The thing about second opinions
If you’re thinking about getting a medical second opinion, you should probably consider what a third opinion might tell you too.
A study of more than 20,000 pathology slides reviewed by the Mayo Clinic in the United States found the clinic doctors’ second opinions differed from original readings of the slides a “very encouraging” 1 percent of the time, Dr Lisa Sanders from the Yale School of Medicine told a Victoria University of Wellington audience.
The discrepancies affected either prognosis or treatment.
However, when 100 of the patients were biopsied again for a third opinion, 13 reverted back to the original opinion.
Sanders’ point, in a public lecture organised by Victoria University of Wellington’s Faculty of Health, is that second opinions play an important role in medicine but the process is neither straightforward nor infallible and should be approached with caution.
“It’s not like going to God to find out what the right answer is. It’s a little bit more complicated and a little bit less effective.”
Sanders teaches topics related to clinical reasoning and diagnosis to medical students and resident doctors at Yale and since 2002 has written the Diagnosis column for the New York Times Magazine. The column was the inspiration for the hit TV series House MD, for which she was an advisor.
“I’m interested in interesting diseases but what I’m really interested in is the cognitive and other kinds of processes that get from mystery to solution,” she said.
Sanders became particularly interested in second opinions after seeking one herself when she was unhappy about being told she would have to wear a colostomy bag for a year or more following removal of a cancerous section of her colon.
“People had told me, my patients have told me, they hesitated to get a second opinion because they were worried their doctor would be mad at them or they thought they were cheating on their doctor like these are some sort of intimate relationship.
“And there is, of course, a very intimate relationship between doctors and patients, but I thought, ‘I don’t care.’ I felt bad but not that bad. And I did feel a little bit like I was doctor shopping. When people wrote to me [through my column] for a second opinion, I knew I was ninth or tenth or twelfth or fifteenth or twentieth on their list. I know many of the people that contact me for a second opinion are doctor shopping. They are looking for a doctor that will tell them what they want to hear.”
Sanders found such a doctor: a respected surgeon who said it would be risky but he would operate without, if possible, a colostomy bag being necessary afterward.
As the Mayo Clinic study showed, and even studies with much higher levels of disagreement, second opinions are more likely than not to support first opinions.
And that in itself is a good reason to seek one, said Sanders. “To get somebody to reassure you that what your doctor told you was right. So you can get the treatment everybody thinks you should get.”
Doctors themselves have been getting second opinions on patients for a long time. Doing so is often routine and in some cases mandatory.
“I know that as a practitioner, whenever you’re up against something where you don’t know what’s going on and somebody’s critically ill, you reach out to the smartest doctors you know and look for second opinions in a more informal way. The halls of any hospital are filled with people getting what’s called ‘curbsided’. You’re just pulling somebody over and saying, ‘I’ve got this really sick woman, she’s got this, that and the other thing, what do you think? Should I do this or do you think it’s this?’ Because medicine is such a story-driven practice, sometimes that one-liner is enough to solve the case.”
In the US, patients are increasingly seeking second opinions. “It was like a piddling five or 10 percent and now it’s about 10–20 percent depending on where you look at it,” said Sanders. There are now companies focused solely on giving second opinions.
In one large study, second opinions agreed completely or had just minor discrepancies in more than 80 percent of cases, although in 16 percent there was major disagreement.
In another study, of men seeking second opinions after prostate cancer diagnoses, there was a change in treatment but not diagnosis 40 percent of the time.
“These are men who had early-stage prostate cancer that hadn’t spread outside the capsule, so this is a very controversial area. Once it’s spread, all bets are off, but while it’s still in the capsule it’s not clear which way it’s going to go. Where there was a discrepancy between the first and second opinion, the second almost always recommended much more aggressive treatment.
“To me, this seems like a bad thing. The treatment for prostate cancer is a really maiming surgery that causes impotence and incontinence in men, sometimes for the rest of their lives. Going to surgery sometimes makes patients feel better – ‘I can’t bear to have that in my body.’ But it may not make the rest of their lives better and it may not have any effect on their mortality.”
For Sanders: “What’s interesting and strange and a little bit disappointing to me is none of these studies really looked at whether the first opinion or the second opinion or the third opinion or the fourth opinion was right. Or whether taking the second opinion advice made any difference at all about whether these people lived or died or got sicker or got better.
“There are ways to check this and they’ve never been done. I don’t know why this is exactly. It seems to be it would be obvious you would want to know. I think the assumption is that of course the second opinion is better. Duh! But there are lots of things we feel ‘Duh!’ about where we were completely wrong.”
Sanders’ original New York Times column was in 2011–2017 joined by one where she provided readers with all the details about a medical case and let them see if they could come up with the same diagnosis the patient actually received.
Taking that a stage further, she has a Netflix series where she does exactly the same thing except with patients who have received no diagnosis yet, their condition continuing to baffle doctors.
What patients want “is a broad differential. This is the broadest possible differential”, Sanders said of her crowd-sourcing approach.
“As someone who saw all the suggestions that came in [from viewers], let me just say most of the suggestions didn’t really make sense. If you really knew medicine and really read the case, you would realise most of them were not right. But there was a core in each of those cases of really smart, thoughtful takes that were worth investigating, and I think when you don’t have the diagnosis that’s all you can ask for – having a different perspective, a different take on what’s going on.”