Search tips
Search criteria 


Logo of jmedethJournal of Medical EthicsVisit this articleSubmit a manuscriptReceive email alertsContact usBMJ
J Med Ethics. 2007 May; 33(5): 308–310.
PMCID: PMC2598124

“Why aren't you doing what we want?” Cultivating collegiality and communication between specialist and generalist physicians and residents


Developing residents' communication skills has been a goal of residency training programmes since the Accreditation Council for Graduate Medical Education codified it as a core competency. In this article, a case that features problematic communication between a generalist and specialist physician is drawn upon, and it is suggested how their communication might become open and effective through a practice of reason exchange. This is a practice of giving reasons, listening to reasons given by others, evaluating reasons and deciding which particulars of situations constitute reasons to act and reasons how to act. Drawing on recent literature in teaching communication to radiology residents, it is proposed that practices of reason exchange are part of the skill set generally referred to as “negotiation skills” that should be cultivated in all residents. Particularly, in cases in which generalist and specialist physicians disagree about the reasons to do something, not do something or do something this way or that way, how well physicians are trained to practice reason exchange depends on whether they can communicate effectively and negotiate disagreement collegially.

A recent article1 offers a model for improving communication training, which was adapted for use in meeting Accreditation Council for Graduate Medical Education requirements in radiology residency programmes. The authors stated their hope to expand their course to include training in negotiation, and they mentioned specific settings for negotiation that were particularly interesting to radiology residents: job negotiations and advocating for change in vacation planning with a residency programme director.

In the article, a list of questions presented to trainees in the pre‐assessment phase of their communication training was displayed. The content of these questions recommend that negotiation skills might be far too narrowly applied only in the contexts of landing a job or improving one's vacation schedule. Several of these questions, particularly those about handling hostile, passive–aggressive, impolite or uncollegial behaviours, suggest that negotiation skills apply much more broadly. Indeed, the skills comprised by the title “negotiation skills” assume central roles in practitioners' success in maintaining good clinical comportment and in communicating effectively with colleagues.

A problem of communication in practice

As mentioned, one of these skills can be described as reason exchange. When people engage in a process of giving reasons, listening to reasons given by others, evaluating reasons and deciding which particulars of situations constitute reasons to act and reasons how to act, they are engaged in a practice of reason exchange. Particularly in cases in which clinicians disagree about whether a procedure is appropriate for a patient, how well clinicians are trained in practices of reason exchange depends on whether they succeed in conducting themselves collegially. Consider the following case, which illustrates a breakdown in communication about whether a procedure should be carried out for (and to) a patient.

Mr A, a 51‐year‐old man, arrives in the emergency department short of breath. He does not have a fever, he is breathing at a normal rate, but he has a few rattles in his lower chest. He has recently come back from a long trip in his car. Dr B, the physician caring for Mr A, suspects that Mr A might have a pulmonary embolus, so he consults Dr Rad, the radiology resident, and requests a CT angiogram (CTA). Dr Rad requests that a D‐dimer be performed first, to (non‐invasively) rule out abnormal or excessive clotting as the cause of Mr A's symptoms. Dr B responds, “Well, we don't typically use D‐dimer.”

“But,” adds Dr Rad, “if a D‐dimer is normal, there's no need to do the CTA. Furthermore, a normal D‐dimer would eliminate any good reason there might be to expose Mr A to the risk of receiving the contrast agent I'll need to use to do the CTA.”

Dr B says, “Look, you're making this quite difficult. We're trying to be thorough; we want you to do a CTA.”

In this case, Dr Rad seems to be seen by Dr B as an obstructionist and difficult when she questions whether a CTA is appropriate and necessary for Mr A. One account of why communication between Dr B and Dr Rad in this case is unproductive considers Dr B's expectations of Dr Rad. That is, Dr B seems to expect that Dr Rad should simply provide what she asks for. If Dr B views Dr Rad more as a functionary who executes orders, rather than as a specialist whose expertise is consulted and whose role is to advise and educate about how radiological diagnostic tools should be applied in the care of patients, then Dr B misunderstands what exactly it is that she's doing when she requests a radiology consultation. But, communication tension between Dr B and Dr Rad might be more fully explained by considering some features of how generalists and specialists, radiologists in particular, gather knowledge about what is going on with patients.

First‐hand knowledge versus second‐hand reports

Generalists, like Dr B in the above case, typically interact intimately with patients, and acquire patients' actual reports of symptoms and other experiences of illness. Invasive radiologists and endoscopists are specialist consultants who also interact with patients first‐hand, in intensely focused ways, to perform particular procedures. But, other specialists, such as pathologists, and other radiologists (diagnostic radiologists such as Dr Rad in the above case, for example) most often do not interact with patients first‐hand. This is an important feature of Dr Rad's relationship with the patient and her role in his care.

From what we know about this case, Dr Rad's knowledge of the patient comes only from Dr B's report. So, we must ask, what is Dr Rad's relationship with Dr B? Do they know each other? Do they trust each other? Trust can play a central role in diagnostic radiology consultation and in effective communication. That is, relative to generalists who have first‐hand knowledge of patients, diagnostic radiologists are often in positions of having to rely on generalists' reports and representations of what is going on with patients. Diagnostic radiologists can examine patients themselves and obtain first‐hand knowledge of what is going on with patients in some cases, but often, generalists' reports and representations of what's going on with patients depends on whether, when and how diagnostic radiologists respond to patients when they are consulted.

