Newsweek’s Sharon Begley pens an excellent piece on why Americans will reject evidence-based medical decisions.
She opens with a vignette, which every emergency physician should be familiar with:
A 4-year-old suffers minor head trauma, perhaps from falling off a swing and hitting her head on the ground. She is dazed, and although she doesn’t lose consciousness her worried parents—visions of subdural hematomas and concussion dancing in their own heads—rush her to the local emergency room, expecting that the doctors there will immediately do a CT scan.
The data, however, suggests that a head CT will not improve care in these cases, and, “in more than 99.9 percent of the cases, the rules accurately predicted which ones did not have a serious brain injury and could therefore have skipped the CT.”
But American patients tend to obsess about that 0.1 percent where the rules missed a serious brain injury.
Ms. Begley rightly sums up that sentiment: “Americans believe that there is no such thing as too much treatment. That, of course, ignores the fact that few treatments are without risk and that every time the medical system gets its hands on you there is another possibility of medical error.”
There is a fundamental belief in our society that more medicine equals better care. But that’s not always the case. Progressive health reformers tend to single out medical providers as the driving force behind ordering more tests. And with all the incentives within the health system favoring more tests and treatments, that’s partially true.
But they need to acknowledge that patients need to shoulder some of that blame too.
{ 4 trackbacks }
{ 30 comments }
People don’t get the fact that a CT scan subjects a patient to a thousand times the radiation of a regular X-ray. When they develop cancer from a excessive number of CT scans they may figure it out.
Well, if after having the situation explained to them, including the costs and the relative cost/benefit analysis, the patient still wants the scan, that’s their choice. Following the EBM guidelines will have insulated the physician from liability, which is most physicians’ primary goal when they order tests, according to them.
If the patients shoulder some of the blame, it’s not much.
Trouble is our culture of medical TV shows lead patients to believe this. And the doctor, fearful of lawsuits, is only too happy to comply.
Let’s say patient’s shouldered a higher percentage of the ER bill, rather than just a $50 co-pay (or whatever). What if they owed a percentage on what tests were ordered. Would they then be so willing to have everything in the book?
Grumpy is corrent. The “problem” of medical consumption is 2 fold. Doctors/ hospitals have the incentives to order every test in the book, and the third party payer system gives patients the incentives to consume as much as possible.
Patients must bear more of the cost of their care–and doctors must bear more of the cost of their failures to treat effectively.
Only then will incentives be properly aligned.
Communication is the key. An effective communication of risks. I agree with John and “Anon”. If you tell the parents the risk of dying when they drive back to their homes, maybe they will realize the true dangers of radiation and sedation needed for the CT scan. Risks are always present whether we like or not. The matter is by how much, how we handle the thresholds, and most importantly, how do we as physicians communicate with patients, with empathy and professionalism.
Dr. Grumpy is spot on.
Communication is important, but even after discussing risks and benefits of doing CT scans in cases like this, more often than not I still get the question “so you’re still going to do the CT, right?”
If patients had more skin in the game, the desire for these extremely low-yield tests would dry up significantly.
The blame remains principally with we physicians.
I’d like to echo the sentiments of the other posters. The fact that a patient chooses to “obsess about that 0.1 percent” is an absolute failure of communication and counseling. In my experience the number of patient’s who will insist upon a non-evidence based test following actual counseling (rather than just throwing out a probability) is exceedingly small.
I also agree that if the patient had to financially shoulder a test where there is a 0.1% of likelihood for useful information we would quickly achieve cost savings AND adoption of evidence based medicine.
This is pure laziness and craveness on the part of physicians.
Risks vs benefits must be EXPLAINED. Caution and caring as opposed to rushing an failing to listen are required to gain the confidence of any worried parent. And of course least reassuring of all is appeal to some statistical point. You must frame care decisions on the basis of each and every individual patient. What factors are present that might make the case deviate from your optimisitc statistics?
What precautions can be taken to minimize the devastating impact of the outlier case?
Parents need to know you aren’t treating a herd but their very own child. That takes work. That takes attention, real observation of the INDIVIDUAL, listening, and explaining.
