Professional Boundary Violations by Physicians

With the widespread recognition that the Hippocratic oath offers no insurance policy against professional misconduct by physicians, the problem of professional boundary violations in medical practice has received increasing attention. Although variously defined, professional boundaries may usefully be considered as “the parameters that describe the limits of a fiduciary relationship in which one person (a patient) entrusts his or her welfare to another (a physician) to whom a fee is paid for the provision of a service”.¹  Although sexual misconduct with patients is perhaps the most egregious and most widely publicized example of professional boundary violations, a variety of other problematic behaviors also require attention. The Massachusetts Board of Registration in Medicine has even issued detailed guidelines on professional boundaries of both sexual and non-sexual types.²  These guidelines apply only to physicians practicing psychotherapy and thus are less readily applicable to other specialties.

Boundary violations of non-psychiatric physicians have received less systematic elaboration. Physician behaviors may violate boundaries if they exploit the patient’s dependency on the physician.¹  These behaviors include sexual relationships, business transactions, large gifts, denigrating language, mishandling of fees, misuses of the physical examination, some types of physical contact and prescribing irregularities that involve dual relationships (where prescriptions are written for employees, family members, oneself, or persons for whom there is no medical record or doctor-patient relationship). Some professional boundary violations do not exploit patients, but involve the treatment of employees, nurses or other allied health professionals in ways that are sexually harassing or otherwise disrespectful of personal space.

In a report on 375 physicians licensed by the Medical Board of California that examined records of discipline from October 1995 to April 1997, 465 separate offenses were identified.³  The most common involved incompetence or negligence and abuse of alcohol or other drugs. However, professional boundary violations involving inappropriate prescribing practices (11 percent) and inappropriate contact with patients (10 percent), were in third and fourth place, respectively. Despite the widespread frequency of this type of professional misconduct, the varieties of professional boundary violations seen in a clinical context have not been well described. In this article we seek to characterize a large series of boundary violations by physicians in the clinical setting of outpatient and/or inpatient evaluations. Only a subgroup of these came to the attention of licensing boards and resulted in discipline.

Since the late 1980s, the senior author has conducted outpatient or inpatient evaluations of physicians and other professionals, first at the Menninger Clinic in Topeka, Kan., and subsequently at the Baylor Psychiatry Clinic at the Baylor College of Medicine in Houston, Texas. Patients have been referred for these multi-disciplinary, three-day outpatient evaluations from many different states and from Canada.

In preparation for this report, we reviewed 159 records of these evaluations and identified professional boundary violations in 100 of the physicians. In our systematic review of these cases involving violations, we recorded basic demographic data, including date of the evaluation, referral source, age, gender, specialty, years in practice and practice setting. This latter category was divided into four categories: private practice, academic, public sector/military and training program. The physician’s cultural background was also recorded. In addition, the following clinical information was also documented: type of boundary violation, presence or absence of substance abuse and Axis I and Axis II psychiatric diagnoses.

The sample was limited to those physicians with a M.D. or D.O. degree. Years of practice included all years of post-medical school, including training programs. Professional boundary violations were defined as they are in the introduction to this communication. We included prescribing irregularities where the person being treated was also a family member or employee who may or may not have paid for his service. Also, where physicians prescribe for people with whom there was no medical record or a doctor-patient relationship established, a professional boundary violation was deemed to have occurred. Finally, physicians prescribing for themselves were regarded as blurring professional boundaries as well. Substance abuse referred broadly to both alcohol and drugs, whether street drugs or prescription drugs. Sexual harassment was considered part of professional boundary violations where there was a hierarchical relationship between the physician and the target of the harassment, whether employee, student or co-worker.

Simple incompetence or negligence was not regarded as a boundary violation. Similarly, irresponsibility, characterized by failing to return calls or poor charting, was not included. Finally, physicians who exploded in anger or committed fraud in terms of billing practices did not meet criteria for inclusion in this review.

Physicians who commit boundary violations do so for a variety of reasons.^1,4,5  Some physicians who engage in sexual relations with patients have fallen madly in love with the patient and are otherwise ethical practitioners. Others are sexual predators and systematically exploit the power differential in the doctor-patient relationship. Prescribing irregularities may include corrupt physicians who function like drug dealers, benevolent overprescribers who try to please their patients by getting them the treatment they desire, and those who are addicts themselves and feed their habit by prescribing for friends and family members, only to use the drugs themselves. In any case, the underlying motive for the boundary violation is not considered in this review of records. Our goal is merely to characterize the types of boundary violations seen in the clinical setting. Some physicians in this sample came to clinical attention because of the threat of disciplinary action by a board, physicians’ health organization, hospital, or lawsuit, while others were self-referred and were seeking help for problems they recognized in themselves and in their practice. Still others were referred by a psychiatrist or other mental health professional.

