Much has been made of Mark Cuban’s medical knowledge since he tweeted, “If you can afford to have your blood tested for everything available, do it quarterly so you have a baseline of your own personal health”. Charles Ornstein shared the tweet and many physicians and others, myself included, weighed in on the costs and potential for harm from unnecessary testing.
I’ll admit that, when I tweeted to him, I expected Cuban to agree. But he didn’t. In fact, he grew increasingly resistant. I stopped responding when he announced that the opposition to his idea his had convinced him he needed to take his proselytizing to his TV show.
Instead of poking the sore, I began to wonder about the origins of Cuban’s conviction. I remembered that he is not alone in wanting tests that clinicians who worry about value, cost, and harm think he shouldn’t have.
But where do these attitudes come from? Is it possible that clinicians are contributing in any way to this situation? Quite the contrary: most Americans want tests, even when you tell them that nothing can be done with the information. Furthermore, Americans are more convinced of the benefits of tests like mammograms than people in other countries, and then go out and get more of them.
I think that we are. My team has studied why patients get so many electively placed coronary stents, when cardiologists readily admit that randomized trials have demonstrated that there are few situations in which such stents improve survival or reduce the risk of heart attacks.
Studies of the beliefs of patients who have just received an electively placed stent give a big clue: 80% thought stenting would reduce their risk of death, even though their cardiologists knew that this was not the case.
One might think that such misunderstandings occur because medicine is complex and we have to give patients lots of information in short period of time. However, I think that lets us clinicians off the hook too easily. There is ample evidence that clinicians and clinical organizations actively lead the public to overestimate benefits and underestimate harms.
A study of news releases from US News and World Report top 50 hospitals, for instance, show that those hospitals often overstate the potential impact of research studies and fail to mention important caveats. For instance, only 10% of news releases about animal studies note that the findings might not translate to humans, while only 1 in 6 news releases about uncontrolled studies of clinical interventions mention that there is no control group.
(One could even ask why a top research institution would ever issue a news release about an uncontrolled study, but that’s what 5% of the news releases were.)
In a recent study of 262 hospitals that offer Transcatheter Aortic Valve Replacement (TAVR, a procedure recently approved by the FDA), researchers found that the hospitals’ web pages devoted to TAVR almost always described the benefits, but mentioned any risk at all less than 30% of the time. There are no length limits on web pages, nor does it cost more to put more information on them, so it is hard to think of why this important issue usually goes unaddressed.
It’s not just hospitals that ignore information that suggests common practices aren’t beneficial. In a recent review of specialty society responses to new data that challenged standards of practice, societies in their communications to members recommended continued use in spite of the new evidence most of the time, while editorial writers in scientific journals did so less than 30% of the time.
So, if we, as clinicians and clinical organizations, tend to overstate the benefits of our work, understate the associated risks, and resist information that suggests we shouldn’t be doing it, those are the messages our patients are likely to receive.
Mark Cuban has a doctor he’s been happily working with for years, collecting unnecessary data without any pushback. When he mentions this on Twitter, a few physicians on Twitter tell him there’s potential for harm, but the vast majority of Twitter docs are silent. Then he reads stories written off hospital press releases and sees clinicians’ web pages, and comes to the conclusion that we can do more than we can.
Unfortunately, in this case, Mark Cuban isn’t being a maverick. Instead, he’s accepting the conventional wisdom as presented for years by us, the clinicians and hospitals around him.
We have to shoulder at least some of the blame for his opinion. And we can start to fix it today.
Adams Dudley, MD is Professor of Medicine and Director of UCSF’s Center of Health Care Value.
Categories: Uncategorized
Good point. In a market without third party payers (whether it’s a government or a private insurer), one can expect scrutiny over whether the expenditures are valuable or not. With third party payers, the focus may just be on will this reduce the patient’s or the physician’s anxiety that an abnormal result has created.
Sunil, I think this particular problem is due to the fact that the usual marketplace decisions cannot exist when government is so involved in the payment process. Under normal market conditions there is a brake that stops a lot of unnecessary care from being done. Government’s brake is to overspend while underfunding the most important things which denies a lot of care and useful innovation.
When it comes to a lot of things physicians do not have solid evidence. Remember, each patient is different with different co morbidities and a whole host of other differences. Thus, the nature of medicine is to see many different ways of managing many different patients.
You are right: I am not blaming patients. But I’m not saying doctors are stupid. Rather, I’m saying they are human, and find it hard to leave abnormal alone. So, as long as they are human, we shouldn’t get tests that might come back abnormal unless their is something specific we are looking for. The video recommended by @BobbyGvegas addresses this.
