Substance Use Disorders in Health Care Professionals


The impact of licit (i.e., alcohol and nicotine used legally) and illicit (including nonmedical prescription) drug use, abuse, and dependence in the United States is well documented in the general population. Overall, a 2006 survey reported that an estimated 20.4 million Americans 12 years of age or older were current illicit drug users, meaning they had used an illicit drug—defined as “marijuana/hashish, cocaine (all forms), heroin, methamphetamine, hallucinogens, inhalants or psychotherapeutics used nonmedically”—during the month prior to the survey interview. This estimate represents 8.3% of the population 12 years of age or older. More specifically, an estimated 5.2 million persons were current nonmedical users of prescription pain relievers, up from an estimated 4.7 million in 2005.

A recent report of abuse of prescription medication in the United States reported that many health care professionals are poorly trained to deal with alcohol or drug abuse. A substantial number of patients served daily by health care professionals in various health care facilities are abusing or dependent on alcohol and or other drugs. On the other hand, the public expects health care professionals to understand the proper use of the medicines they prescribe, dispense, or administer to their patients. Just as in their patients, though, alcohol or drug use also affects the lives of a number of health care professionals.

Starting in college, some health care students develop an attitude of invulnerability and immunity to addiction, fueled by their advanced understanding of the mechanisms of drug action. What begins as recreational college alcohol or drug use may, for some, develop into a complicated pattern of alcohol or drug abuse or dependence intended to attain a “sense of well-being” (p. 17) without an overt manifestation of intoxication or side effects. This concept of balancing drug effects, also called “titration,” or “walking a chemical tightrope,” refers to a practice whereby students or health care professionals use their pharmacological knowledge to balance positive and negative drug actions and reactions by “enhancing, neutralizing or counteracting specific drug effects through ingesting multiple types of drugs” [Dabney D (1997) A sociological examination of illicit prescription drug use among pharmacists. University of Florida, “An Unpublished Dissertation”].

Health care professionals have a significant responsibility that comes with the privilege of using medications to treat patients. Although most health care professionals engage in appropriate prescribing, dispensing, and administration of medication, reports of exceptional cases often receive public attention. A North Hollywood, California, physician, for example, was charged with conspiring to distribute 406 prescriptions of hydrocodone and oxycodone over 2 months after he surrendered his license to the US Drug Enforcement Administration (DEA) in May 2008. This pain management specialist was also being investigated regarding a role that his prescriptions might have played in the deaths of six patients over the past 3 years. A Virginia pharmacist was caught with hundreds of phentermine capsules when he was apprehended by law enforcement authorities, and a Maryland pharmacist was trading sex for drugs. Medication errors caused by substance-impaired pharmacists have been cited as posing a direct and serious threat to the public. Moreover, nurses were reported to be alcohol or drug impaired while committing “dozens of errors leading to patient deaths in Illinois” (p. A1).

Whether by virtue of their drug access or socioeconomic status, most evidence supports the notion that a small but significant proportion of health care professionals do experience personal problems with the use of alcohol and other drugs, which can result in serious consequences to themselves and to the public [Valentine N (1991) Stress, alcohol and psychoactive drug use among nurses in Massachusetts. Brandeis University, Boston, MA, “An Unpublished Dissertation”]. Not only can the economic costs of substance use disorders in health care professionals be considerable, but early identification is essential because patient and provider well-being may be at risk. Given the increasingly stressful environment due to manpower shortages in the health care system in general, alcohol or drug use and misuse among health care professionals has been projected to grow. Treatment of alcohol or drug disorders by health care workers was a policy issue recognized years ago by the professional organizations, and the Joint Commission requires hospitals to monitor and identify matters of health including substance use and abuse by physicians and other health care professionals.

The aim of this chapter is to provide perhaps the most comprehensive review of the problem of drug abuse by health care professionals to date. In addition, although covered in greater detail in other chapters in this book, we also briefly discuss the behavioral signs and symptoms of addiction in health care professionals, the treatment of substance use disorders in this special subpopulation, and the prognosis of sustained recovery and efforts needed to enlighten the various health care professional programs and groups.

Epidemiology of Alcohol and Other Drug Use by Health Care Professionals

The current literature regarding the prevalence of substance use and dependence in health care professionals is limited in both its scope of generalizability and methodological rigor. Lack of empirical data have contributed to an air of skepticism regarding the actual prevalence of substance abuse (abuse as referred to colloquially, not a Diagnostic and Statistical Manual of Mental Disorders , Fifth Edition [DSM-5] diagnosis) and dependence by health care professionals. In fact, evidence of the extent of medication diversion, considered to be the major source of nonprescribed drug abuse by health care professionals, is based primarily on retrospective accounts [Dabney D (1997) A sociological examination of illicit prescription drug use among pharmacists. University of Florida, “An Unpublished Dissertation”], although the actual size of the diversion problem is largely unknown. As a result, the prevalence of inappropriate substance abuse and chemical dependency among health care professionals is inconclusive and, like the extent of prescription opioid drug diversion in the United States, for example, is impossible to estimate at the present time. The fact that the behaviors being measured represent illegal or inappropriate behaviors compounds the problem, as it is difficult to obtain accurate estimates of sensitive variables such as substance use.

A glimpse of the lack of epidemiological knowledge in the field is best illustrated by contradictory prevalence estimates found in the literature. For example, reports have suggested that narcotic addiction in US physicians was as much as 30–100 times the rate found in the general population, but these data were based on data from Germany in the 1950s. In addition, the lifetime estimate of combined substance abuse and dependence among health care professionals was reported by Kessler et al. to be at a rate nearly equal to that of the general population, or 26%. Similarly, estimates from other studies of health care professionals have reported a lifetime prevalence of substance dependence ranging from 3% to 20%. Although the literature provides limited studies of substance use by dentists, Hedge has estimated that up to 15%–18% of dentists could be addicted to drugs and alcohol. In contrast to these rates, however, another report concluded that physicians were at a “greater lifetime probability of developing a substance-related disorder than the general population” (p. 7).