For example, suppose Dr Rad feels pressured to perform the CTA after Dr B accuses her of being uncooperative. Might she reasonably suspect that Dr B has manipulated the facts of Mr A's case to create a false sense of urgency for the CTA? Furthermore, if in Dr Rad's experience, she has come to think that the urgency for diagnostic radiology procedures are frequently or typically exaggerated by consulting physicians like Dr B, she might think that such a suspicion is justified. Regardless of whether it is justified, the mere presence of suspicion suggests that the professional relationship between Dr B and Dr Rad is damaged and their communication is ineffective. Hence, how can their relationship be repaired and strengthened such that they can communicate collegially and assertain what is appropriate treatment for Mr A?

Reason exchange and perception in practice

What counts as a reason, and when and why something counts as a reason, are things about which reasonable and well‐trained colleagues will regularly disagree. Negotiating such disagreement collegially is a skill that can build trust and shore‐up credibility among colleagues; it is indeed a hard‐won, but most valuable skill. One of these skills of negotiation is reason exchange; when people engage in processes of giving reasons, listening to reasons given by others, evaluating reasons and deciding which particulars of situations constitute reasons to act and reasons how to act, they are engaged in practices of reason exchange.

Before Dr B or Dr Rad can practice reason exchange, they both need to understand what they see as a reason for acting and why they see particular reasons prominently. Hence, it is worthwhile to first fully consider the nature of reason perception. One way to characterise how reasons are perceived is this: when a person recognises particulars in a situation, she discerns patterns of salience among them, and construe a reason to act. Discerning patterns of salience among particulars is like assessing a landscape.2 Some particulars configure hills, valleys or streams, for example. Assessing the landscape thoroughly requires discernment of such features of the landscape, however, and appreciation of relationships among them, such as distances, depths and textures; thorough assessment of particulars' patterns of relevance requires comparisons among their dimensions and proportions. Furthermore, a person's perception of particulars and construal of reasons are guided by what he cares about and, of course, what he's have been trained to care about. When a person sees something she cares about as somehow at stake in a situation, that thing acquires salience and prominence; it “protrudes” in the landscape of that situation. In addition to guiding a person's perception of particulars that configure reasons to act, what he cares about guides how he sees an action worth doing.

Hence, in the above case, we might consider Dr B's accusation that Dr Rad is being uncooperative from two different points of view with respect to how both doctors perceive reasons. First, Dr Rad might indeed be behaving uncooperatively or, second, Dr B might be perceiving the landscape of Mr A's clinical situation differently from Dr Rad. If it is the case that the two colleagues see the situation differently, they will see different reasons that there are for acting: different reasons for doing something, different reasons for not doing something or different reasons for doing it in different ways. A lack of agreement about the reasons there are might prompt Dr B to wonder, “Why isn't she doing what we want? Doesn't she see this is important? Why isn't she cooperating?”

One of the goals of reason exchange is to try to forge consensus about the reasons that there are for acting in a clinical situation. The project of forging consensus requires the negotiation of differences in Dr B's and Dr Rad's perceptions of what constitute reasons to or to not perform a CTA . One way to negotiate their differences of perception would be to ask questions that generate discussion about how each physician sees Mr A's needs and vulnerabilities. For example, which particulars in Mr A's situation are clinically relevant? and which particulars are morally relevant as risks, possible harms and possible benefits to him? Dr B's responses to questions like these would probably suggest that what she thinks is most at stake clinically and morally is determining whether Mr A has a pulmonary embolism, but also that he prefers to use a procedure with which he is familiar and has experience. Dr Rad's response, however, might suggest that what she thinks is most at stake clinically and morally is that Mr A is not exposed to any risks she does not think are necessary. Both doctors' responses to these questions can reveal that both of them have the goal of being thorough about deciding an appropriate course of action for Mr A's care, but that they define thoroughness differently, given their background, training and experience (or lack of experience) with D‐dimer and CTA as diagnostic tools.

Additionally, Dr B and Dr Rad could also facilitate better understanding of how each construes his or her obligation to respond to Mr A's needs and vulnerabilities by responding to this question: to which particulars do I feel most motivated to respond? Further discussion could clarify how Dr B and Dr Rad are obliged to respond to one another: which particulars configure reasons to respond to my colleague's worries about Mr A? Dr B's responses to questions like these might suggest to Dr Rad that Dr B is significantly motivated to have her perform a CTA because he is unfamiliar with D‐dimer. Such a thread of conversation might give Dr Rad an opportunity to recognise this and try to familiarise Dr B with the appropriate scope of use of D‐dimer and CTA as diagnostic tests. As a result of having pursued these questions and having exchanged responses, Dr B might become more comfortable with both these diagnostic tools as she cares for future patients in similar situations.

I have tried to show that if Dr B and Dr Rad can articulate responses to the questions I have listed, they can begin to understand what the other doctor sees at stake in Mr A's situation. Together, and in their responses, they exchange reasons, open avenues for subsequent questioning and clarification, and create opportunities for each to propose modifications to plans for what should be done for a patient. The practice of reason exchange cultivates a narrative that adumbrates possible consequences and enables the canvassing and consideration of possible justifications for doing something not doing something, or doing something in different ways. When reasons and the patterns of perception that illuminate those reasons are rendered explicit through conversation and questioning, they can be identified, evaluated, problematised and challenged. Then, communication between colleagues—consulting and consultant doctors—can become clearer, more open and more collegial.


Thanks to Drs Martin Goldman and Richard O'Brien for their assistance with the clinical details of this case.


iThanks to an anonymous reviewer, whose suggestions helped me clarify these distinctions among different specialist consultant clinicians.

Competing interests: None.


1. Lang E V, Sood A, Anderson B. et al Interpersonal and communication skills training for radiology trainees using a rotating peer supervision model (microteaching). Acad Radiol 2005. 12901–908.908 [PubMed]
2. Dancy J. Moral reasons. Oxford: Blackwell, 1993

Articles from Journal of Medical Ethics are provided here courtesy of BMJ Publishing Group