t is well established that within the physician community, anesthesiologists — and most often those in training to become anesthesiologists — are among those at greater risk to become chemically dependent on alcohol or drugs. Although the incidence of dependency on alcohol at some career point does not differ materially from other specialties, the most recent comprehensive national survey of controlled substance abuse in anesthesiology training programs disclosed abuse — principally with fentanyl and sufentanil — by 1.6 percent of residents and 1 percent of faculty. At any given time, the percentage of anesthesiologists or residents enrolled in drug addiction recovery programs, as distinct from other specialists, is about three times the percentage of anesthesiologists included in the physician population as a whole.

In the abstract, these data suggest that possible drug addiction is a problem deserving the serious attention of the specialty. Anyone who has had firsthand knowledge of a drug addiction event, however, also knows that this is hardly an abstract statistical issue. A lot of addicted physicians die from drug overdoses. Both a former president of a major ASA component society and a former president of the American Association of Nurse Anesthetists are among this number. I am told that a projected “star” of a new education videotape on addiction within the specialty — a recovered addict — died from a drug overdose before the film could be made. According to the most recent survey, almost 20 percent of chemically dependent anesthesiologists die or require resuscitation. Among those addicted anesthesiologists who do not die, I understand, the recidivism rate is very high.

The greater incidence of drug addiction of anesthesiologists among physician specialties is attributed to a combination of factors, among them the high addictive quality of anesthetic drugs like fentanyl and sufentanil (which together constitute about 80 percent of the drugs of choice in addiction cases); the ready availability of these drugs and the fact that anesthesiologists, unlike other physicians, personally and routinely administer them intravenously to patients; the relative ease with which these drugs can be diverted to personal use without detection; and the intimate knowledge of drug properties that necessarily is a part of the specialty’s training.

Repeated surveys have demonstrated that addiction occurs more frequently among residents than practicing anesthesiologists; more than 50 percent of addicted physicians in the specialty are under the age of 35, and more than 65 percent are connected with academic programs.

Despite vigorous efforts by ASA and academic anesthesiology leaders to deal with the matter, the incidence of controlled substance abuse among anesthesiologists and residents has reportedly not changed significantly over the past three decades and remains a major problem. Although no comprehensive survey data exist as to the extent of the problem in nonacademic settings, and although existing data show that the greatest incidence of abuse exists among anesthesiology trainees, it is clear that the problem is not confined to academic centers.

Initiatives by organized medicine, and more specifically anesthesiology, to combat substance abuse have principally involved heightened education requirements or opportunities. The Accreditation Council for Graduate Medical Education has required substance abuse education to be included in anesthesiology training programs since 1992. ASA’s Committee on Occupational Heath has published its “Model Curriculum on Drug Abuse and Addiction for Residents in Anesthesiology.” This curriculum, which is in use in several training programs, can be accessed through the ASA Web site at <www.ASAhq.org/clinical/curriculum.pdf> and could serve as a valuable education tool in nonacademic settings as well.

As we are all aware, however, educational efforts have been only part of the response to this societal problem. Beginning in the 1980s, essentially following the drug-free workplace initiative of the Reagan Administration, public and private employers began to turn to employee drug tests as another means of attack.

By the mid-1990s, a majority of employers appear to have put drug-testing programs in place, especially with respect to applicants for employment. More recent data suggest that the enthusiasm for drug-testing in the corporate world has been falling off somewhat, essentially because testing programs are expensive, the number of positive tests in the employment applicant or employee pool have been relatively small, drug testing is not a perfect science and because debate continues to exist as to whether drug testing actually improves overall employee performance.

Although I can find no persuasive national data, it appears that the pattern for health care institutions has been similar; numerous institutions in the early years put testing programs into place, but it also appears that the enthusiasm more recently has been waning. I have been startled to learn that the American Hospital Association (AHA) possesses no survey data on the extent of testing by its members. When I discussed this phenomenon with one of my labor law colleagues last month, she cynically speculated that the “hospitals did not want to know which of their employees were drug impaired — at least not if it meant operating a comprehensive and expensive testing program.” Even the American Medical Association (AMA), while endorsing suspicion-based testing a few years ago, noted that there was no hard data in the health care field correlating drug impairment with diminished professional performance. Patient safety is only part of the problem: the toll on the addicted anesthesiologist and his family (most addicts are male) is severe.