Such statements clearly demonstrate the confusion and misinformation surrounding a meaningful discussion of alcohol or drug use by health care professionals. These generalizations have not only contributed to the uncertainty about the prevalence of substance use, but also to the confusion with regard to risk factors that contribute to substance use among health care professionals. For example, although referring to pharmacists as “drugged experts” (p. 102), Dabney used a measure of questionable reliability and validity to assess substance use in a nationwide sample of pharmacists. Specifically, the measure assessed redundant drug use items, categorized unauthorized use of nonnarcotic medications as addictive drug use, and provided no direction to participants regarding exactly which drugs were included in each drug category. Moreover, as also noted by Baldwin, the frequency data reported by Dabney contained no time frame for reported substance use and were, therefore, not useful in estimating the prevalence of substance use. Although Dabney claimed that the onset of potentially addictive drug use in pharmacists occurred upon becoming a professional, such a conclusion was essentially impossible without longitudinal data or some specific items assessing age at the onset of regular use. The methods were strongly defended by the author ; however, these issues contribute to a suspect interpretation of the data.

Overgeneralizations from methodologically questionable data also exist in the limited amount of literature describing substance use by the dental profession. Except for reporting the number of disciplinary actions taken against Oregon dentists from 1979 to 1984, no known empirical prevalence data for substance use had ever been reported for practicing dentists until recently. However, Chiodo and Tolle, drawing on nonrepresentative disciplinary action data, inaccurately concluded that dentists, like physicians, were at higher risk for substance use and abuse than the general population, and also concluded that the literature had consistently reported higher rates of chemical dependency in health care professionals, a notion unsupported by quantitative self-reported data.

In a series of important analyses, McAuliffe et al. assessed alcohol or drug use by both physicians and pharmacists, and Valentine [Valentine N (1991) Stress, alcohol and psychoactive drug use among nurses in Massachusetts. Brandeis University, Boston, MA, “An Unpublished Dissertation”] assessed alcohol or drug use by nurses. Lack of generalizability to other practitioners outside these two disciplines was a major limitation of these studies. In addition, these studies were conducted in the Northeast, where past-year alcohol or drug use has been reported to be higher than in other areas of the United States. Subsequently, to address some of the methodological shortcomings of these studies, Hughes and colleagues (p. 2333) compared a national sample of physician use of alcohol or drugs with that of the general population. They reported that when compared with the general population, physicians were more likely to use alcohol, benzodiazepines, and minor opioids but less likely to use street drugs such as marijuana and cocaine. Furthermore, contrary to the suggestion made by Chiodo and Tolle, that the literature consistently reported disproportionately higher rates of chemical dependency in health care professionals, Hughes et al. reported that only 7.9% of physicians identified themselves as substance abusers, while the corresponding rate for the general population at that time was 15%–18%. Hughes et al. also noted, however, that physicians were as likely as their age and gender peers to have experimented with illicit substances in their lifetime, an observation also affirmed more recently. Although methodologically rigorous, Hughes et al. acknowledged the narrow focus of their study to physicians alone that subsequently limited their findings due to the lack of comparable national data across other similar professions. In recognition of this limitation, the authors concluded that any comparisons between physicians and other health care professionals in “similar socioeconomic strata may have yielded different results” (p. 2337). Complicating these issues is stigma that accompanies alcohol or drug use in any population, which leads to underestimates of problem use.

Etiology of Substance Use Disorders in Health Care Professionals

Many etiologic factors have been reported to contribute to substance use in health care professionals, such as a family history for drug or alcohol use, college substance use, or age at first alcohol or drug use ; psychological factors such as “pharmacological optimism,” access to prescription medications, self-prescribing, socioeconomic status, and additional factors such as gender (male), lack of religious practices, and social influences.

Drug Access

Drug access, and in particular easy drug access, is generally recognized as a principal factor contributing to substance use by health care professionals. Certainly, studies show that access to prescription medications would explain the higher rates of use of these drugs among health professionals than in the general population. Although research on drug use in the working population in general has been inconclusive, Mensch and Kandel have suggested that drug use by workers was due less to the workplace than to the workers themselves. Clearly, however, a substantial foundation of research indicates that health care professionals are at considerable risk due to their working environment. Drug access is related directly to the job of being a health care professional. As such, the working condition related to medical practice is an important contributing factor enhancing one’s exposure to addicting drugs.

To illustrate this point, dentists have historically had easy access to nitrous oxide, an inhalant commonly kept in dental offices, and a known drug of abuse for dentists. Although the data are now dated (1979–1984), 7.1% of 109 impaired dentists in a study that took place in Oregon were sanctioned for abusing nitrous oxide. The authors concluded that nitrous oxide in particular posed a serious hazard for dentists. Although dentists have access to nitrous oxide for procedures, access to other drugs such as minor opioids and anxiolytic drugs is limited. For example, dentists were the only health care professional group who did not report personal use of samples; the study, nonetheless, indicated that they found other sources for addicting prescription medications.

Different researchers have developed measures to assess the impact of drug access by health care professionals on drug use [Dabney D (1997) A sociological examination of illicit prescription drug use among pharmacists. University of Florida, “An Unpublished Dissertation”]. A pilot study by Trinkoff and Storr suggested that easy access to drugs contributed to misuse. This was more firmly supported in a later, more extensive study of nurses ( n = 3917), wherein the ease of access correlated positively with past-year misuse. Three workplace dimensions were measured (availability, frequency of administration, and workplace controls), and, summed as an index, nurses with easy access on all dimensions were most likely to have misused prescription-type drugs (odds ratio = 4.18; 95% confidence interval [CI] 1.70–10.30). Furthermore, access continued to show the same correlation to misuse, even when knowledge of substances was also controlled in the analysis, thereby showing that access was not explained by nurses’ knowledge of substances used.

