Dr. Strasburger is assistant clinical professor of psychiatry at Harvard Medical School in Boston, MA, and attending psychiatrist at McLean Hospital in Belmont, MA.
Acknowledgments: The author reports no financial, academic, or other support of this work.
Boundaries are defined as the structures and limits of psychiatric treatment which enable treatment to proceed. Boundary keeping is an essential aspect of medical and psychiatric treatment—a process which protects the patient and informs the clinician. Boundary crossings occur from time to time and, if properly attended to, can further the treatment. However, if not dealt with effectively, boundary crossings can become boundary violations—a destructive form of unethical behavior on the part of the clinician. This article will discuss various forms of boundary crossings and violations and offer suggestions for risk management through preventive measures.
The concepts of boundaries and boundary keeping have attracted attention in recent years, partly as a result of media focus and partly as the result of a series of malpractice awards for purely emotional damages.1 Yet, this awareness needs to be renewed and deepened with respect to training about boundary issues (D Norris, MD, TG Gutheil, MD, LH Strasburger, MD, unpublished data, 2002).2 The clinical concept of boundaries makes possible the professional work of medical practice and psychotherapy, permitting them to flourish. The handling of boundary issues is as important to therapeutic technique as it is to risk management.
Understanding the concepts of boundaries and boundary keeping is central for medical practice, psychotherapeutic technique, and risk management.
One can think of a treatment boundary as a spatial metaphor describing the interpersonal frame defining the edge of appropriate behavior in the therapeutic relationship. The metaphor “Good fences make good neighbors,” expressed in Robert Frost’s poem “Mending Wall,”3 is a good example.
In our work as psychiatrists, boundaries constrain the doctor in order to enable treatment. This constraint on the clinician imposed by boundaries contains and protects the patient, the clinician, and the therapeutic frame. Once the structure of treatment is established and the limits are formed, the process of human interaction can carry the treatment forward. Boundaries outline the stage on which the drama of clinical interaction can be played out. However, where the boundaries are breached or nonexistent there is no theater because the patient does not experience the safety and freedom to conduct an inner exploration.
The conscious application of a clear sense of boundaries should be thought of as a technique to enable and enhance treatment rather than merely as ethical protection for the patient or legal protection for the treating clinician. Keeping boundaries thoughtfully and carefully allows the clinician to experience situations redolent with the patient’s issues in the immediate present of the consulting room. These encounters can be discussed in therapy in terms of their meaning to the patient; it is just this discussion that preserves the therapeutic force of the doctor-patient relationship.
The issue of boundary keeping is a significant one whether the treating clinician is engaged in dynamic psychotherapy, general medicine, behavioral treatment, or psychopharmacology. All areas of medicine will be involved in a power imbalance with the patient and may be the object of magical wishes and attributions. Any doctor or clinician has an ethical responsibility to act in the patient’s interest and not to exploit a vulnerable patient for personal gain.
Although not every physician practices Freudian psychotherapy, Freud’s ethical principle of abstinence is the paradigm case for the theory and analysis presented here, and the ethical principles apply to all. According to Freud’s principle of abstinence, the clinician must refrain from improper gratification of himself from the medical relationship. No matter how “creative” his approach, the doctor must stick to business, remaining in role and at task.
Patients who are trusting (sometimes excessively) and frequently regressed often do not act self-protectively. Expecting to receive help, these patients assume they can let their guard down, creating vulnerability. Compounding this is an immediate power imbalance in the relationship in which the treating clinician has education, experience, status, and authority. For example, doctors in our society can ask almost any imaginable question, no matter how personal, to a patient. To someone who is unhappy or in trouble, the treating clinician appears to be a strong, successful person who has all the answers. The asymmetry exists even when one is treating a peer (ie, a fellow clinician).
Medical treatment is a system in which mere knowledge of the process confers power. Through experience, the clinician knows what to expect and has the power to influence the patient’s development in many ways. He can confer a sense of personal value simply through his paying attention, inspire emotional attachments, persuade the patient to do what is best for himself, and, alternatively, create confusion and conflict.4
Intimacy and the intensity of strong feelings in the clinical relationship may foster regressive longings, needs, and fantasies for both clinician and patient while weakening the good judgment required for self-control. When a patient’s infantile yearnings are revived, they often take the form of sexualized dependent striving. If the patient or treating clinician confuses these with adult sexuality, “familiarity breeds attempt” (SJ Welpton, MSW, oral communication, 1982). Patients more likely to exhibit this dynamic are those with borderline personality disorder5 or those who may have been previously abused “sitting ducks.”