Drug testing is, of course, controversial, legally and otherwise, because it necessarily involves an invasion of personal privacy. In the case of public employers (and public employers can include not only governmental agencies but also publicly owned entities such as hospitals), it raises the specter of an unreasonable search prohibited by the Fourth Amendment to the federal Constitution.

Private employers must be concerned with a claim based on a privacy provision contained in some state constitutions, or more often on the common law tort, created by the courts, of invasion of privacy: the so-called “right to be left alone.” In general, the protections of the Fourth Amendment are better defined than the prohibition against invasion of privacy. This means that private employers tend to enjoy greater legal flexibility in operating drug-testing programs than public employers, but as we shall see, both types of employers must establish a credible basis for deciding to operate a program that can withstand legal scrutiny.

Not all drug-testing programs are targeted at the same group of individuals or carry the same legal risks. Programs include pre-employment testing, suspicion-based testing of existing employees (or those engaged in safety-sensitive work), random or periodic testing of employees and postincident or postaccident testing. Some forms of testing are more controversial than others, but all have the common denominator of requiring invasion of an individual’s personal privacy. All forms of testing also involve the need to balance the common good, whether among railroad passengers or hospital patients, against the reasonable privacy expectations of the test subject.

Without a doubt, pre-employment testing is the least controversial testing regimen, especially when conducted as part of a pre-employment physical examination. The courts’ rationale, of course, is that job applicants have less of a vested interest in a position than do those already employed. If applicants do not want to be tested, they are free to seek employment elsewhere. Although pre-employment testing has received widespread public acceptance, employers instituting such a program, especially public employers, will be well advised to state to all applicants the rationale for the program. At least one court has struck down pre-employment testing by a public entity when the employer failed to establish any job-related basis for testing. To the extent, moreover, that an employer adopts a uniform policy for all applicants, or at least all those seeking safety-sensitive positions, the greater the likelihood the pre-employment program will be upheld.

At the other end of the testing spectrum, from a legal perspective, is random testing of existing employees. Statutes in 12 states prohibit random suspicionless testing, but many of these statutes do not apply to testing of those holding safety-sensitive positions. Unless constrained by statute, courts have to date uniformly upheld random testing when the employer established that positions involved were safety-sensitive and that selection of subjects was truly random; that is, did not involve administrative discretion. In essence, the public interest has been found by the courts to outweigh the privacy interests of individuals, including health care workers, engaged in safety-sensitive occupations.

Falling in the same “suspicionless” category as random testing is so-called mandatory or periodic testing of some or all categories of employees. This form of testing can involve preannounced testing of employees, without regard to suspicion, or, for example, mandatory testing of any employee involved in a work-related accident.

Although it may be easier to justify postincident mandatory testing than random testing of every employee, both forms of testing survived scrutiny by the Supreme Court in 1989 in two companion cases.

One case involved after-accident testing of railroad crews, and the other involved testing of customs service employees seeking transfer to positions involving drug interdiction and requiring the carrying of firearms. Suspicionless testing was upheld in both cases as not violating the constitutional protection against illegal search and seizure. Taken together the decisions stand for the proposition that in evaluating suspicionless programs, the special needs of the public entity will be balanced against the reasonable privacy expectations of the individual.

Where a special public interest is not present, however, the individual will prevail. In 1997 the Supreme Court struck down a Georgia law requiring testing of candidates for public office; the court noted that there was no evidence before the legislature concerning drug abuse among political candidates and noted, moreover, that candidates are rarely involved in safety-sensitive undertakings.