That’s a good video by Aaron Carroll (@aaronecarroll https://twitter.com/aaronecarroll)
@Adam Dudley
Quoting you :
“If there is anything abnormal (whether by lab error or random variation or a true illness), patients are likely to show it to doctors. Doctors consistently then begin working it up, even though, for most patients with no symptoms, it’s an error or random variation. ”
It seems there is a problem with the doctors not the patients. So basically the summary of the response is
1. Doctors don’t have the time to deal with the lab tests and so order follow up.
2. Doctors are stupid and order follow up testing.
3. Or in the current system , the only rational choice is to order follow up testing.
This is a call to change the current system, not to fault the lab testing.
https://www.youtube.com/watch?v=bW_V7EZNW84
Yes, I think that is Mark Cuban’s point, and it makes sense, except for human behavior.
The problem arises when you actually look at the results of the testing. At that point, you have to account for both lab error and emotion. If there is anything abnormal (whether by lab error or random variation or a true illness), patients are likely to show it to doctors. Doctors consistently then begin working it up, even though, for most patients with no symptoms, it’s an error or random variation.
Study after study shows that this approach, on average for a group of patients undergoing the tests, almost always wastes money and causes harm. The harm comes in many forms—radiation from x-rays, complications from procedures done without justification, side effects from drugs taken “just in case”. Now, for an individual patient, there is no way to know what the outcome will be. You may take that drug without harm. But that’s why we do studies and compare the average harm to the average benefit.
There is a group, the United States Preventive Services Task Force (USPSTF), whose job it is to look for exceptions to the no-testing-without-symptoms rule—specifically for instances in which testing without symptoms makes sense. They recommend tests where the chance of disease is high enough to outweigh the risk of harm from getting the test. The USPSTF members are trained scientists who have dedicated their lives to the concept of preventing disease by preparing for it or recognizing it early, so they are a group we would expect to be in favor of testing whenever the benefits outweigh the harms. Their recommendations can be found here:
http://www.uspreventiveservicestaskforce.org/Page/Name/recommendations
In all other situations besides those where the USPSTF recommends testing, we either know that testing does more harm than good, or we have no reason to believe it will help. The tests Mark Cuban suggests are not on the USPSTF list. In fact, nothing is recommended quarterly by the USPSTF.
I think everyone is missing Mark Cuban’s point. Everyone seems to be thinking that he’s using quarterly blood tests to reach an early diagnosis.
I think he’s trying to get a handle on his average blood levels and a standard deviation so that he does start feeling really sick or is heaven forbid diagnosed with cancer or some other major ailment, he can compare his post-diagnosis blood work to his pre-diagnosis blood work to get a sense for how bad it is.
It’s possible this is a pipe dream to be able to do this, but I don’t get the sense he’s trying to diagnose himself through this practice. Everyone’s responses seem to be knee-jerk “too much testing is a bad thing” without engaging with him. Mark Cuban is a really really smart dude, I give him the benefit of the doubt that what he’s advocating is not stupid.
The current over-emphasis on benefits and “don’t just stand there, do something” is a reflection of the incentives providers have to do more and get paid more. It is hard to get people to act against the incentives in the system.
Now there are payment reforms being pushed that change those incentives. These may lead providers to de-emphasize benefits and discuss risks more, but we’ll need to monitor that an incentive to do less doesn’t lead to trying to talk patients out of things they need. Would be easy to go too far the other way.
Generally, the best things to fight personal interests are having a strong culture and monitoring behavior. In this context, having a strong culture means making sure medical education and organizational leadership emphasize evidence-based medicine and seeking value in health care. Monitoring behavior means measuring performance, looking for evidence that patients aren’t being informed, and calling it out.
“The over examined life is not worth living.”
Aristotle
Well, it’s easy to see how different people — individual clinicians, researchers, journal editors, medical societies — have an interest in highlighting things that sound like success. Even if, upon closer scrutiny, it’s not very successful in the way people assumed it was.
Adams, how do you propose we go about fixing this? Seems it’s an uphill push against human emotions and also financial motives in some cases.
The problem, in two words, “anticipatory medicine.”
It started when a wise man and president of the American Medical Association, Wendell Phillips, pronounced with a permanence that only the hubris of an intellectual can pronounce:
“Physicians must give a new significance to the word patient, for in the new order of things both sick and well people must be recorded in the list of their patients.”
Philips wanted physicians to give patients “exuberant” health.
Right!
The root of the problem may well be physicians but physicians alone cannot solve it. To pretend otherwise is futile, but morally pleasing, self-flagellation.
Note this Bloomberg story on LabCorp’s recent decision to allow customers to order their own tests —
http://www.bloomberg.com/news/articles/2015-04-20/the-doctor-is-out-labcorp-to-let-consumers-order-own-tests