A lot of concern about professional boundaries has emerged because of the wish to prevent mild, nonsexual transgressions leading to frank sexual exploitation of patients. Sex with patients is a boundary violation; one would think that this hardly needs saying, yet it continues to occur and typically begins with minor boundary crossings.4
Boundary crossings are benign interactions which vary from traditional medical interactions but do not harm the patient. When doctor and patient are able to discuss these interactions and they are not part of a repetitive pattern, they may improve the relationship and advance the treatment. What seems to be a minor boundary crossing might be therapeutically beneficial (if properly handled), neutral (innocuous), or a boundary violation (improper, possibly dangerous in itself and possibly leading to more serious violations). Of course, a crossing that is not itself an unethical boundary violation may still be harmful to both the patient and treatment depending on the patient’s sensitivities and the doctor’s skill in handling them.
Answering two important questions can determine whether an interaction is a boundary crossing or a boundary violation: (A) Is this action a part of the professional role? and (B) Can the interaction be discussed? Consider some examples: a patient is unavoidably late after tending to a sick child and the doctor allows the appointment to extend beyond its usual limit; a patient stumbles upon leaving the office, falls to the floor, and the doctor helps the patient up making sure he/she is not hurt; a patient announces that his/her son has just died and the doctor accepts the patient’s tearful embrace. Each of these interactions represents a boundary crossing but probably not a boundary violation. Flexibility, human responsiveness, and warmth are still professional.
Other situations that rub the edge of boundaries include chance encounters, use of first names, and a variety of contact possibilities. It is a small world, and one may easily encounter a patient at the movies, in a museum, or at a party. It would be rude not to say hello, but ill-advised to get into an extended conversation. The significant risk of boundary violation here would be breaking confidentiality through openly acknowledging treatment. For chance encounters, a humane response is indicated.
The use of first names can become a boundary issue. How should the patient be addressed? To many patients, the use of first names is a familiarity that is expected because we live in an informal society. To some, the use of first names by a doctor feels condescending and infantilizing or overly familiar. Asking how someone wants to be addressed turns the issue into one of proper respect, though explicit permission does not always resolve the issue. The doctor must always use clinical judgment. A patient may beg to be called by a first name yet be unable to handle the intimacy which is engendered. A similar issue forms around what the patient calls the doctor.
Context frequently determines whether behavior is a boundary crossing or a boundary violation. For example, a doctor who gives a patient a ride home in a blizzard might be judged differently depending on whether public transportation was still running, whether the patient felt coerced by the doctor to accept the ride, and whether the doctor also gave the patient rides in mild weather. Assessment of the propriety and therapeutic appropriateness of this gesture also depends on the doctor’s conduct in the course of the ride (maintaining professional bearing, avoiding undue familiarity, not conducting therapy in the car) and whether the doctor explored the unusual encounter with the patient in a subsequent session. The latter brings the encounter into the context and boundaries of treatment, restoring the boundaries that have been breached.
Physicians almost always see their patients when hospitalized, but a psychotherapist deciding whether to call or visit an ill patient in a general hospital has to use careful clinical judgment. The benefits would have to be weighed against the patient feeling exposed or intruded upon. The context might depend on the patient’s history, the seriousness of the illness, the nature of the therapeutic alliance, and the patient’s preferences. Frank discussion clarifying the issue with the patient would be of considerable assistance.
The type of treatment is a factor in considering the appropriateness of professional behavior. Case management and most medical treatments are more (figuratively) hands-on than traditional psychodynamic psychotherapy. It might be entirely appropriate outreach for a case manager to render practical assistance to patients, arrange appointments, encourage socialization, set up transportation, or engage with families. For example, drug and alcohol counselors may attend Alcoholics Anonymous meetings with their counselees, a game of catch with a latency boy might be an important approach to forming an engagement and therapeutic alliance, a behavioral doctor might assist an agoraphobic patient in going to a shopping mall to overcome his fear, or a child custody evaluation might involve a home meeting. Each of these situations could be regarded either as a boundary violation or acceptable behavior depending on context or type of treatment. Different kinds of treatment have different boundaries, so that what might be a boundary violation in psychodynamic therapy would not even be a crossing in another context.