In a survey study performed comparing alcohol or drug use by pharmacy and nursing students and with pharmacists and nurses, predictors of lifetime illicit drug use by pharmacists and nurses included having a family history of drug problems, greater amount of past-month alcohol use, lack of religious affiliation, and notably greater access to drugs. Predictors for use of an illicit drug (any Schedule I or unprescribed drug use) by pharmacy and nursing students included a family history of drug problems, less drug access, and cigarette use in the past year. Of interest, lower drug access was a significant predictor for lifetime illicit substance use by pharmacy and nursing students, suggesting that when substances were unavailable in the workplace, students were more likely to obtain them elsewhere. Despite a reassurance of anonymity, students may also have been reluctant to admit to such behavior due to the fear of being discovered. In support of this notion, none of the students in the study reported diverting any medications from where they work, yet a greater number of pharmacy and nursing students in the same sample reported use of prescription medications than among the general population. We know that various sources for drug use include the home and friends [Dabney D (1997) A sociological examination of illicit prescription drug use among pharmacists. University of Florida, “An Unpublished Dissertation”], but we also know that sources include the workplace as well [Dabney D (1997) A sociological examination of illicit prescription drug use among pharmacists. University of Florida, “An Unpublished Dissertation”].

Where pharmacy students worked did not appear to be related to disproportionate drug use; however, a greater number of retail pharmacists reported illicit drug use than pharmacists in other pharmacy practice areas. When parsing out comparisons of individual drugs, except for marijuana, consistent with the data from Hughes et al. a higher proportion of the general population reported use of street drugs such as cocaine, hallucinogens, and inhalants. A greater number of health care professionals and students, however, reported use of drugs to which they typically had access, such as opioids and anxiolytics. In sum, quantitative and qualitative studies have all demonstrated that increased drug access in an unrestrictive environment provides an important substrate permissive of drug use by health care professionals. The available studies are consistent for studies of nurses, pharmacists [Dabney D (1997) A sociological examination of illicit prescription drug use among pharmacists. University of Florida, “An Unpublished Dissertation”], certain types of physicians, and health care professionals in general that report drug access to be a key element leading to drug misuse and abuse in health care professionals. Efforts to restrict drug access in every setting, as well as increased vigilance to monitor drug procurement and drug disposition by clinicians who dispense from their offices, should be considered a priority.

Family History of Alcohol and Drug Use

Without a doubt, the greatest concern for health care professionals, as for the public, are alcohol use disorders. Lifetime prevalence of alcohol abuse in the United States is 17.8% and alcohol dependence is 12.5%. Past 12-month prevalence of alcohol abuse is 4.7%, while alcohol dependence over the same period is 3.8%. Alcohol dependence is significantly more prevalent among men, whites, and younger unmarried adults, and lifetime alcohol abuse is highest among middle-aged Americans.

Twin studies of alcoholics have highlighted the possibility of genetic components of alcoholism while other researchers have also sought genetic markers for individuals with a positive family history for alcoholism. Studies have demonstrated that first-degree relatives (parents, siblings, or offspring) are more likely to use alcohol, become alcohol dependent, and are at substantially higher risk of developing problems with alcohol at some point during their lives. Family history of alcoholism has been estimated to be approximately 38% in the United States.

A retrospective review of substance use and addiction in medical students, residents, and physicians suggested that the most predictive factor for alcoholism in physicians was a positive family history for alcoholism. Kenna and Wood reported that significant bivariate correlations between positive family history and pattern of alcohol use ( r = 0.31), as well as positive family history for drug problems and current drug use ( r = 0.55), existed for physicians alone. There is the possibility that there were genuine relationships between those physicians reporting a positive family history for alcoholism and their alcohol use and between a positive family history for drug problems and drug use. Physicians are trained diagnosticians and can putatively accurately assess the presence or lack of alcohol and drug use problems by family members. These diagnoses may have led to a more accurate assessment of family members, thereby reducing measurement error in this particular group.

Numerous studies also demonstrate that first-degree relatives are at substantially higher risk of developing problems with alcohol at some point during their lives. Coombs proposed that the health care professions attract “people vulnerable to drug abuse because of emotional impairment due to alcoholic and emotionally abusive parents” (p. 192). Several studies of dental students previously speculated that many dentists perhaps come from dysfunctional families or families with a history of alcoholism or chemical dependency. Sammon et al., for example, reported that 35%–39% of students at two dental schools had an alcoholic parent or grandparent, and Sandoval et al. reported that 15% of all dental students at the University of Texas had a family history of alcoholism and 17% of illicit drug use. In a more recent study, however, dentists reported the fewest family members with alcohol problems of any health care professional group, suggesting that there is little evidence that dentists are at greater risk than other health care professionals to report a family history of alcohol problems.

Several other studies have also reported high rates of positive family history for alcoholism for health care students and health care professionals as well. For example, in a comparison of chemically dependent and nondependent nurses, Sullivan reported that 62% of chemically dependent nurses reported an alcoholic family member, compared with 28% for nonchemically dependent nurses. In addition, in a sample of recovering pharmacists, Bissell et al. reported a positive family history for alcoholism rate of 55%–58% in recovering pharmacists, slightly higher than the 47.4% prevalence estimate reported by Kenna and Wood in a survey. What of course must be considered between the two rates are the differences between the two study populations: one clinical and the other population based. In college students, Tucker et al. reported a positive family history for alcoholism in 28.1% in a sample of pharmacy students, and Kriegler et al. established that a positive family history for alcoholism was reported by 38.3% of nursing student respondents. In a measure including eight close relatives (other studies typically included parents, grandparents, and siblings), Kenna and Wood reported a positive family history for alcoholism in 46% of pharmacy students and 74.5% of nursing students surveyed.

