Case and Commentary
Aug 2005

Frame Feedback to Improve Professional Performance: Colleague-to-Colleague Communication.

Barbara F. Sharf, PhD
Virtual Mentor. 2005;7(8):547-550. doi: 10.1001/virtualmentor.2005.7.8.ccas3-0508.


You are a radiologist in a hospital-based group practice. While one of your partners is on vacation, an internist calls you because your partner read a head MRI for one of his patients a short while ago. The internist is concerned because his patient's headaches are becoming worse, and she is now developing visual field deficits. Before considering a second MRI, the internist would like you to review the first one. When you get off the phone, you pull the patient's (Mrs. Kirk's) MRI; there is a suspicious "spot" in the pituitary gland that you think called for further evaluation. This difference of opinion has happened twice before with radiographic studies that were originally read by your partner and reviewed by you over the last 6 months. You think he is making some obvious omissions in his readings, and you recognize that something must be done. Patient safety could be at risk. You decide that the best course of action is to talk to your partner as soon as he returns from vacation. What do you say to him and how do you say it?


There's no getting around it—this is an unpleasant interpersonal situation, and one of ethical concern for patient well-being. Typically, physician colleagues enact a professional relationship of co-equal status and power. You have no authority to monitor your partner (Dr. M), nor do you have any desire to regulate your fellow physicians. On the other hand, the practices of clinical consults and second opinions have developed to provide a system of checks and balances in patient care so that the judgment of one practitioner is open to scrutiny, questioning, and feedback from others. The hoped-for result is the best care available for the patient. This situation with your partner is functioning like a second opinion. It is not unusual for physicians to differ in their judgments, thereby triggering dialogue and reconsideration of options. It is unusual to discover a pattern of obvious clinical errors; and, in this case, you feel ethically obliged to do something about this problem before it is repeated with other patients. But what constitutes "doing something"? Calling your peer to task looms as a distasteful event.

Define goals.

The first knotty question to be considered is: what is the goal of this conversation? Put another way, what do you hope to accomplish by confronting your colleague? An overarching goal is to get Dr. M's full attention and to mutually negotiate a constructive response. Examples of constructive responses might include a promise from Dr. M to conduct a supervised review of his work and to contact treating physicians if they've been given incorrect information. You may also have concerns about your colleague. While you have no more desire to be his therapist than his police officer, it would be useful to know to what he attributes this pattern of errors; if there is a difficulty beyond carelessness (eg, incompetence, depression, substance abuse, unusual stress, etc.), then it will become important to urge Dr. M to seek help for himself.

Such goals assume 2 willing participants engaged in civil conversation. What if Dr. M is defensive, angry, in denial, or otherwise unwilling to talk? The intimidating tape of a tense phone interchange between Dr. M and Dr. Y (alias, you) keeps playing in your head:

Dr. Y: Dr. M, this is Dr. Y. I'd like to talk with you about your findings on Mrs. Kirk, a patient whose MRI we both reviewed recently.
Dr. M: OK, shoot. What's on your mind?
Dr. Y: I was hoping we might be able to talk in person about this matter. Should take about 20 minutes or so.
Dr. M: (beginning to sound annoyed.) Sorry, I'm really booked today. If you have something to say, let's hear it.
Dr. Y: (also a bit annoyed.) Well, to be honest, when I reviewed your report on Mrs. Kirk, I found a spot on her scan that you had overlooked. The internist who ordered the MRI called me to discuss further testing. And, unfortunately, this isn't the first time I've found similar errors in your work.
Dr. M: (abruptly) Fine, Dr. Y. I'll take a second look at Mrs. Kirk's MRI. But in the future, I'd appreciate your concentrating your efforts on your own work. I'm perfectly capable of overseeing my write-ups, thanks!

What steps can be taken to avoid this type of hostile exchange and to facilitate as productive an encounter as possible? There are no guarantees, of course, when it comes to dealing with individual personalities, but there are well-established guidelines for communicating feedback, including critical constructive or corrective feedback. Though these guidelines have been developed primarily in the context of educational and managerial settings,1,2 the practice of medicine has often been framed as a lifelong educational venture. Thus, it's helpful to realize that even seasoned practitioners can benefit from peer commentary and advice, if provided in a sensitive and thoughtful manner. So how can critical feedback be communicated constructively?