Self-disclosure, albeit a common precursor in cases of therapist-patient sexual exploitation, can be a means of adapting the frame to connect with the patient.6 With some patients who have difficulty talking, and particularly with adolescents, one might want to discuss sports or movies as a way of engaging the patient in the treatment process. In the course of this discussion a doctor may reveal a great deal about personal interests and activities. The boundary issue is not whether the treating clinician self-discloses, since people self-disclose whenever they open their mouths. The vital question is what is disclosed and for what purpose? When a doctor burdens a patient with his own problems, self-disclosure has gone too far. To establish alliance, what does the patient need? What withholding will the patient tolerate? Whether a clinician’s self-disclosure is a boundary crossing or a boundary violation may be subjectively determined by the patient’s needs.
It is important that the clinician not misinterpret the concept of boundaries as prohibiting frank clinical inquiry about emotionally charged issues. One must take a sexual history from a patient, explore a patient’s fantasy life, and understand erotic or sexual transference feelings that can attach to any doctor. Anxiety about sexual boundary violations should not lead to avoiding discussions of a sexual nature with patients. In the clinical setting, open discussion should be the rule of thumb. Context and type of treatment are crucial determinates.
Boundary violations represent a failure to define and carry out the interaction as professional, not personal, social, or sexual. Boundary violations that stop far short of sex can still compromise treatment; they comprise many of the problematic interactions which precede frank sexual encounters.7 While boundary violations do not, in and of themselves, prove sexual misconduct, it is worth noting that juries give more credence to allegations of sexual misconduct where other boundary violations have occurred. The argument has been that where there is smoke, there must be fire.
The following examples of nonsexual boundary violations are taken from consultations with a series of over 125 clinician-patient sexual abuse cases: the patient runs seemingly inconsequential errands for the doctor which include mailing letters, returning library books, picking up the clinician’s laundry and groceries, doing household chores, and picking up the doctor’s children after school. In other cases, patients have been engaged to do the doctor’s professional billing, secretarial work, and to write academic papers for publication under the doctor’s name.
The list of inappropriate behaviors include going to lunch, dinner, movies, or family parties with patients and escorting patient to professional functions. In one case, a patient was made head secretary for the Department of Psychiatry which her doctor chaired. Dubious financial transactions have included charging purchases on a patient’s credit card, selling a boat to the patient, borrowing money from a patient’s retirement funds, and using “insider” investment information gained from the patient.
Sexual exploitation of patients does not usually begin de novo. The usual pattern is a gradual blurring of boundaries. In a typical scenario, the action begins with small and seemingly inconsequential behaviors such as extending sessions beyond scheduled time, excessive telephone conversations, or inappropriate self-disclosure leading to role reversal where the sessions begin to serve as “therapy for the therapist.” Perhaps gifts are exchanged or the doctor begins to direct work or life choices for the patient. In an early court case of “undue familiarity,” the patient was invited to the therapist’s home and asked to participate in parties, skating, and nude swimming. The judge ruled that the behavior would have been equally egregious if it had consisted of ballroom dancing and swimming with suits on.8
Harms to the patient, although common and (both clinically and legally) reasonably foreseeable, are not universal and vary in severity depending on the patient as well as the nature of the offense. The damage does not necessarily take the form of dramatic personal trauma. That happens in some cases of sexual transgression, but in a larger set of cases, sexual or nonsexual, there is first an undermining of treatment. If the relationship is sufficiently compromised by inappropriate personal and social communications, meetings, gifts, and the like, the doctor will no longer be able to serve in a professional role and minister to the patient’s health. These are substantial losses. Even where transgressions are nonsexual and relatively innocuous, they set up a “special” role for the patient. The “special” relationship may be initially gratifying, feeding the patient’s self-esteem. However, when it accelerates and finally crashes, the “cessation trauma”9,10 can be devastating.
Patients may not immediately recognize when they have been abused and exploited, especially when there has been no overt sexual relationship. When there has been a sexual relationship, the effects have been well documented.11 The patient is caught in a deep ambivalence toward the doctor, wishing to escape from a destructive process but fearing separation. The patient’s guilt and sense of having caused the doctor’s behavior is similar to that experienced by incest victims. Often, patients develop a sense of pervasive isolation and feelings of emptiness and confusion which affect their ability to relate to others. Their ability to trust is severely compromised.