In a follow-up study of 479 licensed health care professionals (68.7% response), researchers sought to ascertain whether positive family history for alcoholism and positive family history for drug problems were more prevalent among nurses than among dentists, pharmacists, and physicians and if an association between positive family history for alcoholism or positive family history for drug problems and current alcohol or drug use, respectively, existed. Nurses reported a significantly higher prevalence of positive family history for alcoholism than other groups of health care professionals ( P < 0.001) ( Fig. 68.1 ), and nurses also reported a significantly higher prevalence of positive family history for drug problems than dentists and physicians ( P < 0.01), but not pharmacists ( Fig. 68.2 ). The study also demonstrated that positive family history for alcoholism in nursing was not associated with either amount of current alcohol use or abstinence. On the other hand, as noted previously, among physicians alone, relationships between alcohol use and positive family history for alcoholism as well as between drug use and positive family history for drug problems were significant. The results of this study support the notion that positive family history for alcoholism and positive family history for drug problems differ across groups of health care professionals.

Fig. 68.1

Family history of alcoholism among health care professionals ( n = 479).

Fig. 68.2

Family history of drug abuse among health care professionals ( n = 479).

While speculated, no one truly understands why a significant number of people with a positive family history for alcoholism appear to select nursing as a profession. Some have suggested that the desire to go into nursing emanates from the family of origin and that nurses assume parental roles taken on in childhood. For example, in a study of the characteristics of chemically dependent nurses, 48% indicated that while growing up, they had acted in some type of parental role compared with only 22% of nondependent nurses. In order to delineate the association between nursing and family history of alcoholism, more research into the familial dynamics or individual differences of nurses and nursing students needs to be performed.

Professional Invincibility

Many health care professionals assume that their education, intelligence and knowledge of pharmacology will provide immunity from substance-related impairment. This self-deception of professional invincibility is an attitude of denial of impairment. More importantly, intervention is difficult in health care professionals as denial to the existence of a substance-related problem contributes to continued substance use, abuse, and dependence. Hankes and Bissell referred to this air of invincibility in physicians as “MDeity” (p. 890). The attitude that health care professionals are selectively immune to the pharmacological actions of addictive medications—based primarily on their knowledge of drug action—has been the subject of retrospective accounts given by many health care professionals [Dabney D (1997) A sociological examination of illicit prescription drug use among pharmacists. University of Florida, “An Unpublished Dissertation”]. The health care professionals may believe that their education, particularly with respect to drug titration, makes them impervious to physical or psychological dependence, or the unconsidered equivalent, drug addict. Health care professionals are good at hiding their addiction by walking a pharmacological tightrope ; they tend to take greater amounts and a wider variety of drugs, making them more difficult to treat. Perhaps, then, it is this attitude of pharmacological invincibility that becomes the fundamental problem with substance use experimentation and addiction in health care professionals, particularly those who choose to treat themselves or who continue treatment beyond the period of illness or at dosages escalating beyond those required to circumvent tolerance.

Pharmacological Optimism

Pharmacological optimism, or the anticipated benefits of medication use, has also been suggested to be a contributing factor in the development of chemical impairment by health care professionals. In 2000, pharmaceutical companies spent 15.7 billion dollars on drug promotion that grew to almost 30 billion dollars by 2005, with most of the expenditure aimed toward health care professionals. Although pharmaceutical companies relentlessly directly target consumers, the impact does not compare to the decades-long indoctrination that health care professionals have received. Moreover, in view of a health care professional’s training and education, familiarity with drugs is an important aspect of professional competency. Development of beliefs about the anticipated outcomes using drugs can be assumed to be a logical extension of knowledge of a drug’s effect.

The literature has been inconsistent in defining pharmacological optimism. For example, one researcher defined pharmacological optimism as “a generalized positive belief about the efficacy of drugs for managing symptoms as measured by an individual’s willingness to use psychoactive drugs under varying circumstances” (p. 48). Although reported to be based on alcohol expectancy literature, the measure developed did not assess specific beliefs about the effects of psychoactive drugs but misoperationalized the concept as a general willingness to use psychoactive drugs.

A qualitative study of health care professionals suggested that pharmacological optimism was synonymous with the concept of “better living through chemistry” (p. 187), which suggests that all ills can be cured with a medication. Although pharmacological optimism was one of several key factors considered to contribute to substance use and abuse by health care professionals, the author did not further define or quantitatively assess the construct. Trinkoff and colleagues suggested that pharmacological optimism may occur as a result of highly specialized knowledge about drugs. For example, self-administration practices by health care professionals may occur as a result of the development of attitudes and or beliefs that drugs may be the quickest route to change one’s feelings and mood. Trinkoff et al. measured pharmacological optimism with the combination of access to drugs and knowledge of drugs, reporting pharmacological optimism to be significantly associated with past-year substance use, but only when access was not included in the analysis.

Other researchers suggest that pharmacological optimism is more specifically based on beliefs of a drug’s anticipated effect and conceptually similar to alcohol and other drug expectancies. Alcohol, marijuana, and cocaine expectancies are beliefs about the effects of these specific drugs that develop prior to and as a result of their use and show significant variability across levels of use.

To test this theory related to prescription medications, Kenna and Wood developed scales to measure pharmacological optimism and one’s willingness to use drugs of abuse. The authors administered a self-report cross-sectional survey to upperclassmen and graduate pharmacy and nursing students as well as comparable non–health care students ( n = 401). The results demonstrated that although pharmacological optimism predicted unique variance in drug use over a person’s willingness to use a drug, no differences were demonstrated between health care and non–health care students on pharmacological optimism, and pharmacological optimism was not associated with greater drug use by health care students. In sum, although the results support the existence of pharmacological optimism, these beliefs ultimately did not appear to facilitate drug use by health care students over and above experiential or occupational circumstances such as workplace access to substances.

