 1990;263(7):958–960.

88.Federation of State Physician Health Programs, Inc. Evaluation, Appendix I, in The 2005 physician health program guidelines. 22–26.

89.Schneck SA. “Doctoring” doctors and their families. JAMA 1998;280(23):2039–2042.

90.Stoudemire A, Rhoads JM. When the doctor needs a doctor: special considerations for the physician-patient. Ann Intern Med 1983;98(1): 654–659.

91.Fayne M, Silvan M. Treatment issues in the group psychotherapy of addicted physicians. Psychiatr Q 1999;70(2):123–135.Alcohol Alcohol 2002;37(4):370–374.

92.Sifneos P, Apfel-Savitz R, Frankl F. The phenomenon of “alexithymia.” Psychother Psychosom 1977;28:47–57.

93.Gabbard GO. The role of compulsiveness in the normal physician. JAMA 1985;254(20):2926–2929.

94.Dilts S. The Colorado Physician Health Program: observations at 7 years. Am J Addict 1994;3(4):337–345.

95.Reading EG. Nine years experience with chemically dependent physicians: the New Jersey Experience. Md Med J 1992;41(4):325–329.

96.Smith PC, Smith JD. Treatment outcomes of impaired physicians in Oklahoma. J Okla State Med Assoc 1991;84(12):599–603.

97.DuPont RL, McLellan AT, Carr G, et al. How are addicted physicians treated? A national survey of Physician Health Programs. J Subst Abuse Treat 2009:37(1):1–7.

98.Skipper GE. Treating the chemically dependent health professional. J Addict Dis 1997;16(3):67–73.

99.Galanter M, Talbott GD, Gallegos K, et al. Combined Alcoholics Anonymous and professional care for addicted physicians. Am J Psychiatry 1990;147(1):64–68.

100.Buhl A, Oreskovich M, Meredith C, et al. Prognosis for the recovery of surgeons from chemical dependency: a 5-year outcome study. Arch Surg 2011;146(11):1286–1291.

101.Skipper G, Campbell MD, DuPont RL. Anesthesiologists with Substance Use Disorders: a 5-year outcome study from 16 state physician health programs. Anesth Analg 2009;109:891–896.

102.Florentine R, Hillhouse MP. Drug treatment and 12-step program participation. The effects of integrated recovery activities. J Subst Abuse Treat 2000;18:64–74.

103.Enders LE, Mercire JM. Treating chemical dependency: the need for including the family. Int J Addict 1993;28:507–519.

104.International Doctors in Alcoholics Anonymous. IDAA History and Objectives. Accessed December 10, 2012.

105.The sick physician. Impairment by psychiatric disorders, including alcoholism and drug dependence. JAMA 1973;223:684–687.

106.Nelson HD, Matthews AM Girard DE, et al. Substance-impaired physicians probationary and voluntary treatment programs compared. West J Med 1996;165:31–36.

107.Fletcher CE, Ronis DL. Satisfaction of impaired health care professionals with mandatory treatment and monitoring. J Addict Dis 2005;24(3):61–75.

108.Merlo LJ, Greene WM. Physician views regarding substance use-related participation in a state physician health program. Am J Addict 2010;19:529–533.

109.Jaffee WB, Trucco E, Levy S, et al. Is this urine really negative? A systematic review of tampering methods in urine drug screening and testing. J Subst Abuse Treat 2007;33(1):33–42.

110.Skipper GE, Weinmann W, Thierauf A, et al. Ethyl glucuronide: a biomarker to identify alcohol use by health professionals recovering from substance use disorders. Alcohol Alcohol2004;39(5):445–449.

111.Wurst FM, Skipper GE, Weinmann W. Ethyl glucuronide—the direct ethanol metabolite on the threshold from science to routine use. Addiction 2003;98(Suppl 2):51–61.

112.Wurst F, Dresen S, Allen J, et al. Ethyl sulphate: a direct ethanol metabolite reflecting recent alcohol consumption. Addiction 2005;101:204–211.

113.Aradottir S, Asanovska G, Gjerss S, et al. Phosphatidylethanol (PEth) concentrations in blood are correlated to reported alcohol intake in alcohol-dependent patients. Alcohol Alcohol2006;41(4):431–437.

114.Lloyd G. One hundred alcoholic doctors: a 21-year follow-up.

115.Earley PH, Gallegos K, Howell E, et al. Georgia Composite State Board of Medical Examiners: guidelines for problem physicians. Georgia Composite Board of Medical Examiners, 1995.

116.DuPont RL, Skipper GE. Six lessons from state physician health programs to promote longtTerm recovery. J Psychoactive Drugs 2012;44(1):72–78.

117.Earley P. MPARI History, Part One. 2009: Accessed June 10, 2012.

118.Gallegos KV, Lubin BH, Bowers C, et al. Relapse and recovery: five to ten year follow-up study of chemically dependent physicians—the Georgia experience. Md Med J1992;41(4):315–319.

119.Ganley OH, Pendergast WJ, Wilkerson MW, et al. Outcome study of substance impaired physicians and physician assistants under contract with the North Carolina physicians health program for the period 1995–2000. J Addict Dis 2005;24(1):1–12.

120.Weiss RD, Potter JS, Fiellin DA, et al. Adjunctive counseling during brief and extended buprenorphine-naloxone treatment for prescription opioid dependence: a 2-phase randomized controlled trial. Arch Gen Psychiatry 2011;68:1238–1246.

121.Menk EJ, Baumgarten RK, Kinsley CP, et al. Success of reentry into anesthesiology training programs by residents with a history of substance abuse. JAMA 1990;13;263(22):3060–3062.

122.Collins GB, McAllister MS, Jensen M, et al. Chemical dependency treatment outcomes of residents in anesthesiology: results of a survey. Anesth Analg 2005;101:1457–1462.

123.Paris RT, Canavan DI. Physician substance abuse impairment: anesthesiologists versus other specialties. J Addict Dis 1999;18(1):1–7.

124.Merlo LJ, Greene W, Pomm RM. Mandatory naltrexone treatment prevents relapse among opiate-dependant anesthesiologists returning to practice. J Addict Med 2011;5(4):279–283.

125.Oreskovich M, Caldeiro R. Anesthesiologists recovering from chemical dependency: can they safely return to the operating room? Mayo Clin Proc 2009;84(7):576–580.

126.Vaillant GE. What can long-term follow-up teach us about relapse and prevention of relapse in addiction? Br J Addict 1988;83(10):1147–1157.

127.Benzer DG. Healing the Healer: a primer on physician impairment. Wis Med J 1991;90(2):70–79.

128.Talbot GD, Benson EB. Impaired physicians—the dilemma of identification. Postgrad Med 1980;68(6):56–64.

129.Younger JW CL, D’Arcy NT, Trott KE, et al. Prescription opioid analgesics rapidly change the human brain. Pain 2011;152(8):1803–1810.

130.Malan, David. Individual psychotherapy and the science of psychodynamics. Oxford University Press, New York, 2001.