AMA Journal of Ethics^®

Case

For the first presentation of his radiology elective, Scott was to select a film and discuss what he saw with the other students, residents, and faculty. He selected a chest film of a 57-year-old man, Mr. Walters, who had come to the Emergency Department several days before with a cough.

Scott decided the film was consistent with a diagnosis of chronic obstructive pulmonary disease. The next time he saw the attending radiologist, Dr. Carlson, Scott asked him to look at the film with him to see whether he concurred. Dr. Carlson began working through the systematic reading of the film with Scott—soft tissue, then bones. Then he pointed to some spots on a lower left rib and asked Scott, "What are those?" "They look like mets," Scott answered, and Dr. Carlson agreed they could be.

Scott pulled Mr. Walters' chart to verify his diagnosis and see whether the bony lesions were noted in the report. The radiologist's report said only: No acute pulmonary disease. Scott read the full chart entry. Mr. Walters had been in town visiting his daughter when he developed the cough. Because of his chronic lung condition, she had prevailed upon him to go the emergency room. The physician ordered the chest X-ray to rule out pneumonia. The discharge note said that, given his underlying lung disease, Mr. Walters saw his pulmonologist every 6 months. The chart said nothing about a chest X-ray that revealed possible bony metastases.

Scott asked Dr. Carlson whether they should call the radiologist who had read the film. "Hell, no. This guy probably knows all about his cancer. He came in to find out whether he had pneumonia and we told him he didn't. The discharge note says he'll be seeing his pulmonologist soon. I'm sure he's getting proper care."

Scott wasn't satisfied. Knowing that the examining radiologist notes all pertinent positive findings, Scott thought it possible that no one yet knew that Mr. Walters possibly had metastatic processes in his rib. Mr. Walters had gone home, and Dr. Carlson had said he couldn't be calling around the country to the docs of everyone who came into the ER to see what they knew and didn't know about that patient's overall health. That made sense. And certainly Scott couldn't take it upon himself to find out who Mr. Walter's physician was and call him. The guy would think he was nuts.

Scott mentioned the case to his wife that night, mostly to let her know about his diagnostic "catch." But Becky's response was all about Mr. Walters, and she didn't see things the way Dr. Carlson did.

"It's okay for you to practice on those patients," Becky said, "to peer and poke at them and talk about them. All for the good of medical education and the benefit of society. Now here comes a case where somebody might benefit right here and now from the fact that a student and a second radiologist took a look at his X-ray. That could only happen in a teaching hospital. You have to do something, Scott."

Scott didn't care for any of his options. Dr. Carlson had been clear that no follow up was necessary. He and Scott had no patient-physician relationship with Mr. Walters, this was not an emergency, and so on. If Scott went back to Dr. Carlson and received the same reply, that would have to be the end of it. Dr. Carlson was there to assess all of Scott's performance including his ability to follow instructions. Yet Scott was uncomfortable taking no action in Mr. Walter's behalf.

Commentary 1

This is a case that involves the ethical question of duty to warn. The core conflict, however, is less about duty to warn than about how one should handle disagreements with superiors, in this case a disagreement between a medical student on a radiology clerkship and his attending. The relevant medical history is of a 57-year-old man with a history of a chronic lung condition who presents to the emergency department (ED) with a cough. The ED attendings want to rule out pneumonia, so they order a chest X-ray that comes back negative. The student, Scott, has not met the patient, Mr. Walters, but chose his X-ray as one to interpret and present to his attending.

In doing the presentation, Scott and the attending observe what appear to be bony metastases, a diagnosis that was not made by the original radiologist or mentioned in the ED chart, either as having been reported from the patient during medical history taking or told to the patient as a present finding. Scott wants to follow up with Mr. Walters, but the attending believes there is no duty to warn because the patient probably already knows of the cancer, and, if he doesn't, someone else will probably tell him. The attending also notes that neither he nor Scott has a patient-physician relationship with the patient, implying either that there is no obligation to warn or that it may be inappropriate to warn. Scott is justifiably frustrated by this answer.

Why Warn at All?

The first issue of concern here is the possibility of a preventable medical error occurring, namely that a patient could have a diagnosed disease and not be aware of it. In the multi-factorial process by which errors occur, it is not hard to see where this one may have started. In the ED, the chest X-ray was ordered to rule out pneumonia, not to screen for bone cancer, and it seems possible that both the radiologist and the ED could miss the bone cancer because of their focus on the question of pneumonia; a classic example of tunnel vision. Because documentation is stressed so heavily for legal reasons, it is hard to believe that a possible cancer diagnosis was discussed and not documented.

