Human relationships are complex, and the relationship between patient and physician is no exception. In different ways, two articles in this issue of the Journal address the nature and consequences of interpersonal expectations and social norms. In the first article, Farber et al. explore physicians' experience and response to boundary violations by patients.1 Professional boundaries constitute a particular type of social norm with legal and fiduciary ramifications. In the second article, Schmittdiel et al. examine the effect of physician gender on patient selection of physicians and satisfaction with care.2 The results of this study also touch on expectations of normative social behavior.
Farber et al. have reported the frequency and physician response to nine patient boundary violations: use of the physician's first name, personal questions, social invitations, expensive gifts, overt affection, sexually explicit language, verbal abuse, sexual contact, and physical abuse. The most serious offenses were rarely encountered; other patient behaviors were common. Physicians generally responded to serious offenses by making a note in the chart and occasionally by dismissing the patient from care. Responses to the other patient behaviors were varied. One third of physicians responded to personal questions with anger, while almost two thirds expressed annoyance at the use of their first name.
To understand this pattern of response, we wondered, what is the nature and purpose of professional boundaries? Professional boundaries have been defined as “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.”3 Boundaries are important in relationships in which one party has greater potential (perceived or real) power and acts to protect the vulnerable participant from exploitation. Indeed, protection of patients from abuses of power has raised the principal of autonomy to so vaulted a place in our society that it has become the guiding value in modern bioethics and medical law.4
Farber et al. add two concepts to the traditional definition of professional boundaries. First, they go beyond the concern of patient protection to consider professional boundaries as protection against exploitation of both the patient and the physician. Second, by broadly defining boundaries as “mutually understood rules and roles which are found in relationships,” they link professional boundaries to universally understood social norms. Each of these concepts deserves separate consideration.
First, do physicians need protection from exploitation by patients? For the most serious offenses (including physical and sexual abuse), we believe the answer is yes. Physicians are vulnerable to these behaviors and, indeed, our society affords legal recourse. It is likely that the charting and dismissal practices of physicians when encountering these events are attempts at documentation for legal protection. The answer in regard to the need for physician protection from exploitation as a result of the other patient behaviors is less obvious.
Let's consider the issue of names. Physicians generally address patients by their title and last name as a demonstration of respect and regard for patient autonomy. Some consider formal address an important psychological protection against the infantalizing effect of over-familiarity and its potential for coercion and paternalism. Physicians may wish that a patient use a similarly formal manner of address, but it is difficult to imagine that physicians are exploited if a patient chooses not to do so. Although the majority of physicians in the Farber study were annoyed by this behavior, their autonomy was not threatened by it.
How about personal questions? Although one third responded with anger, the majority of physicians were not angered when they encountered personal questions from patients. This indicates that many physicians feel comfortable allowing their patients to know them somewhat personally. It may be that the angry physicians perceived personal questions as attempts to develop a relationship that is too intimate. This may also be true of first names, social invitations, expensive gifts, and overt affection. Perhaps these behaviors represent a patient's attempt to control the relationship or diminish the high degrees of respect generally afforded to physicians. If a physician is made uncomfortable by these behaviors, he/she can communicate his/her expectations by requesting a different manner of address, redirecting the patient's question, or politely refusing the invitation. While these instances may be awkward, we do not believe they meet a standard of exploitation.
Most interesting, from the perspective of social norms and interpersonal expectations, is the pattern of gender-related responses to these patient behaviors. Female physicians reported more objectionable behavior and responded more negatively in almost every category, including patient dismissal for serious offenses and anger or annoyance at more trivial ones. The fact that women in this study experienced more boundary violations and were less tolerant of them is no surprise. Women physicians, as all women, are vulnerable to physical and sexual assault, sexual harassment, and sexual exploitation. Female physicians experience these attacks more often than do male physicians, not only from their patients but also from their professors and colleagues. Female physicians may also experience more breaches of normative social behavior and greater challenge to their professional status than their male colleagues.
In defining professional boundaries as mutually understood rules and roles found in the patient-physician relationship, Farber et al. have linked social norms to protections against exploitation. This may not always be the case. Social norms are important; they affect interpersonal expectations and behavior. Violations of social norms can be uncomfortable, undeserved, or unkind but are not necessarily exploitation.
