How a Cancer Doctor Really Feels About Second Opinions

Andrew C. Neuschatz, MD


December 31, 2018

As a radiation oncologist in private practice, the question of getting a second opinion comes up fairly regularly with patients during consults. My standard answer to patients is that they shouldn't trust any doctor who is afraid of a second opinion. I'll admit, however, that the topic of second opinions is probably discussed 10 times or more for any one patient who actually goes on to get one.

The idealized second opinion rarely happens.

Second opinions can be both rewarding and frustrating for an oncologist in private practice. No one can keep up with the expanding knowledge base for every disease site, and the volume of unusual and obscure diagnoses and pathologies is surprisingly high in a busy private practice. Whether it's radiosurgery for a craniopharyngioma or mediastinal radiotherapy for an anaplastic large cell lymphoma with prior breast augmentation, there are situations in which I'm happy to hear what someone else has to say. Additionally, grading and interpretation of biopsy specimens can be subjective, and a differing conclusion can change both treatment options and prognosis.

Unfortunately, the idealized second opinion rarely happens. The patients who you'd be happy to see get a second opinion never seem to request one, even when suggested. Common things happen commonly, and so it is with second opinions as well. Node-positive colorectal cancer patients, hormone receptor–positive breast cancer patients, and other common scenarios end up dominating the second opinions in my practice, as they dominate the number of consults in most practices. Requesting a second opinion on a pathologic specimen or a second read of an MRI is easy to request and will rarely cause a significant delay. A full second opinion—one in which the patient essentially has a complete second consult, where pathology and radiology are independently reviewed, and the case is re-presented at a second tumor board—is a considerably more time-consuming endeavor. A recent study suggested that at a National Cancer Institute (NCI) cancer center, over 40% of breast cancer patients seen for a second opinion had a change in diagnosis.[1] I do not think this is consistent with the experience of a private practitioner in any good-quality practice that I know. Some of this depends on what constitutes a change and how it is presented to the patient.

I have gotten second opinions noting that their pathologist feels the margin is 2 mm rather than 4 mm, or that they'd suggest a boost dose 2-3 fractions higher than I was planning. These may sound significant to a patient but are probably more stylistic and institutional differences than anything statistically proven to be meaningful. Additionally, suggested changes may include ROS1 or ALK testing when this was attempted but there was insufficient tissue on a lung biopsy and repeat biopsy has potential toxicity. Last, second opinions often call attention to "cover-your-ass" radiology and pathology reports for which there's little to do. A 3-mm lung nodule seen on a staging CT scan in a colon cancer patient does indeed need follow-up, but it's too small to do anything about right now, and sidetracking the patient into believing that something was missed is often not helpful when initial treatments need to be completed.

Another occasionally irritating situation: Despite the patient's expectations that they will be seen by well-known senior physicians, I often find that second opinions at academic medical centers are assigned to the most junior faculty. This leaves the original physician in the sometimes less than desirable position of taking advice from a physician with considerably less experience and occasionally a lack of real-world insight. I had a patient with a limited-stage small cell lung cancer whose second opinion note derided me for not considering twice-daily radiation. A simple discussion with the patient would have easily revealed that this was infeasible for him due to transportation and distance from my facility. Other times I've seen recommendations for agents suggested by genomic profiling but not approved for that indication and thus highly unlikely to be covered by the patient's insurance. A more experienced physician doing a second opinion would probably appreciate more nuances of patient care, including performance status, expected tolerance, and insurance coverage, as well as logistics and feasibility.

All of this is to say that second opinions are potentially very useful in almost all situations, but I think many private-practice oncologists see the difference between a good second opinion and a bad one. Good second opinions should provide clarity in decision-making and confidence to the patient. Bad second opinions muddy the waters and make patients believe that their initial doctor "missed" something, thereby injuring the doctor-patient relationship. A second opinion should not be an academic exercise in maybes and what ifs but rather a real-world recommendation on treatment.

Physicians who attempt to poach patients when doing second opinions deserve special mention. Everyone knows this occurs, and this may be one reason why some private-practice physicians hesitate in otherwise recommending second opinions. Certainly, there are occasional good reasons for a doctor giving a second opinion to offer to take over care. Some patients may legitimately benefit from access to less common technologies and treatments, such as protons or allogenic bone marrow transplants that are unavailable in the community. Many high-quality private practices have access to clinical trials, but no one has access to everything, and the availability of a promising trial is another excellent reason to offer a transfer of care to a patient. On the other hand, sometimes motivations are less noble. I had an older patient with low-risk ductal carcinoma in situ for whom I recommended observation. She flew to another state for a second opinion at an NCI-designated comprehensive cancer center, where the physician not only suggested treatment which I thought to have marginal benefit, but recommended that she stay there for weeks to have whole breast radiotherapy at their proton facility.

When done well, second opinions can be a wonderful thing. They can improve patient care, reassure patients and families, and help to deepen doctor-patient relationships by inspiring confidence and demonstrating to the patient that their physician is open to any ideas or suggestions that will improve care. Second opinions can serve to help disseminate new treatments and data from academia to the community. When done poorly, however, they can unnecessarily undermine patients' faith in their physicians and sow confusion over relatively minor details at a time when patients and their families are better served by focusing on the big picture.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.