Clinical Concerns in Boundary Issues

August 1, 1999

It is relatively recent that boundary issues in clinical practice have attained professional awareness, although some early concepts portended the issue. Several decades ago, for example, the term boundary violations referred almost exclusively to what we now term sexual misconduct.

It is relatively recent that boundary issues in clinical practice have attained professional awareness, although some early concepts portended the issue. Several decades ago, for example, the term boundary violations referred almost exclusively to what we now term sexual misconduct. Within psychoanalysis existed the notion of parameters (e.g., actions, interventions or responses that lay outside the fundamental analytic terrain of verbal free association)-a term that embraced some of the boundary issues we identify today, though in slightly different conceptual terms.

Moreover, Gabbard and Lester (1998) have pointed out that the developmental notion of ego boundaries, a term relevant to the understanding of psychotic thought among other things, may have been an early expression of boundary theory; however, the sense of "where 'you' leave off and 'I' begin" has a more intrapersonal focus compared to the dyadic context in which we now think of boundary issues (Simon, 1989).

As the mental health professions expanded their knowledge and understanding of trauma and its effects, victims of therapist sexual misconduct began to come forward in increasing numbers. Clinicians who assessed, treated or testified for such patients began to observe that sexual relations between these patients and their therapists were almost inevitably preceded by, accompanied by and surrounded by a nimbus of behaviors that constituted either encroachment on patients' personal space in some way or transgressions by the professional of some stated or unstated limits of professional conduct.

Since emergence of these early nebulous perceptions, a more systematic vision of boundary theory has emerged-if only in the last decade (Epstein and Simon, 1990; Gutheil, 1989; Gutheil and Gabbard, 1998, 1993; Simon, 1992, 1989). Since problems with boundaries still seem to bedevil our field and our patients, the time may be ripe for a review and update of this topic.

Some Definitions

What is a boundary? While the term is easier to intuit than to explicate, a boundary is the edge of appropriate behavior in a given situation (Gutheil and Gabbard, 1993). But modern boundary theory goes beyond behavior to matters of language, time, place and space, money, self-disclosure, and receptivity to input from outside the dyad.

Therapists breach boundaries in one of two major ways (Gutheil and Gabbard, 1993): boundary crossings and boundary violations. In a boundary crossing, the therapist steps out of the usual framework in some way, but this action neither exploits nor harms the patient; indeed, it may advance the therapeutic alliance or the therapy itself. Examples that clearly fit this description include offering a crying patient a tissue; helping up a patient who has fallen; disclosing some facts about oneself (e.g., that one is a psychiatrist rather than a psychologist); and, under some circumstances, giving a patient a ride to safety in a blizzard.

None of these actions would constitute ordinary psychotherapy; moreover, the examples involve physical action, physical contact, self-disclosure and extra-office encounters, respectively-all issues that could, with slight alterations, represent boundary breaches. However, none of these particular events, as described, harms or exploits the patient.

Boundary violations, in contrast, are defined by their exploitative or other harmful effects on the patient. In a boundary violation, the therapist inappropriately takes advantage of the dynamics and potential dynamics that are present naturally in the therapy situation: transference, intimacy, dependency, idealization, rapport, empathy and the closeness one feels with a confidant. Rather than serving as essential elements of the therapeutic alliance for therapeutic goals, these natural elements of therapy are, in a boundary violation, put to the service of the therapist's gratification rather than the patient's welfare.

Note that patients, too, may breach or attempt to breach boundaries, but the responsibility for maintaining and defining boundaries always belongs to the therapist.

Robert I. Simon, M.D., and I (1995) suggested that boundary breaches may first manifest themselves in the transition zone "between the chair and the door," i.e., when the session appears to be over. If put to use as risk management, this interval may serve as an early warning system of boundary problems if carefully scrutinized by the therapist.

The Centrality of Context

This is an opportune moment to stress one of the most commonly misunderstood aspects of boundary theory: the centrality of context in assessing boundary issues in general and, most importantly, in distinguishing a boundary crossing from a boundary violation. In parallel with the examples above, giving gifts to a patient, erotic physical contact, highly personal self-disclosure and planned meetings outside the office might well constitute boundary violations, but context is always determinative. For example, a psychiatrist might give an impoverished patient a medication sample; a patient might have an unexpected erotic reaction to an inpatient physical examination, even with a chaperone present; a therapist might have to explain that a recent absence was for needed surgery; and a case manager might have to drive a patient to a welfare office.

Some Problem Areas

One point that comes up with great frequency in these discussions is the matter of children; clearly, the aforementioned issues are completely different for children. For instance, play therapy may involve rolling on the floor with the child-patient, an action hard to defend with adults. Thus, boundaries in child psychiatry are harder to define. As child psychiatry boundary issues are rarely discussed in the literature, surely this represents an area for more investigation and discussion by child psychiatrists.