Also in this issue of the Journal, Schmittdiel et al. similarly explore the consequences and dynamics of social norms on interpersonal expectations. Regardless of patient gender, the investigators found a bias favoring the selection of male physicians. While female patients chose female doctors three times (36%) as often as male patients (12%), the majority of both male and female patients chose a male physician. This may be because female physicians violate the social norm by which professional status is ascribed to male gender. The role demands of an authoritative, competent physician may be at odds with the stereotypical female role that implies nurturance, positive expressiveness, and equality. Physicians are generally regarded as male; women physicians may cause confusion. This confusion may be reflected in the study results.
Although not linked directly to gender preferences, communication studies in primary care have demonstrated broad conversational differences between physicians of different genders.5,6 Female physicians have longer visits and engage in more partnership building, emotionally focused talk, positive talk, and psychosocial exchange than male physicians. Patients of female physicians disclose more biomedical and psychosocial information and are more positive in their talk. The communication behaviors associated with female physicians are those generally valued by patients and predictive of positive patient outcomes, including satisfaction, recall of medical information, and compliance with medical recommendations7 as well as health status improvements.8 Nevertheless, the literature directly relating physician gender and patient satisfaction is mixed, with some studies finding higher satisfaction with female physicians9–11 and others finding the opposite or no effect.12
In our own studies, we have reported findings different from those reported here. On the basis of analysis of two independent studies, we found that both male and female patients of young female physicians reported lower ratings of satisfaction than other patients.13 An intriguing element of our finding was that these young female physicians engaged in more of the communication behaviors generally valued by patients than did their male colleagues. We speculate that other patient values and prejudices, perhaps an inferred lack of authority or expertise because of youth and gender, may offset whatever advantage the female physician might have by virtue of her communication performance. Alternatively, expectations for positive communication skills (including partnership and emotional support) may have been so high that patients were disappointed despite the superior performance of their female physicians.
In the current study, highest satisfaction ratings were evident for a select minority of males. Only 42% of males chose a physician at all, and only a very small number of this group (12%) chose a female physician. Let's think about how this very small population of males might differ from the vast majority of male patients who, if opting for a physician at all, opted for a male physician. There are two hypotheses that can be forwarded. First, this small group of male patients may hold nontraditional values that support the choice of a non-normative (female) physician. Second, these males may have chosen their female doctors for some unmeasured consideration such as personal referral.
The situation for female patients is somewhat, but not entirely, different. Half of the female patients opted to choose a doctor and, although more likely than male patients to choose a female doctor, only 36% of female patients made that choice. Female patients who choose female physicians may, like their male counterparts, hold nontraditional views about professional status and gender. However, Schmittdiel et al. found that these female patients differ from males in placing a higher value on physician communication skills. The import of this distinction may be in raised expectations for female physician performance. As noted in our own empirical studies of medical dialogue, female physicians do engage in more of the communication behaviors generally valued by patients. Nevertheless, this may not be sufficient to meet the high expectations of female patients, particularly under the tight scheduling constraints typical of managed care.
Given such variability in expectations and norms, Farber et al.'s conclusion that physicians must communicate clear expectations to patients is compelling. Patients must also communicate expectations to their physicians. This, unfortunately, is not easy to do. Physicians cannot rely on subtle, unspoken, or universally understood rules to dictate patient behavior, nor can patients expect physicians to meet unreasonable and unspecified expectations. But because norms are responsive to societal pressure, we are optimistic in believing that our society is undergoing a sea change. This is evident in broad shifts in the workforce, where female gender is more normative than in the past. Also, generational changes anticipate an increased forthrightness by which baby boomers are distinguished from their predecessors and genXers from their mothers and fathers. We can perhaps look forward to changes through which more direct expression of patients' and physicians' expectations regarding their relationship will emerge.
The article “Factors Influencing the Selection of General Internal Medicine Fellowship Programs: A National Survey” by Caiola and Litaker should have been listed as a Brief Report rather than under the section heading “Innovations in Education and Clinical Practice.”