  • Effective feedback is timely, but not sprung on the recipient as a surprise attack.
  • It is communicated in an environment conducive to an undisturbed and private interpersonal exchange.
  • It is framed as problem-solving between allies with common goals, as opposed to one person attempting to condemn or control the actions of another.

In the preceding imagined dialogue, Dr. Y had an appropriate instinct to meet with Dr. M personally to talk about Mrs. Kirk's case; a planned face-to-face meeting would have been preferable to an off-the-cuff phone conversation, which lacks many nonverbal, relational cues (an even bigger problem in e-mail messaging). However, Dr. Y didn't realistically assess the difficulty of setting up such a meeting, especially with the opening gambit of discussing "your" (Dr. M's) findings, which does not connote conjoint problem-solving. Timeliness was out of Dr. Y's control, since Dr. Y had already been contacted by the treating physician.

  • Feedback should be conveyed in descriptive, rather than evaluative, language.
  • It should be anchored in firsthand specific data, limited to a few key points, rather than vague generalizations or overwhelming amounts of information.
  • The person offering feedback may include subjective reactions, as long as these are acknowledged as personal feelings and not statements of fact.

Even in this brief interchange, there is a good deal of evaluative language, in which Dr. Y labels Dr. M's work as error-prone. Had this meeting occurred in person, there would have been opportunity for these 2 To jointly examine the MRI (the firsthand data). Dr. Y may also have been more successful in engaging Dr. M had she expressed her personal concerns for both the well-being of Mrs. Kirk and Dr. M.

  • The recipient should have adequate opportunity to respond and contribute to problem-solving.
  • It should be focused on remediable behaviors, resulting in improved decisions and outcomes.

After having been negatively evaluated, rather than engaging in problem-solving for the benefit of the patient, Dr. M is focused on exiting this conversation—not on addressing potentially serious problems. There's been no discussion about making future positive changes.

Using these guidelines to inspire a dialogue that doesn't arouse defensiveness, let's imagine another possible exchange between Dr. Y and Dr. M:

Dr. Y: Thanks for agreeing to meet with me to discuss our differing interpretations of Mrs. Kirk's MRI. I'm glad we could find a mutually agreeable time.
Dr. M: Yeah, I appreciate the heads up. I've had a chance to compare our 2 reports. I agree there's a spot that probably bears further investigation. I've called the treating physician to discuss my revised assessment and to let him know that you and I concur.
Dr. Y: I see. Well, that relieves a large part of my concern.
Dr. M: OK.
Dr. Y: Something else is nagging at me.
Dr. M: OK, shoot.
Dr. Y: Well, I've noticed that our readings have differed in 3 of the last 4 images we've both examined. It concerns me because these differences raise questions about optimal patient care.
Dr. M: Really? What cases? Were they like what we've been discussed about Mrs. Kirk's case?
Dr. Y: Yes, roughly the same kinds of things.
Dr. M: I don't know what to say.
Dr. Y: In the past, we've agreed upon our readings quite well. I'm wondering whether something out of the ordinary is going on with you recently.
Dr. M: Well, I have to admit that, lately, I've had some major stress at home with a very sick child. I like to think of myself as a very careful reader, and I don't want to put anyone at risk.
Dr. Y: I wonder if it would help to go through the previous 2 images together and see if we can come to agreement on them as well? Also, we should discuss if there's anything I or other folks in the department can do to help while you're having this extra stress at home.
Dr. Y: That would be a great help. I appreciate your taking the time to talk to me about this.

Even though this conversation went very well, what should you do if Dr. M continues to make clinical errors? Reporting him to a quality assurance committee is always a final resort, but needn't be the first course of action. This case reveals a real and important tension between a physician's ethical obligation to ensure accuracy in patient diagnosis and treatment while providing corrective feedback to colleagues in a constructive, collaborative manner. Thoughtful communication that does not put colleagues on the defensive is most likely to result in improved performance.


  1. Ende J. Feedback in clinical medical education. JAMA. 1983;250(6):777-781.
  2. Gibb JR. Defensive communication. J Commun. 1961;11(3):141-148.


Virtual Mentor. 2005;7(8):547-550.



The people and events in this case are fictional. Resemblance to real events or to names of people, living or dead, is entirely coincidental. The viewpoints expressed in this article are those of the author(s) and do not necessarily reflect the views and policies of the AMA.