Table 1 reflects the demographic and diagnostic information relevant to this sample. Ninety-six percent of the physicians seen were male and 92 percent were Caucasian. The ages ranged from 27–74. Many specialties were represented, but psychiatry was by far the most common, with 41 percent, followed by family practice at 16 percent. Of the practice settings, 86 percent were from the private sector, with small numbers coming from training programs, academic settings and the public/military sector.

The most common referral source was a physicians’ health organization, often functioning independently from a licensing board and more interested in providing some form of monitoring and rehabilitation than in disciplinary action. Twenty-three percent were referred from licensing boards. Eleven of the physicians were self-referred because of their own concerns. Other sources of referral included attorneys, ethics committees, hospitals, residency training programs and practice partners.

Psychiatric diagnoses were common in this sample. Fifty-two percent met criteria for an Axis II personality disorder. Twenty-six were diagnosed with a mood disorder. Seventeen physicians had substance abuse disorders; eight of those were in remission at the time of the evaluation. Paraphilias and sexual disorders were found in 13 of the physicians. Seven had anxiety disorders. Some of them had multiple diagnoses reflecting the fact that there was considerable co-morbidity in our sample. Twenty-seven had no psychiatric diagnosis.

In organizing the types of boundary violations committed by the physicians in our sample, we divided them into three overall categories: 1) boundary violations with a patient, 2) boundary violations with non-patients (employees, co-workers, family members) and 3) prescribing/treating irregularities. We recognize there is some degree of overlap across these categories. Boundary violations with patients are further subdivided according to whether they are sexual or nonsexual. Sexual misconduct in physicians is usefully regarded as involving one of three categories delineated by the Medical Council of New Zealand (1992).⁶  Sexual impropriety refers to gestures or expressions disrespectful to the patient’s privacy and sexually demeaning to the patient. Many cases of sexual harassment involving unwanted advances, sexually explicit remarks and denigrating comments would fall into this category.

The second category is sexual transgression, which involves sexualized and inappropriate touching of the patient that falls short of actual sexual relations. Such behaviors as kissing, touching of the breasts or genitals not appropriate for the physical exam, or performing a physical exam without gloves, all would be included under this grouping.

Finally, sexual violation proper is the third category, and this refers to physician-patient sexual relations. This category applies regardless of whether the patient or the physician initiates the contact and irrespective of whether the two profess love for one another. Included in this category are oral sexual relations, anal intercourse, genital intercourse and mutual masturbation.

Using these categories, sexual boundary violations proper with patients were identified as problems in 53 of the physicians in the sample, making it by far the most common boundary violation (see Table 2). Of these violations, 43 involved a current patient or patients, and 10 were related to a former patient. Fifteen of the sexual violations were transgressions. Fourteen cases of sexual impropriety occurred in the sample and ranged from a physician who unzipped his fly in front of a patient to another who walked into an examination room and told his female patient that she “smelled like a French whore.”

Eighteen nonsexual violations occurred with patients in this sample of physicians. Five involved financial matters, such as soliciting a donation for one’s own building or borrowing money from patients. Six were social in nature and ranged from staying in a patient’s home to taking a patient and her child to dinner. Only three confidentiality boundary violations were reported. Some nonsexual violations were not easily categorized and involved such things as helping a patient move, frequent extensions of psychotherapy sessions, using drugs with a patient and simultaneously treating and supervising a psychiatric resident.

Turning to boundary violations with non-patients, all 22 were sexual in nature (as there is very little agreement on what would constitute a non-sexual boundary violation involving non-patients). Ten involved sexual boundary violations proper, wherein a physician had sex with a supervisee, employee or co-worker beneath the physician in the hierarchy of the institution or office, or with a family member of a former patient. Two involved transgressions and 10 were cases in the sexual harassment spectrum that are classified as impropriety.

The category of prescribing/treating irregularities addresses instances where there is a blurring of boundaries about one’s role. Nine cases of prescribing for oneself were reported. In some cases, prescribing narcotics for oneself was identified as a problem, while in other cases physicians prescribed drugs for a patient and then had the patient turn them over to the physician for his/her own use. Prescribing for employees occurred six times in this sample. Prescribing for and treating family members accounted for three instances. In one case, a physician performed a pelvic exam on his teenage daughter. In another case, controlled substances were prescribed for the physician’s spouse. Trading sex for drugs occurred on two occasions. Another problem that was reported six different times involved prescribing for or treating someone without a medical record. These instances ranged from prescribing medications for friends and family, treating female employees at a massage parlor that a physician frequented, to treating the 13-year-old daughter of a woman a physician was dating.

