AMA Journal of Ethics®

Illuminating the art of medicine

Journal of Ethics Header

AMA Journal of Ethics®

Illuminating the art of medicine

Virtual Mentor. June 2005, Volume 7, Number 6.

Clinical Cases

  • Print
  • |
  • View PDF

Doctor and Friend

When a dual relationship exists between a physician and a patient, the physician has an obligation to put the patient's health first even at the risk of ending the personal relationship.

Commentary by James Hallenback, MD

Dr. Cleveland has been treating Mr. Neezer for 20 years, and they've been fishing buddies for at least 15. Two years ago Mr. Neezer began consistently complaining about lower back pain. Initially Dr. Cleveland tried to treat it with muscle relaxants and referred Mr. Neezer to a physical therapist. Mr. Neezer went the first time, but failed to show up for the second appointment. When Dr. Cleveland asked him about it, Mr. Neezer just said he wasn't into "that physical therapy thing."

"And besides," he said, "Medicare won't cover all of it."

As the back pain continued, Dr. Cleveland noticed that Mr. Neezer moved more stiffly and had particular trouble getting onto and off the exam table. He began including a prescription analgesic along with the muscle relaxants. For the last several months, Mr. Neezer has been making appointments every 6-8 weeks. He consistently asks Dr. Cleveland "What're we going to do about this pain?" and requests stronger pain control, while refusing to schedule the surgery consult that Dr. Cleveland has recommended.

"Look, Doug, with you as my doctor I don't need to go see some surgeon, you're doing a great job taking care of me."


by James Hallenbeck, MD

This case raises 2 ethical issues, both involving patient-physician relationships. One might first ask, "How should the patient's refusal of recommended care affect the provision of care by the physician?" The second issue relates to the dual relationship shared by these individuals, which is both professional and personal. In this case these issues overlap to create a serious problem.

At the simplest level, competent patients have a clear right to refuse any medical therapy, based on the ethical principle of respect for autonomy [1]. Legally, within the United States this right is based on battery statutes that guarantee freedom from unwanted touching [2]. So there is no question but that the patient is within his rights to refuse a surgery consult. The trickier question is how the exercise of this right should affect the physician's decision making and obligations to the patient. In many cases, patient refusal is not a major problem; acceptance or refusal of recommended therapy is well within a range of reasonable choices with minimal implications for care. Sometimes, however, patient refusal (or less direct noncompliance) can have more serious implications. In such situations, it is recommended at a minimum that the physician approach the problem as a matter of informed consent [3,4]. While informed consent is too often narrowly defined in terms of procedures or therapies the physician wishes to do to the patient, a broader interpretation suggests a professional obligation to inform the patient of the potential consequences of any action by either the physician or the patient that are important to the health of the patient [5]. Thus, for example, if a patient has a solitary lung nodule suspicious for cancer, and a biopsy is suggested and refused by the patient, the physician has an obligation to present possible benefits of the patient's choice not to have a biopsy (eg, avoiding possible complications and costs associated with the biopsy of a possibly benign lesion), burdens or risks of not having the biopsy (if the nodule is a curable cancer, this opportunity for cure might be missed, resulting in a terminal illness), and possible alternatives (serial chest x-rays or sputum cytologies).

Refusal of care may also have significant implications for decisions by the physician. While competent patients have the right to refuse any therapy, this does not translate into a right to receive any therapy they wish. In this case what should the physician do about the request for stronger pain medications in light of the patient's refusal to see the surgeon? While not explicitly stated, the wording of the case suggests that the physician is being pressured to prescribe opioids in a situation where they would not be appropriate—especially given the patient's refusal to consider other diagnostic and therapeutic options. Would the prescription of opioids be within the bounds of reasonable practice? It is impossible to say from this brief vignette, although there are warning flags that this might not be appropriate.

What about the dual relationship between the doctor and patient? Dual relationships exist whenever physicians treat individuals with whom they have other, non-patient-physician relationships [6]. They vary in intensity from minor—treating a member of a common social organization such as a church or work group—to major—treating a family member. Dual relationships can even exist if and when the physician shares the same illness as the patient [7]. They are not necessarily bad; sharing a common bond can improve mutual understanding and empathy. Friendship may in fact be something that patients need from physicians and can be a positive professional attribute. The risk inherent in dual relationships, however, is that objectivity can become blurred by emotions or extraneous concerns—financial interests, for example, or one's status within a group or on the job. It is too simplistic to state that the relationship should not exist; the question, rather, is how does one best guard against a dual relationship resulting in harm?

I suspect that the dual relationship between Dr. Cleveland and Mr. Neezer developed slowly over time. A particular risk in their case (and arguably in many friendships) is that a "slippery slope" may be encountered, in which "special considerations" insidiously lead from small acts of friendly kindness to requests for favors that lie outside the bounds of propriety. Each step down the slope seems reasonable enough, but, at a certain point, one realizes he is in trouble, and climbing back to safety seems impossible. I worry that this may be exactly what has happened here—unbeknownst to either the patient or physician.

How do you know when a dual relationship is on a dangerous slippery slope? I think the best safeguard against the danger is to abide by 2 principles: "the patient comes first," and "first, do no harm." The very nature of a dual relationship implies that the physician has some investment in the relationship beyond his or her professional role. This is not necessarily a problem unless that investment creates such a conflict of interest that professional judgment is compromised. Friendship may serve the patient, if the physician is motivated to "go the extra mile" and has a better understanding of the patient as a person. It is not hard to imagine, however, that the friendship might result in harm—the possibility of which is strongly suggested in this case—if interactions with the patient are driven more by the need to maintain the friendship and not offend than by professional judgment.