Just to demonstrate how far the current Supreme Court is prepared to go, though, note should be taken of its most recent decision in the area, in 2002, in which the court upheld a school district’s requirement that all students wishing to participate in an extracurricular athletic activity be subjected to a drug test. The court said that the students, by attending a public school and seeking to participate in interscholastic activities, had a limited expectation of privacy and that the school district had a legitimate interest in ensuring that the students were drug-free.

Aside from pre-employment testing, testing based upon a reasonable suspicion that an employee at work is drug- or alcohol-impaired today enjoys the greatest level of acceptance. Both AHA and AMA have supported suspicion-based testing as well as pre-employment testing; both organizations have stated opposition to random testing. There is obviously something offensive about the idea that an employer, without individualized suspicion, could require that an employee allow his or her bodily fluids to be analyzed.

One of the virtues of a random program is that, properly administered, it subjects individuals to testing truly at random. Not so, however, the suspicion-based program, which singles out an individual for testing on the basis of perceptions. It is clear that for such a program to survive attack, a protocol must be in place defining the basis for singling out and identifying those administrators, preferably more than one, who must agree that reasonable suspicion exists.

What is reasonable suspicion of drug-based impairment? One court has said that reasonable suspicion requires “some quantum of individualized suspicion as opposed to an inarticulate hunch.” That is catchy but not terribly helpful. The California Society of Anesthesiologists suggests that there is reasonable suspicion when “the party making the request [for testing] has determined, in his judgment and on the basis of personal observation or apparently reliable information, that there is a reasonable basis for suspecting that an individual may be under the influence of alcohol or illegal or inappropriately used drugs.”

The part relating to “apparently reliable information” is important: It does not appear that the courts will require, in either the public or private context, that the reporting party personally observe indicia of impairment but instead may base the decision on apparently reliable information offered by others.

A recent Missouri case is instructive in this regard. In Reeves v. Fulton State Hospital, a hospital for the criminally insane, the institution received an anonymous tip that two security guards were illegal drug users. Hospital administrators demanded that the two provide a urine sample for testing, and when they refused, they were summarily discharged. The Missouri appellate court held that the reasonable suspicion requirement could not be met solely on the basis of an anonymous tip, and when the hospital administrators failed to corroborate the information anonymously received, they were off base in discharging the employees for refusing a drug test.

ASA’s model curriculum on drug abuse contains a comprehensive description of the symptoms of addictive disease, exhibited both inside and outside the hospital, and there are numerous other publications that do the same. Awareness and accurate identification and documentation of these symptoms in individual cases is important from a legal point of view in order to avoid potential liability for false allegation of impairment in a suspicion-based program. Stated otherwise the courts are going to require that suspicion be based on observation of the classic indicia of drug impairment or reliable hearsay as to those indicia, and although they will surely allow some margin of error when the position involved is safety-sensitive, not all erratic behavior can be assumed to be based on drug addiction.

It is very clear that in order for a drug-testing program to work, the terms of the program should be carefully defined and announced in advance. One of the failures in the Missouri case just discussed was that the hospital had no such policy placing employees on notice that they might be subject to testing. Testing procedures should be designed with a sensitivity to limiting invasion of personal privacy consistent with ensuring accurate results. Obviously the use of experienced testing personnel and accredited testing laboratories enhances judicial approval of the testing process.

The bottom line here is that the courts, especially in cases where safety to others is an issue, have consistently sustained well-crafted drug-testing programs of several varieties. For this specialty, where the risks of abuse are clearly higher that in almost any other occupation, this is good news. I think the question remains as to whether anesthesiology departments and the institutions to which they belong will use testing as well as education as means of aggressively attacking the problem.

Although I have given you an overall outline of the way in which legislatures and the courts have dealt with employment-related drug-testing, you should remember that any program should be constructed in a way that is consistent with the law of the state in which it will be operated. Certainly no hospital or department should establish a testing program without having it reviewed by counsel prior to implementation.

Michael Scott, J.D., is the former ASA Director of Governmental and Legal Affairs. He retired from ASA in December 2004.