Victims fluctuate between rage and despair but are often unable to talk about the source of their rage. Suicidal risk increases and cognitive deficits such as inattention, intrusive thoughts, flashbacks, and nightmares are common. Simple daily tasks may seem impossible to perform. A variety of diagnoses afflict these victims, including posttraumatic stress disorder, major depression, anxiety states and panic disorder, eating disorders, and drug and alcohol abuse. Also, personality disorders which had lain dormant may become aggravated. It is abundantly clear that no inappropriate bodily violation can be called “treatment.”
The clinician who has committed serious boundary violations faces real trouble in terms of liability. The professional damage often falls under five major areas of concern:
(1) Civil lawsuits may be filed against the doctor and the institution in which the boundary violations occurred. Allegations may include negligent hiring, negligent supervision, and negligent retention;
(2) Complaints may be registered with the clinician’s licensing board, resulting in suspension or revocation of professional license;
(3) In many states there are criminal penalties for sexual boundary violations;
(4) The clinician’s professional society may sanction him for ethics violations;
(5) The lingering pain of bad judgment and the loss of professional respect endures long beyond the incidents and sanctions themselves.
How can physicians help themselves and their colleagues manage the boundary risks inherent in interactions with patients? Monitoring one’s own behavior and reviewing the slippages with colleagues and supervisors is an ongoing task which is essential for treating clinicians. The following self-examination indices are behaviors which should raise red flags and trigger self-reflection:12,13
• Recurrent erotic feelings about patients;
• Being lax about late-fee payment;
• Attempting to impress a patient;
• Tardiness in starting appointments;
• Allowing appointments to run overtime;
• Gossiping about a patient with colleagues;
• Frequent drowsiness during appoint-ments
• Encouraging overdependence;
• Excessive fear about a patient’s leaving treatment;
• Asking patients to do favors;
• A need to argue a point with a patient;
• Unnecessary or intrusive self-disclosures;
• Having appointments outside the office.
Epstein and Simon’s Exploitation Index13 also contains valuable questions that clinicians can ask themselves in order to forewarn against incipient boundary violations. The index includes searching questions such as:
• Do you find the chronic silence or tardiness of a patient a satisfying way of getting paid for doing
• Do you accept gifts or bequests from patients?
• Do you disclose sensational aspects of your patient’s life to others?
• Do you recommend treatment procedures or referrals that you do not believe to be necessarily in your patient’s best interests, but that may be to your direct or indirect financial benefit?
• Do you find yourself trying to influence your patients to support political causes or positions in which you have a personal interest?
• Do you find it painfully difficult to agree to a patient’s desire to cut down on the frequency of appointments, or to work on termination?
Similar self-examination questions can be found in Pilette, Berck, and Achber’s Nursing Boundary Index.14
The best prevention against boundary violations are (1) having a satisfying personal life and (2) remaining in role and at task. Anyone who works with others in a helping capacity needs to be able to distinguish between normal human feelings and unethical acts. Are you ever attracted to clients? Of course you are. The problem arises when one either acts on the feelings or allows too much energy to be drained by the feelings so that professional effectiveness declines.
Responses to Provocation
Provocative gestures, overtures, and leading questions from patients require a response. First, acknowledge your feelings to yourself. If needed, get supervision aimed at helping to understand your feelings evoked by the provocation. Confide in your peers, supervisor, or a professional consultant. One can deal with these feelings with help from colleagues or a therapist, learn to set them aside, and formulate strategies to deal with the provocative behavior. Such confidences to peers effectively remove the option to act out one’s impulses. Establish a record of your having addressed the situation in an ethical, professional manner. Above all, never worry alone. Do not make the client’s problems your own; overidentification impairs judgment. Do not make your problems the clients’ problems; they have all they need. Do not share your sexual interests with patients. Nothing should be done when alone with a patient that could not be reported to a colleague or supervisor.