Negative Proscriptions

Winick proposed a theory that the incidence of substance abuse is highest in those groups in which easy drug access, role strain, and disengagement from negative proscriptions exist. Disengagement from negative proscriptions regarding substance use may be an important correlate to one’s level of association with conventional institutions and subsequent risk for substance use. Although difficult to measure directly, both religiosity (internal factors) and social networks (external factors) have been hypothesized to be important conjoined factors to measure negative proscriptions. Religiosity has been hypothesized to be an internal factor that may mitigate substance use. One’s social network has been found to be an important external factor, linked to one’s reference group, norms, and peer group choices that may promote drug use. Trinkoff et al. tested the utility of this theory in 3600 nurses and reported that nurses were more likely to use drugs when drug access increased, social networks contained drug users, and religiosity decreased. These data also suggest that weak attachments to negative proscriptions (low religiosity and social networks that promote drug use) and high drug access are influences related to one’s drug use.

Social and Professional Influences

Social influences have been hypothesized to play a central role in models of substance initiation and are considered among the strongest correlates of alcohol use and misuse. Within the social influence framework, two types of social influences (active and passive) are proposed. Active social influences consist of explicit offers to use drugs or alcohol that require an immediate response from the individual offered the substance and are seen as important sources for substance use initiation. Passive social influences include both social modeling and perceived peer norms. Social modeling involves observing drug or alcohol use by one’s family or friends. Perceived norms are beliefs surrounding what referents consider or perceive as normal drinking or drug use that may affect both behavior and attitudes about alcohol or drug consumption. In short, these social influences are the means by which a person may gain information simply by observing another’s behavior or developing a sense or misperception of what level of substance use is ongoing and acceptable by peers. It is thought that this information may influence future behaviors. For example, social modeling by family members has been hypothesized to be a risk factor in nurses. Significant differences were found when comparing drinking behaviors in the families of chemically dependent and non–chemically dependent nurses; 32% of the chemically dependent nurses reported heavy drinking at home during childhood as compared with only 10% of the non–chemically dependent nurses. Moreover, Dabney [Dabney D (1997) A sociological examination of illicit prescription drug use among pharmacists. University of Florida, “An Unpublished Dissertation”] reported from the qualitative arm of his study of 50 recovering pharmacists that 30% of these health care professionals were encouraged by peers and supervisors that it was acceptable to use drugs to be able to function and perform at work. Kenna and Wood demonstrated that social influences, primarily active offers of drugs and to a lesser degree social modeling of drug use, were strong predictors of current drug use in college students. These analyses suggest that peer groups and social networks play an important role in the use of drugs in this population.

As an occupational hazard, health care professionals frequently receive offers to attend dinners at which alcohol is served without cost. These dinners are intended to inform health care professionals about specific treatment protocols or new drugs that have recently become available. For example, Kenna and Wood reported that physicians received significantly more offers to drink alcohol than other health care professionals offered alcohol by pharmaceutical companies ( P < 0.001). In the same study, it was also found that dentists received significantly more offers to drink alcohol in social situations than did other health care professionals ( P < 0.01). As for prescription medications, Clark and colleagues reported that physicians wrote more prescriptions for family, friends, and colleagues immediately upon receiving prescriptive authority. Although not statistically significant, these results do suggest that physicians, dentists, and potentially other health care professionals are asked to provide prescriptions or medications to friends, family, and other colleagues.

Age and Substance Use

The data from studies consistently demonstrate that older health care professionals drink significantly more alcohol than younger health care professionals, a finding that is supported by both quantitative and qualitative data in health care professionals. For example, McAuliffe et al. first reported that among physicians, heavy alcohol use (five or more drinks at one time on five or more days during the past 30 days) increased with age. Moreover, heavy alcohol use by pharmacists declined only slightly after peaking at ages 31–40 years. Notably, each professional group in the McAuliffe et al. study had their own distinctive trend of heavy alcohol use across age. Drinking habits among doctors were not associated with medical specialty or type of practice but were positively related to gender (men greater than women) and to age (older were more apt to qualify as heavy drinkers than younger doctors). Hughes et al. reported that physicians were more likely to report past-year alcohol use than age- and gender-matched cohorts to the general population. On the contrary, after peaking at ages 21–25 years, past-year alcohol use, binge use, and heavy alcohol use all decrease with age among US adults.

Qualitative studies also report that older substance-impaired health care professionals have a tendency to be more alcohol involved than younger ones. Retrospective studies with health care professionals suggest that younger health care professionals tend to use a greater variety of drugs in addition to alcohol than older health care professionals. For example, Talbott et al. examined the substance abuse patterns and specialties in 1000 substance-impaired physicians referred for treatment in Georgia and reported that younger physicians tended to be more polydrug involved than older physicians, who tended to use alcohol alone. General population data would suggest that combined alcohol and illicit substance use peaks in the range of 18 to 25 years of age. Similarly, in a study of nurses, Sullivan et al. reported that alcohol was the most common drug of dependency and that women and older registered nurses tended to use alcohol alone while younger registered nurses tended to report narcotic dependency more frequently. In a study of recovering pharmacists, Bissell et al. found that by the age of 49 years, the majority of pharmacists in their study abused alcohol alone.