I
I have had three CT scans. I never requested them, the risks/benefits were never explained to me. One of those CT’s were for defensive medicine. So is the doctor responsible for the cancer?
And where are the studies that show that a majority of patients insist on extensive testing?
The trouble is that most people think anecdotally not statistically. It is hard when your own child is injured not to be obsessed about that 0.1 percent.
In cases like this where the evidence advises against the intervention I tend to let the parent know the risk in terms of anecdotes too: the rare kid that gets a cancer from radiation from CTs.
Generally once people understand that you are trying to protect their child not ignore their concerns they are fine with going home with head injury precautions.
I would also add that this would be an excellent area where comparative effectiveness research could easily identify best practices. As a father of three young children I have had 2 visits to an ER following a fall and possible head injury. During one occasion we received a CT scan that (in retrospect) was truly unnecessary; and in another case we were sent home with head injury precautions.
(In both cases, the ER physician was “unblinded” to the fact that my wife and I are health care professionals)
As a consumer, I actually preferred the physician who explained the precautions and did not (reflexively) order the radiologic study.
As a physician and scientist, I certainly understand the cost savings and prospect for enhanced utilization of resources with the two different practice patterns.
It’s somewhat odd to hear physicians criticize patients for acting based on tiny risks or on anecdotes considering physicians do the very same thing and call it defensive medicine.
SarahW -
I’ve carefully read many of your ubiquitous comments on this site and am beginning to wonder whether you have one comment that you just cut and paste into the newest post. The theme throughout your comments can be summed up as followed: Many doctor’s are incompetent, negligent morons who treat “the herd” rather than the individual patient, which is why they are sued as often as they are.
So I pose this question to you: What are these “individual” attributes that you so often speak of, which take precedence over evidence based guidelines? Are we talking about the general anxiety of the patient/parent? How vocal/forceful they are? Insurance status? Their general level of medical knowledge? What about professional status as either a health care professional (as the commenter above) or lawyer? Or VIP status such as hospital donor or celebrity? I could go on but that’s a good sampling of common non-clinical, “individual” attributes that often guide medical decision making. Do any of these characteristics (particularly the first, which is an extremely common reason for ordering unnecessary tests), truly justify clinical decisions? Does a patient who just “wants the test” deserve it, even if it offers little to no benefit, when the same test would not be ordered for a patient who wasn’t as vocal or worried? The clinical guidelines proposed by evidence based medicine already account for many “individual variations” such as mechanism of injury, age, co-morbidity, symptoms (or lack there-of), and physical exam. How do you identify the 0.01 percent of bad outcomes not detected with these algorithms (”cook-books”) without testing 100% of people? Do you know of a way where this 0.01 percent can be identified out of the tens of thousands of people (”the herd”) who present in the same way without testing everyone? If you can, I’d be glad to hear it, as would every other physician reading this post. Or should we just scrap all evidence if it doesn’t succeed in 100% of cases? I eagerly await your answers.
Matt,
The patient and the doctor are making different calculations when they do this. To a rough approximation, whether or not you should have a diagnostic test can be calculated as the risk of having the condition to be ruled out, multiplied by the seriousness of the condition, minus the risk of complication from the test, multiplied by the seriousness of the complication.
When a physician is practicing defensive medicine, he is adding extramedical factors to this calculation. He adds the risk of being sued and the damage it will cause to his own career to the equation and often decides to do a test which would not be justified with purely medical reasoning.
I think that patients, assuming that they are adequately counseled and understand the risks, may also rely on extramedical factors to make their decisions. They may think that the added peace of mind they will get from the CT scan outweighs the risk from the CT scan. Or they may want to feel empowered by using every option that they have, even if there is little hope of benefit.
The problem is not with the patients or the physicians per se, its in the framework of our health care delivery system. It may be rational for individual physicians to practice defensive medicine and for individual patients to demand medically unnecessary tests, but iterated thousands of times a day, it leads to an unsustainable health care system, unhappy physicians, and patients with needless and avoidable complications.