Seductive patients are testing limits. Recognize the line between compassion and passion. Set limits but do not be rejecting. Do express nonsexual caring, the basis of empathy. The patient should not have to act out to obtain warmth and caring, but one need not answer personal questions. Become adept with boundary-keeping responses. For example the patient’s question, “Are you married?” may be answered with a response such as, “What if I am and what if I am not?” Explore whether the question is related to treatment or an invasion of privacy. Clarify and emphasize that a personal relationship cannot be negotiated. “What is going on here?” and “That is not what we are here for” can be appropriate responses.
Try not to refer out—the patient does not need abandonment added to other problems. When the trainee or doctor says, “I am falling in love with the client” rather than referring the patient out, the supervisor should explore what is interfering with professional performance. Where referring the patient out is inevitable, accept responsibility for being unable to provide support and allow time to process the termination. With regard to follow-up care of former clients, the client should be able to return to a professional relationship.
Trainees, staff, and patients all need education about maintaining boundaries. Staff meetings that focus on personal feelings and issues of boundary violations combat isolation. An open sharing of feelings about patients among staff is helpful. Personal feelings about patients are normal and inevitable and need not be repressed, denied, or explained away. Staff must learn to experience, identify, tolerate, and use their responsive feelings toward patients.
Educating patients is an important step that should not be forgotten. Patients need to know that strong feelings are likely to develop toward clinical staff, that these feelings are normal and natural, and that discussing them is likely to be helpful. Teaching about the limits of acceptable behavior can be done by example, like responding to patient invitations or gifts with, “Let us discuss it.”
Supervisors must recognize that when staff is having difficulty or not having needs met, guidance is essential. Reaching out for help is a healthy sign, and staff seeking supervision should be praised, not pathologized. Clinical supervisors should be good role models who offer examples of appropriate behavior. Supervision should not just take place during a crisis when there is a suspicion of wrongdoing, but on an ongoing basis. Everyday vigilance keeps individuals and organizations out of trouble.
To come full circle, recall that a boundary is a protective, containing structure that enables clinical work. Boundary crossings may be beneficial if dealt with properly. Boundary violations are destructive because they do direct harm to patients, and they divert and prevent good treatment. In Frost’s poem “Mending Wall,”4 he and his neighbor go out and pile stones to build a wall. In our work, the maintenance of boundaries is a one-person job—it is always and solely the responsibility of the professional. Still, the ultimate principle is clear: good boundaries make good treatment. PP
1. Gutheil TG, Gabbard GO. The concept of boundaries in clinical practice. Am J Psychiatry. 1993;150:188-196.
2. Gutheil TG, Gabbard GO. Misuses and misunderstandings of boundary theory in clinical and regulatory settings. Am J Psychiatry. 1998:155;409-414.
3. Robert Frost. Complete Poems of Robert Frost. New York, NY: Henry Holt Co.; 1949:47.
4. Simon RI. Sexual exploitation of patients: how it begins before it happens. Psychiatric Annals. 1989;19:104-112.
5. Gutheil TG. Borderline personality disorder, boundary violations and patient-therapist sex: medicolegal pitfalls. Am J Psychiatry. 1989;146:511-602.
6. Psychopathology Committee of the Group for the Advancement of Psychiatry. Reexamination of therapist self-disclosure. Psychiatr Serv. 2001;52:1489-1493.
7. Strasburger LH, Jorgenson L, Sutherland P. The prevention of psychotherapist sexual misconduct: avoiding the slippery slope. Am J Psychotherapy. 1992;46:544-555.
8. Landau v. Werner. 105 Sol. J. 527 (Cal 1961).
9. Gutheil TG, Gabbard GO. Obstacles to the dynamic understanding of therapist-patient sexual relations. Am J Psychotherapy. 1992;46:515-667.
10. Gutheil TG. Patients involved in sexual misconduct with therapists: is a victim profile possible? Psychiatric Annals. 1991;21:661-667.
11. Epstein RS. Keeping Boundaries. Washington DC: American Psychiatric Press; 1994.
12. Menninger K, Holzman P. Theory of Psychoanalytic Technique. 2nd ed. New York, NY: Basic Books; 1973.
13. Epstein RS, Simon RI. The Exploitation Index: An early warning indicator of boundary violations in psychotherapy. Bull Menninger Clin. 1990;54:450-465.
14. Pilette PC, Berck CB, Achber LC. Therapeutic Management of Helping Boundaries. J Psychosoc Nurs Ment Health Serv. 1995;33:40-47.