Time issues frequently emerge in discussions of boundaries. Many clinicians believe that a regular appointment for a standard time is a valuable aspect of the frame of psychotherapy (Langs, 1982). However, sometimes it is necessary to extend the time of an appointment on an emergency basis. This situation can be distinguished from becoming so engrossed with a patient or a patient's narrative that the therapist loses track of time and repeatedly runs over the designated appointment time.

Trips outside the office are a common way station on the path to greater boundary problems. Can you do psychotherapy over lunch in a restaurant? If you do not talk with your mouth full, it seems theoretically possible, but 1) it is not a confidential setting (people passing by could overhear the conversation), and 2) it does not look like therapy-it looks like a date and thus is not above suspicion. On the other hand, certain behavior-therapy approaches to variants of social phobia might well involve leaving the office to confront, say, the phobic situation as the last step in a desensitization program.

Hugs are another source of confusion among mental health care professionals. In the "old days" when everyone wanted to be Leo Buscaglia, Ph.D., hugs were seen as a form of benign, occasional physical contact punctuating the surrounding therapeutic expanse of talk. Needless to say, it is different now. All hugs have become suspect because of their potential erotic content, an effect intensified among victims of trauma.

An exception might be made for HIV-positive patients because of their status in many clinical settings as "lepers," untouchable without latex armor. I have suggested that clinicians who work with HIV-positive patients should contrive to touch such patients nonsexually at some point during each therapeutic encounter, e.g., by a handshake, pat on the shoulder or squeeze of the arm, specifically to counter the social ostracism such patients may feel.

Finally, self-disclosure remains an area of controversy. A fundamental tension appears to exist between two notions. The first notion is the clinical value of the therapist's relative anonymity, even to the point of becoming the "blank screen" envisioned in early analytic writings. The second notion is the emphasis in some therapeutic circles on the value of the "real relationship" with the therapist, one component of which may be the therapist's self-revelation of personal details, real-world advice or expression of personal values.

Arguments for the first-more reticent-approach stress how information about the therapist may burden patients and constrain or even foreclose certain areas of free discussion. To use a straightforward example, a patient who is told that her therapist is Roman Catholic may find it more difficult to discuss a contemplated abortion. Opponents of this view note that all therapists are constantly disclosing information about themselves willy-nilly through channels ranging from body language and speech to choice of office decor.

Proponents of a more self-revelatory approach also suggest that the reality of the therapist is humanizing, demystifying and de-idealizing in a manner that fosters greater autonomy and less dependence in the patient with greater therapist-patient egalitarianism and less authoritarianism from the therapist.

This matter almost certainly requires decision-making on a case-by-case basis. One author (Maroda, 1991) has suggested that-in certain specific therapeutic impasses-even disclosure of countertransference dynamics may be constructive. Self-disclosure may facilitate authentic engagement in the therapeutic process in one context, but seductive, coercive or destructive intrusion on a patient in another. The therapist's sensitivity to the issue of boundary transgression and reliance on the traditional standbys of documentation and consultation provide the greatest security against the potential harmful impacts on patients and preservation of clinical effectiveness.

Finally, therapy works best when it is seen as work, gratifying for its own sake. The time-honored principles of abstinence and neutrality are still essential components of this endeavor.




Epstein RS, Simon RI (1990), The Exploitation Index: an early warning indicator of boundary violations in psychotherapy. Bull Menninger Clinic 54(4):450-465.


Gabbard GO, Lester EP (1998), Boundaries and Boundary Violations in Psychoanalysis. New York: Basic Books.


Gutheil TG (1989), Borderline personality disorder, boundary violations and patient-therapist sex: medicolegal pitfalls. Am J Psychiatry 146(5):597-602.


Gutheil TG, Gabbard GO (1998), Misuses and misunderstandings of boundary theory in clinical and regulatory settings. Am J Psychiatry 155(3):409-414.


Gutheil TG, Gabbard GO (1993), The concept of boundaries in clinical practice: theoretical and risk management dimensions. Am J Psychiatry 150(2):188-196.


Gutheil TG, Simon RI (1995), Between the chair and the door: boundary issues in the therapeutic "transition zone." Harv Rev Psychiatry 2(6):336-340.


Langs R (1982), Psychotherapy: A Basic Text. New York: Jason Aronson.


Maroda KJ (1991), The Power of Countertransference: Innovations in Analytic Technique. New York: John Wiley and Sons.


Simon RI (1992), Treatment of boundary violations: clinical, ethical and legal considerations. Bull Am Acad Psychiatry Law 20(3):269-288.


Simon RI (1989), Sexual exploitation of patients: how it begins before it happens. Psychiatric Annals 19:104-112.