All true. But at the end of the day the point remains that we’ve got people, both patients and doctors, making decisions resulting in “unhappy physicians. . . and needless and avoidable complications” as well as allegedly billions of dollars, based on risks they barely comprehend. The physician typically has no clue what his risk of being sued is, nor whether X or Y test reduces it in this situation. Likewise, the patient typically has no clue what the real risk of the test being performed is, or the cost/benefit analysis. Both are relying pretty much on anecdotes.
I just don’t see why physicians would criticize patients for doing the same thing they do.
Having to shoulder more of the cost would probably tip the scales for a lot of people in favor of not having the test. But as soon as that one child who’s the 0.1% has a bad outcome because his parents didn’t have a several-hundred-dollar scan, the headlines will be “child dies because parents couldn’t afford care”.
In this situation of a child with a sustained head injury, there is more than just one correct answer because there are many different types of parents and many ways that physicians practice medicine. I generally categorize patients in one of four classes: (1) patients who completely trust their doctor and want the doctor to decide the treatment, (2) patients who want to understand the possible diagnoses and treatments but who want the physician to choose the best, (3) patients who want to understand the possible diagnoses and treatments and make the decisions about treatments for themselves, and (4) patients who want to understand and want to work with the physician to reach a mutually agreed upon decision. I think that the physician should try to understand which category best fits the patient or parents and then try to work with them to reach a mutually agreed upon course of treatment. This is a tall order, I realize, given the generally limited amount of time of interaction between patient and physician.
Matt,
I don’t think this post is about criticizing patients as much as it is to understand frustrating patient behavior. Physicians are trained to follow evidence, so things that prevent them from doing this, such as fear of litigation and patient insistence, are frustrating. Hopefully tort reform can help mitigate the first problem, but I think that human nature will perpetuate the second.
I wasn’t referring to the post so much as the comments.
Tort reform, in its current forms, has existed for decades, and does not appear to have changed a thing.
It seems to me that patients desire low-yield tests precisely because their skin is literally in the game. If the ‘bad outcome’ is death in 0.1% of the cases, that’s a non-infinitesimal number when I consider my own life. The test’s cost seems small when compared to my survival instinct.
Let’s flip the statistics upside-down. Would anyone here take $1,000 (ballpark CT scan cost) to join a pool of 1,000 people in a death-row lottery?
Even if the statistics sound small to the physician, why shouldn’t someone desire a potentially life-saving test? The astronomical cost of treating the heard is important for the political debate, but the consequences for the individual are clearly more profound than for the heard. The heard will survive long after the individual is dead.
Tort reform has affected the exodus of physicians from some states reduced frivolous suits and malpractice costs. Look at the effects in Texas.
Matt L,
You’re forgetting that the CT scan itself comes with risks that outweigh the benefit. If it were merely a question of cost, it may still make sense to have a low-yield test, but if the test has a greater chance of hurting than helping…
There is no evidence that tort reform has affected physician population. If that were a true driver of access California would have more physicians per capita than anywhere in the country having had “reform” for three decades.
Texas physician growth trails it’s population growth. Is it surprising that there are more physicians in states with a growing population? What it is surprising, if you think tort reform works, is why California is the second highest source of physicians moving to Texas.
Matt L makes an excellent point:
“Would anyone here take $1,000 (ballpark CT scan cost) to join a pool of 1,000 people in a death-row lottery?”
A lot of people see this debate in these terms–high financial costs vs. safety.
I think it is the physician’s job to point out that their is skin in the game on both sides. 1,000 CT’s may find one clinically invisible bleed (which often would not actually be fatal). But how many cancers would 1,000 CT scans cause? How many benign abnormalities might be mistaken for problems and lead to risky surgeries?
The people who write these recommendations study these risks intently. We aren’t talking about a benign thing when we irradiate the brain of a 4 year-old. Not only may low-yield CTs cost money but lives.
The skin game cuts both ways. Unfortunately everyone has a tendency to accept risks while doing something big and invasive that they would never accept while doing nothing. Both physicians and patients have a dangerous bias toward “doing something” rather than waiting, even when waiting to see is the best decision.