This sample of physicians is by no means representative of all physicians who are referred for clinical evaluations by licensing boards, physicians’ health organizations and other agencies. The sample is skewed in the direction of psychiatry and boundary violations because of the senior author’s extensive writings on boundary violations among mental health professionals.^7–10  Nonetheless, the 100 physicians seen in evaluation in a clinical setting provide useful information about the types of boundary violations seen and the psychopathology of the physicians. Of note in this regard is that 52 percent of the physician sample met criteria for an Axis II personality disorder. By far the most common personality disorder was “Mixed, or personality disorder not otherwise specified,” meaning that criteria were not met for a pure personality disorder in most cases, but a mixture of personality characteristics of different personality disorders was present. In a recent presentation at the Annual Conference of the Academy of Organizational and Occupational Psychiatry, Schouten presented data from his experience with 82 cases of physicians who had been referred for evaluation because of disruptive behavior.¹¹  He too found that personality disorder not otherwise specified was the most common diagnosis.

A substance abuse disorder was diagnosed in 17 of the cases, and at least 13 had some type of paraphilia or sexual disorder. As noted above, comorbidity was quite common in the sample, so that one diagnosis could rarely be regarded as the explanation. Twenty-seven of the physicians had no psychiatric diagnosis at all, indicating that boundary violations are not simply the outgrowth of psychopathology.

In reviewing the varying types of boundary violations, it is clear that multiple violations are common. The well-known “slippery slope” phenomenon is often at work, in which a nonsexual boundary violation leads to increasingly egregious violations and finally results in sexual relations between doctor and patient.⁸  Sexual boundary violations are by far the most common in this sample, likely because of the senior author’s writings. Forty-three of the sexual violations proper occurred with current patients and 10 with former patients. While the American Psychiatric Association Ethics Code states unequivocally that sex with former patients is unethical, the AMA Council on Ethical and Judicial Affairs (1991) regards sexual relationships between former patients and physicians as potentially unethical if exploitation of a still emotionally dependent patient is involved.¹²  Whether the doctor and patient are “in love” or married is not relevant to whether exploitation has occurred. The inherent power differential between doctor and patient makes it difficult for a patient to give fully informed consent to a sexual relationship.

Some of the nonsexual boundary violations, such as having coffee with patients, may seem trivial, but in some of these instances, the patient misconstrued having a ride with the physician or going for coffee as a sexual overture. This misunderstanding reflects the fundamental principle that the physician’s intent may be different from the impact on the patient. Physicians may cross boundaries with a perfectly reasonable and benevolent intent, but patients can easily experience the departure from normal professional role as violating, particularly if the patient has a history of sexual abuse.

Most of the boundary violations occurring with non-patients, such as employees, co-workers, and family, involve variations on sexual harassment. Power differentials exist between doctor and employee, and they also exist in a hospital hierarchy even when there is not a specific contract of employment between a physician and another employee of the hospital who might be a nurse, an X-ray technician, a medical student or a secretary.

The last major category, prescribing/treating irregularities (see Table 2) addresses a pervasive problem in the medical profession. Many physicians do not place themselves under the care of a colleague, and they often use samples to treat themselves. Many also prescribe for employees, family members and others with whom there is no doctor-patient relationship. The American Medical Association’s Principle of Medical Ethics has long advised against the practice of treating oneself or one’s family members because of a lack of professional objectivity. Nevertheless, in our sample, nine physicians prescribed drugs for themselves, including controlled substances. Six prescribed for employees. Moreover, some of these transgressions involved treating family members that went beyond prescribing. One physician performed a pelvic examination on his teenage daughter, for example, and another physician traded sex for drugs. Treating or prescribing for people without a medical record or a doctor-patient relationship was also common. In one case, a physician treated the 13-year-old daughter of a woman he was dating. In another case, a physician routinely prescribed informally for friends who had complaints.

Space considerations do not allow for detailed discussion of prevention or treatment. Suffice it to say that education on professionalism in medical school is crucial. Moreover, identification of medical students with problematic professional behaviors and remediation for those students are badly needed.¹³  Many physicians who commit boundary violations are amenable to rehabilitation, but a careful evaluation is needed to rule out those with antisocial or severe narcissistic personality disorders who are unlikely to respond to a rehabilitation program.⁹