Alcohol, Tobacco, and Other Drug Use in the Health Care Professions


Although there is no overwhelming evidence that dentists are at a higher risk for developing alcohol dependence than the general population, alcohol use/misuse appears to be the most notable substance use problem facing dentistry. As noted previously, dentists engage in other social interactions that promote their alcohol use. Nearly every major alcohol use standard assessed indicated that dentists were significantly more likely than other health care professionals surveyed to self-report use and misuse of alcohol. Compared with other health care professionals, dentists reported a higher lifetime prevalence of alcohol use, significantly greater past 30-day quantity and frequency of alcohol use, greater past-year and past-month binge drinking (five or more drinks for men or four or more drinks for women at one time), as well as greater daily use. When compared with the US general population data for individuals 35 years of age or older at the time, dentists reported a greater prevalence of lifetime alcohol use and past-year binge drinking. These data are consistent with retrospective treatment records of 2015 health care professionals in Georgia’s substance abuse treatment program in which dentists, in addition to physicians, were more likely to be exclusively alcohol dependent.

Information is limited as to why alcohol use, misuse, and problems appear to be higher among dentists compared with other health care professionals. Hankes and Bissell pointed out that good data regarding substance use for dentists were extremely limited; this has not changed. Putatively, causes could be linked to several risk factors previously noted in the general population, such as gender, family history, income, and social factors.

Certainly, one could assume that alcohol use differences may be related to the gender imbalance of more men in the dental profession, and men report more alcohol use than women. More so than in any other health care professional group surveyed, male dentists (85%) outnumbered female dentists, which was consistent with American Dental Association data, which reported that almost 83% of dentists were men. However, it is important to note that 74% of the physicians in the Kenna and Wood study were men, and physicians reported the least amount of alcohol use of any of the professions. In addition, although there were no differences between men and women regarding regular alcohol use, analyses demonstrated no gender difference in weekly quantity and frequency of alcohol use in health care professionals in general.

One potential explanation for the increased alcohol use by dentists may be related to higher income. In other words, the use of alcohol may increase with increased incomes. Previously, Hughes et al. noted that the prevalence of alcohol use was greater among physicians than in the general population by virtue of their socioeconomic class and not related to the profession of medicine. Consistent with Hughes et al., lifetime prevalence and past-year binge drinking reported by physicians were higher than reported for the general population. Although dentists did report the highest mean income of all health care professional groups, they also reported income only slightly greater than that of physicians, who reported the least amount of alcohol use of all health care professional groups. In short, dentists drank significantly more alcohol than physicians, yet dentists and physicians reported essentially the same income. Furthermore, nurses and pharmacists reported a substantially lower income than dentists and physicians and still reported greater lifetime prevalence of alcohol use and past-year binge drinking than the general US population and physicians. The results reported by Kenna and Wood support the notion that alcohol use may, indeed, vary by virtue of health care profession and independent of socioeconomic status.

As noted, social factors might significantly affect alcohol use by dentists. An underlying social structure defines and shapes a relationship between alcohol use, alcohol involvement, and group membership. One must consider that dentistry is also a business, and networking through memberships in various organizations is an important part of building a successful dental practice. Kenna and Wood reported that, during the past year, dentists were offered alcohol by friends and colleagues significantly more often than any of the other health care professional groups ( P < 0.01). In addition, although nurses knew significantly more heavy drinkers or alcoholics than pharmacists and physicians did in their social networks, there was no significant difference between dentists and nurses, perhaps suggesting that a strong social interaction component, more than any other risk factor, may contribute to alcohol use and misuse by dentists.


In 1991, Nancy Valentine’s study of Massachusetts nurses led her to conclude that alcohol use by her sample was low compared with use by the physicians, pharmacists, and students in the McAuliffe et al. study. More recently, however, alcohol use by nurses overall was surprisingly high given the overwhelming proportion of women compared with the other health care professional groups. Nurses used less (mean use) alcohol than only dentists surveyed, but not significantly less. Although nurses reported fewer mean drinks per day than dentists, they also reported more mean monthly alcohol use than pharmacists and physicians. Moreover, although nurses reported less past-year binge drinking than the other three health care professional groups, they reported more past-year binge drinking than the general population 35 years of age or over, despite the fact that they are largely female. Results from the Kenna and Wood study report higher lifetime (67%) and past-year (22%) binge rates compared with rates reported by Trinkoff and Storr in a comparably sized study (54.4 and 19.3%, respectively). In a larger study performed by Trinkoff and colleagues, 17% of nurses reported binge drinking during the past year ( n = 3919). In the Kenna and Wood study, 55% of nurses reported that they used alcohol on at least 4 days or more a month during the past year, and only 20.2% of nurses reported that they were nondrinkers.

Surprisingly, one of the primary substance use concerns for nurses compared with other health care professionals continues to be cigarette use. Padula previously suggested that the level of smoking in the nursing profession was unacceptably high, higher than other health care professionals, and should be cause for concern within the profession. Consistent with Padula’s findings, more recent research reported that past-month cigarette use by nurses was significantly greater than any other health care professional group, although the past-month rate was less than half the rate reported by similarly aged peers in the general population.

A significant number of nurses also report use of illicit drugs. For example, Trinkoff and Storr reported significant rates for lifetime (41%) and past-year (3%) marijuana use by nurses in their study. Lifetime marijuana use self-reported by nurses in the Kenna and Wood study was 57.4%, and past-year use was 4.7%, which was less than the 61% lifetime use reported by experimenters in the Sullivan et al. study of nurses, but consistent with the 37.3% lifetime use reported by nurses surveyed in New York and by 37% of nurses surveyed in Massachusetts [Valentine N (1991) Stress, alcohol and psychoactive drug use among nurses in Massachusetts. Brandeis University, Boston, MA, “An Unpublished Dissertation”].