If the dual relationship poses a risk of harm to the patient, what should the physician do? It almost goes without saying that, when it is clear from the outset that a dual relationship poses a serious risk, professionalism requires that the physician not serve in the professional role. More difficult, as likely happened in this case, is the situation in which the professional relationship was entirely appropriate initially, but where, over time or due to changing circumstance, a potentially harmful relationship evolves. In clear-cut situations, the patient must transfer to another physician following discussion as to the reason for the referral. In borderline cases, the potential conflict of interest should be disclosed and discussed with the patient, at a minimum, and a continuation of the relationship weighed against transfer of care.

While I have addressed these 2 ethical issues—the patient's refusal of recommended treatment and the patient-friend-physician relationship—separately, they come together in terms of the communication skills needed to manage the situation. If it is clear that the professional relationship should not continue, then the major question is how best to break this news to the patient and explore the implications both for continued care (referral options to other physicians) and their friendship.

If the situation is less clear-cut and continuation of care is contemplated, then a discussion must occur regarding their relationship, and future care plans must be negotiated [8]. While the patient in the above vignette indirectly refers to their friendship status ("with you as my doctor…"), their friendship has likely remained a subtext to their clinical conversations. The positive and negative implications of this for the patient's health care must be addressed more directly. If continued care by this physician is contemplated, the physician should consider establishing certain rules regarding the overlap between their friendship and professional relationship and negotiate a mutually agreeable plan for addressing the patient's back pain [9]. If either of these attempts fails, there is little choice but to transfer the patient.

Negotiation in health care is an underappreciated art, a detailed discussion of which is outside the scope of this text [10,11]. The biggest risk in this case is that the issues in dispute will be personalized. Indeed, the patient has already done so, by dismissing consideration of the surgery consult because "Doug" is such a great doctor. Should Dr. Cleveland challenge the status quo—either their relationship or his approach to Mr. Neezer's back pain—he should not be surprised if the personalization turns negative. "Doug, I thought you were my friend! Do you think I'm some kind of drug addict?" While the physician cannot control the response of the patient, he can avoid making the same mistake of personalizing the situation. Using the language of Fisher and Ury in their book, Getting to Yes, separate the people from the problem [12]. Here, it is important to separate the people—patient and doctor—from the problem—that a conflict of interest can compromise care. Fisher and Ury also stress the importance of using objective criteria and mutual interests, rather than "positioning" in negotiating. In this case, the patient has taken the position that he does not want to go to the surgeon and he does want more painkillers. The physician could use more objective standards of care in supporting both his concerns about their dual relationship and his argument that the patient see the surgeon, based on their shared interest in maximizing good health outcomes and maintaining personal and professional relationships.

Fisher and Ury also introduce the term, BATNA (best alternative to a negotiated agreement). Prior to having the suggested discussion with the patient, the physician must be clear on his bottom line(s), his BATNAs. One bottom line might be, "I am only willing to consider a change in pain medications if you agree to see the surgeon and the surgeon concurs." Another might be, "I am agreeable to continuing as your physician, but only under the following conditions…." In establishing one's bottom line, one must be prepared for the consequences if it is not met. In this case, the friendship may be a casualty, one which the physician must be willing to sacrifice for the good of the patient, if necessary.


  1. Beauchamp T, Childress J. Principles of Biomedical Ethics. 4th ed. New York, NY: Oxford University Press; 1994.
  2. Renquist W. Vaco v Quill. In: Battin M, Rhodes R, Silver A, eds. Physician Assisted Suicide. New York, NY: Routledge; 1998:423-430.
  3. Bogardus ST Jr, Holmboe E, Jekel JF. Perils, pitfalls, and possibilities in talking about medical risk. JAMA. 1999;281:1037-1041.
  4. Braddock CH III, Edwards KA, Hasenberg NM, Laidley TL, Levinson W. Informed decision making in outpatient practice: time to get back to basics. JAMA. 1999;282:2313-2320.
  5. LeBlang TR. Informed consent and disclosure in the physician-patient relationship: expanding obligations for physicians in the United States. Med Law. 1995;14:429-444.
  6. Gabbard GO, Nadelson C. Professional boundaries in the physician-patient relationship. JAMA. 1995;273:1445-1449.
  7. Hallenbeck J. A dying patient, like me. Am Fam Physician. 2000;62:888-890.
  8. Rourke JT, Smith LF, Brown JB. Patients, friends, and relationship boundaries. Can Fam Physician. 1993;39:2557-2564.
  9. Eggly S, Tzelepis A. Relational control in difficult physician-patient encounters: negotiating treatment for pain. J Health Commun. 2001;6:323-333.
  10. Quill TE. Recognizing and adjusting to barriers in doctor-patient communication. Ann Intern Med. 1989;111:51-57.
  11. Rodning CB. Coping with ambiguity and uncertainty in patient-physician relationships: III. Negotiation. J Med Humanit. 1992;13:211-222.
  12. Fisher R, Ury W, Patton B. Getting to YES - Negotiating Without Giving In. 2nd ed. New York, NY: Penguin; 1991.

James Hallenbeck, MD, is assistant professor of medicine at Stanford University. He is board certified in internal medicine and hospice and palliative medicine. He is also the hub-site director for the Department of Veterans Affairs Interprofessional Palliative Care Fellowship Program.

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 on this site are those of the authors and do not necessarily reflect the views and policies of the AMA.