The central ethical question in this case pertains to the duty to warn. The attending presumably agrees that the patient has a right to know about his cancer diagnosis, but also believes that because (a) the patient probably already knows of the diagnosis, and (b) there is no established relationship with the patient, that there is no duty to warn. I disagree with this reasoning because I see little harm in telling the patient that he may have cancer, whereas I can see great potential harm in not telling him. I believe there is a duty to warn irrespective of whose patient Mr. Walters is. If the patient already knows he has cancer, I doubt he will be upset at having someone go to the trouble of contacting him to make sure he knows. In fact, I think it more likely that he will be appreciative, knowing that someone was paying such attention to his health, an action based on the ethical principle of beneficence. However, if he does not know of his cancer, withholding that information could delay potentially beneficial treatment until someone else makes the diagnosis. This is essentially contradictory to the ethical principal of non-maleficence (do no harm) because withholding the diagnosis could do harm if it delays treatment.

The only potential harm that I can see for the patient is if the diagnosis is false. If there is a high rate of false positives in X-ray diagnosis of bone cancer, it may be prudent for Scott to try to contact Mr. Walters' pulmonologist or another physician to try to confirm the diagnosis instead of speaking directly to the patient. If the concern is of a breach of confidentiality in the student viewing the X-ray, I would think that the informed consent provided when entering an academic medical center would be sufficient to justify the student's participation in this way.

In regard to whether a duty to warn actually exists in this case, I think it does. According to Opinion 8.12, "Patient Information" of the AMA's Code of Medical Ethics, "Patients have a right to know their past and present medical status and to be free of any mistaken beliefs concerning their conditions." Moreover, as to whether it matters that the parties involved were on the patient's treatment team, Opinion 8.12 continues, physicians' "ethical responsibility includes informing patients of changes in their diagnoses resulting from retrospective review of test results or any other information."

How Do You Warn?

Accepting that there is an ethical duty to warn, the second question in this case, and the one that is the basis for my recommendation, is how might the student proceed in contacting the patient, given that the attending has instructed him not to do so. Learning how to handle these situations is particularly important because subordinates may frequently be hesitant to question their superiors' judgment. Furthermore, as I stated above, I believe not telling the patient is a potential medical error. It is my belief that we will only stem the epidemic of such medical errors if we maximize a cooperative team approach to medicine.

As I see it, the student has the following options, starting with the least desirable. First, the student could accept his attending's instructions, with the risk that the patient delays potentially beneficial therapy. Second, the student could attempt to contact the patient's physician without the attending's knowledge. This could jeopardize the student's performance review if the attending discovered that his instructions had been overruled. Third, the student could discuss this matter with another, perhaps more senior, attending, or even request an ethics consult. Some students may fear that if they take action contrary to the attending's instructions, the attending will be angered and might punish the student with a bad evaluation. A member of the ethics team at my medical school said that an attending who would take such punitive action has no business at an academic medical center. While this may be the case, I sympathize with this student's concerns, and suggest a fourth course of action that is really the basis for what we all must do to work effectively on the same team—learn to communicate in a professional manner, respect that we may disagree, and strive to keep the patient's best interests our primary objective.

My recommendation, therefore, is that the student politely tell the attending that he is unconvinced that the patient has been notified of his potential cancer diagnosis, and is not comfortable with assuming that someone else will make or has made the diagnosis. I would also tell the attending that, as a medical student, I have the luxury of the time to follow up with the patient or the patient's physician. I would explain that patients in academic medical centers accept the inconveniences of being "taught upon" by medical students, residents, and other health professionals in training, in part because of the better care and comprehensiveness that such teams provide. This a key example. If possible, I would communicate this to the attending by e-mail, because it removes the confrontational nature and gives the attending time to think about a response instead of just reacting to perceived second guessing by the student. In certain time-sensitive circumstances, the student might have to make this confrontation in person. If this information is not well received by the attending, the student might have to return to the possibility of calling an ethics consultation. The worst case scenario is that the student receives a less than favorable evaluation, but I would rather defend a poor evaluation then an avoidable poor patient outcome.

For a medical student and young physician, it is a vital skill to learn how to communicate and make suggestions to one's attendings without being threatening or questioning their seniority and experience. Of course, the student must know when to this, (eg, for a patient's benefit, not merely to point out the attending's having mistaken the embryological origin of the ligamentum arteriosum) otherwise he or she will quickly become the least appreciated student on the team. However, the notion that attendings are infallible and always right, is not consistent with the teamwork environment that we must foster in modern medicine if we want to reduce the thousands of patient deaths due to medical errors each year. This is truly a lifelong skill physicians must learn, for once we, the medical students and young physicians become the senior attendings, we will need to be able to accept this feedback from our own students.