Nurses also report extensive nonprescribed prescription medication use. For example, 14% of the nurses in the Kenna and Wood study reported ≥61 nonprescribed medication taking episodes, which was more than any of the other professions surveyed. Lifetime nonprescribed opioid use reported by nurses ranges from 52% [Valentine N (1991) Stress, alcohol and psychoactive drug use among nurses in Massachusetts. Brandeis University, Boston, MA, “An Unpublished Dissertation”] to 45.7 % to 21%. In addition, when combining one or more episodes of lifetime drug or medication use, 63.6% of nurses reported use of one or more drugs, which is less than rates reported by 68.6% of nurses surveyed by Trinkoff and Storr and 73.7% surveyed by Valentine [Valentine N (1991) Stress, alcohol and psychoactive drug use among nurses in Massachusetts. Brandeis University, Boston, MA, “An Unpublished Dissertation”]. However, when compared with the general population, with a few exceptions (e.g., cigarettes consistently, minor differences with cocaine and hallucinogens), lifetime, past-year, and past-month substance use rates were higher among nurses.

Although still high compared with the general population, one possibility for the reduced drug use among nurses over the last 20 years may be twofold. First, the mechanism for access to prescription medications in many facilities has changed a great deal. The link between access and drug use has been noted by many researchers [Dabney D (1997) A sociological examination of illicit prescription drug use among pharmacists. University of Florida, “An Unpublished Dissertation”]. In order to meet Joint Commission requirements to maintain strict controls over medications to promote patient safety, most hospitals use automated machines that control and dispense medications to nurses for patients. One of the major advantages of these dispensing units is to maintain accurate counts of controlled drugs that must be verified at each shift change. Potential unauthorized access can be more readily detected. Although speculative, the automation of dispensing has probably reduced the access-prescription medication use link for many nurses. Second, the steady decline of drug use in society in general is also likely an important factor in this observation.

A number of nurses in the Kenna and Wood study reported iatrogenic drug exposure as they had been prescribed medications such as minor (e.g., C-III–C-IV) and major (C-II) opioids by practitioners. This finding was consistent with Valentine’s [Valentine N (1991) Stress, alcohol and psychoactive drug use among nurses in Massachusetts. Brandeis University, Boston, MA, “An Unpublished Dissertation”] study that reported that prescribed drug use was much higher in nurses than other health care groups used for comparison, which led Valentine to conclude that “the path to dependence for nurses is use of drugs as a consequence of treatment under another provider’s direction” [Valentine N (1991) Stress, alcohol and psychoactive drug use among nurses in Massachusetts. Brandeis University, Boston, MA, “An Unpublished Dissertation”; p. 651]. Although the issue regarding gender has generally not been considered, data suggest that there is an increased likelihood that more women than men will visit a physician and receive a prescription. In a report on the ambulatory care of patients in the United States, women were 33% more likely than men to visit a doctor, even after accounting for pregnancy-associated visits. Moreover, the rate of physician’s office visits for such reasons as annual exams and preventive services was 100% higher for women than men. Furthermore, prescriptive habits differed as well. Women were more likely than men to receive medications such as nonnarcotic analgesics and antidepressants. The likelihood that a group such as nurses that is predominately female would report more iatrogenic contact with various addictive substances is consistent with this notion.


The data suggest that pharmacists are not more inclined than other groups of health care professionals to drink more alcohol. The mean drinks per month reported by pharmacists surveyed by Kenna and Wood was 18.4, which was comparable to the rate of 21.2 drinks per month reported by McAuliffe et al. in 1991 for pharmacists. One possible reason accounting for the slight difference might be attributed to the dominance of men (84%) in the McAuliffe et al. study compared with only 59% of men in the Kenna and Wood study. Twelve percent of pharmacists also reported past-month use of five or more drinks or binge drinking, which was comparable to 9.3% of physicians reporting binge drinking by Hughes et al.

Based on qualitative studies, alcohol remains one of the salient drugs of choice for substance-impaired pharmacists, although it is important to note that alcohol was rarely the sole drug of choice. Bissell et al. reported that only 21% of the pharmacists in their study were addicted to alcohol alone, whereas 77% were addicted to a combination of alcohol and other drugs, most always prescription medications.

As noted by Bissell et al., the combination of alcohol and prescription drugs presents a more formidable threat to pharmacists, and the pathway to addiction for most pharmacists who may become impaired is probably through polydrug use. That is, alcohol in combination with medications, most notably minor opioids and anxiolytics, comprise the bulk of current substance use by pharmacists. Seldom are pharmacists addicted to just a prescription medication.

Studies using varying designs suggest that pharmacists do use drugs and suffer the consequences of their use. These include quantitative, qualitative, retrospective, and combination designs [Dabney D (1997) A sociological examination of illicit prescription drug use among pharmacists. University of Florida, “An Unpublished Dissertation”]. There is currently no evidence that total lifetime or past-year drug use by pharmacists significantly exceeds that of other major groups of health care professionals or the general population. Yet, pharmacists by virtue of their ease of access may be at greater risk to use prescription medications than the general population. It appears that the greatest threat to pharmacists is nonmedical opioid, stimulant, and anti-anxiety drug use due to easy availability. McAuliffe initially reported what he called “nontherapeutic” opioid use by health care professionals. The health care professionals interviewed described their progression of opioid use that led to becoming addicted. A high rate of self-treatment with tranquilizers was also noted in the McAuliffe et al. study that included pharmacists. Most nonprescribed opioid and anti-anxiety drug use seems to be for self-diagnosed ailments and most stimulant use is reported to be for facilitating performance, such as staying awake, performing better, and studying.

As reported, lifetime stimulant use by pharmacists was 15.8%, and was greater than twice the general population rate. Access to drugs by pharmacists would most likely explain the differences in prescription medication use between the general population and pharmacists. Consistent with studies of resident physicians who gain access to prescriptive privileges at that stage of their medical career, opioid and benzodiazepine self-treatment represented the bulk of prescription medication use by resident physicians. The majority of pharmacists who report the unauthorized use of prescription medications, not surprisingly, initially did so after leaving college. Dabney reported that 40% of pharmacists surveyed had used prescription medications without a physician’s authorization, and 20% reported that they had done so five or more times in their lifetime. It was proposed that diversion was the primary source for obtaining these medications, and that access to medications is therefore a prerequisite to use for many pharmacists. However, as noted by Trinkoff and Storr, perhaps access alone is necessary but may not be sufficient to foster the conditions that promote drug use in health care professionals. Access in a permissive environment, coupled with drug use knowledge; the lack of education of the developmental dynamics of addiction ; and peer, academic, or occupational influences that do not dissuade substance use, appear to be the primary factors contributing to illicit prescription medication use [Dabney D (1997) A sociological examination of illicit prescription drug use among pharmacists. University of Florida, “An Unpublished Dissertation”]. Among other legal and ethical concerns for such behavior is the concern with self-diagnosing of their condition by pharmacists. A common theme with pharmacists who were interviewed for the Bissell et al. study was that although pharmacists sought help, they often misdiagnosed their substance use problem and sought ineffectual or misplaced support and did not see, or were unable to correctly diagnose, their own addiction.


With the exception of anxiolytics, physicians appear to be less likely than other health care professionals or the general population to use alcohol, tobacco, or other drugs. For example, McAuliffe et al. reported that physicians drank 20.3 drinks per month in their study and concluded that there was no reason to suspect that alcohol use by physicians differed from that by other professionals. More recently, physicians reported consuming an average of only 17.9 drinks per month, which was also the lowest monthly mean total of any health care professional group. Not only was alcohol use lower than in other health care professional groups surveyed, alcohol use was lower compared with previous data reported by Hughes et al. and McAuliffe et al. Physicians’ lifetime prevalence (92.3%) was similar to that of the general population (85.8%) as was past-year binge drinking (22.1% vs. 18.6%, respectively), although past-month binge drinking was much lower (7.7% vs. 16.2%, respectively).

As noted earlier, an important social-professional contributor to alcohol use includes alcohol served by pharmaceutical companies at education seminars. A Kaiser Family Foundation study of physicians found that of 2608 physicians polled, 61% received free meals, drinks, travel and tickets from pharmaceutical companies. Pharmaceutical companies are providing alcohol at continuing education seminars as a part of a marketing approach targeting practitioners. Some estimate that 12% of a pharmaceutical company’s marketing budget (12.5–15 billion dollars a year) is targeted at physicians. Although the use of alcohol by physicians does not appear to be a problem, alcohol continues to be served by pharmaceutical companies to provide sales representatives the opportunity to engage with health care professionals and facilitate conversations regarding the use of their particular medication. Kenna and Wood reported that physicians were offered alcohol by pharmaceutical companies, significantly more often than all other health care professionals.

As for other drug use, Kenna and Wood reported that the prevalence of lifetime street drug use among physicians exceeded rates reported by the general population, some other health care professional groups in the study, and rates reported previously by Hughes et al. For example, lifetime prevalence of marijuana use by physicians (51.9%) exceeded rates reported by the general population (31.6%), by physicians in the Hughes et al. study (35.6%), as well as rates reported by pharmacists (44.4%) and dentists (48.7%) in the Kenna and Wood study. Furthermore, Kenna and Wood reported that the lifetime prevalence of cocaine (17.3%) and hallucinogen (11.5%) use by physicians was the highest among health care professional groups used for comparison and higher than both the general population (11.8 and 10.1%, respectively) and the physicians surveyed (10.3 and 7.8%) by Hughes et al. Results regarding past-year use of street drugs by physicians from the study of Kenna and Wood, however, were consistent with data reported previously by McAuliffe et al. and Hughes et al., and self-prescribed minor opioid use among physicians was far lower in the Kenna and Wood study. In addition, past-year minor opioid use (1%) was only one-eighth the rate reported by pharmacists (8.3%). On the other hand, prevalence of anxiolytic use by physicians was the highest among health care professional groups and much higher than rates reported for the general population. Despite this, reported rates have been dropping. For example, past-year use by physicians was 4.8%, a rate higher than that reported by pharmacists (4.5%), nurses (3.1%), dentists (2.7%), and the general population, whereas past-year use of benzodiazepines (anxiolytics) was 11.4% as reported by Hughes et al. and 9% in the McAuliffe et al. study.

In terms of onset of use, initiation of street drugs often begins in college, high school, or earlier; only opioid and anxiolytic use began during residency. Hughes et al. suggested that unsupervised opioid and anxiolytic use could contribute to substance abuse or dependence in physicians, particularly in light of the results in their study that found that physicians were more likely than age and gender peers to have used alcohol, minor opioids, and anxiolytics.

Identifying Drug Problems in Health Care Professionals

Job performance issues such as excessive absenteeism, errors, frequently changing jobs, calling in sick or offering to work overtime, frequent wasting of medication, sleeping on duty, and always giving maximum doses of medications are a few of the important behavioral signs useful in identifying substance abuse problems in health care professionals. In addition, there are several symptoms of alcohol or drug abuse that coworkers may experience, such as memory blackouts, emotional lability, withdrawal from family or coworkers, depression, insomnia, slurred speech, disappearing frequently, or the odor of mouthwash or mints on their breath ( Table 68.1 ). In fact, some of these markers may hold the key to the type of impairment. For example, unexplained work absenteeism may indicate an alcohol abuse or dependence problem, since, in addition to hangover, home is the more convenient place to store and access this drug of choice. By contrast, consistently volunteering to work overtime, or arriving at work when not scheduled, provides an impaired coworker the opportunity to access controlled substances not available at home. Unfortunately, such symptoms are more easily identified with hindsight than before a coworker is identified with an alcohol or drug use disorder.

Jan 19, 2020 | Posted by in PATHOLOGY & LABORATORY MEDICINE | Comments Off on Substance Use Disorders in Health Care Professionals

Full access? Get Clinical Tree

Get Clinical Tree app for offline access