Head CTs is only one example. Look at the massive prescription of antibiotics for flu-like symptoms if you really want to see this bias in effect.
Radiologist, just what do you think I mean? I mean every individual is unique and every case is unique. What is good for most patients isn’t good for all patients. What most patient’s have isn’t what all patients have. It’s your job to know which is which, and not to excuse the use of judgment with relianceupon some expert “system” or evidence based guideline. Perhaps the patient in this case looks awful and/or Has a family history of bleeding disorder and/or had a similar blow the previous day.
I don’t think physicians can be excused by a practice guideline for missing something right under their noses.
Thanks for the reply SarahW. Of course you’re right, but what you are saying is not incompatible with practicing evidenced-based medicine. I think your comment exhibits a lot of the misconceptions regarding the practice. While I do think that doctors should be excused for bad outcomes when practicing under evidenced-based guidelines, the argument that that this would excuse doctors in all situations is false. It would certainly be malpractice for a doctor to apply a specific guideline when such a guideline may not be appropriate for a given patient. For example, guidelines regarding head injuries in children lay out very specific criteria for who can forego a CT. These are:
“For children under 2, if there is normal mental status, no scalp swelling, no significant loss of consciousness, no palpable skull fracture, and no behavioral changes (as reported by a parent), and if the injury occurred in a nonsevere way (something other than a car crash), it is safer to skip the CT scan.”
If anything in the patient’s history, presentation or physical exam deviates from these criteria (i.e. bleeding disorder, “looks bad”, focal neurological signs, etc.) than a CT would be warranted and a doc who didn’t order it would be remiss in his responsibility to the patient (and liable if a bad outcome occurred). Doctors would never be FORCED to use a guideline in all circumstances. In just the same way as guidelines cannot exist without clinical judgement, clinical judgement really can’t exist without evidence to guide it.
The thing is…and this is the part where I’m sure to lose you…most patients do not have a unique characteristic that sets them apart from the 10,000 other people in the study. Most kids fall off their bike, don’t lose consciousness or vomit, have normal exams, look fine, have normal vitals, and aren’t on blood thinners or have a bleeding disorder. In other words…for the vast majority of patients presenting with a given symptom (in this case 99.9%), they are, for the purposes of triage and diagnosis, enough like the many other people presenting with the same problem, that guidelines are useful. As long as the child did not have some identifiable characteristic that rendered the guideline inappropriate, the doctor, who did not order a CT in that one patient, should not be sued for the bad outcome. Like I asked, if you know a way to identify the 0.1% without scanning everyone (which, I would argue, is much less individualized than following guidelines) I would like to know – but it’s not going to be some “unique” feature that only an astute clinician can identify. This may work on “House, M.D.”, but not in the real world.
Bottom line…what you say is true, and this is the reason most guidelines are very specific as to which patients they apply to. In fact, many guidelines (such as the Fleischner Criteria for Lung Nodules or the Consensus Statement on the Management of Thyroid Nodules, to use two radiology examples, include in the algorithm itself some caveat about high risk factors or other clinical features which would render the criteria moot).
You may still not like evidence-based medicine even after all of this but just so you know – next time you go to a doctor and he orders some test or prescribes a certain drug, it’s because there’s a compilation of EVIDENCE out there in the literature GUIDING the BASIS for that decision. That is, if he’s a good doctor.
Frankly, I wish I had $20 for every time I’ve had to hand my physicians reports with evidence-based medicine about which THEY knew nothing. And I’m a patient. So, I don’t buy that line that “next time you go to a doctor and he orders some test or prescribes a certain drug, it’s because there’s a compilation of EVIDENCE out there in the literature GUIDING the BASIS for that decision. That is, if he’s a good doctor.” Even good doctors don’t know it all. And often they know a lot less than they think they do.
That is not criticism. It’s a statement of fact, and until we learn to work together with respect instead of pointing fingers, we aren’t going to do things any better.
Just my two cents (and I have plenty more if you want to hear it. )
I think EBM guidelines gained the notorious reputation from being unstable and rapidly becoming outdated. According to Twila Brase (2008) in (EBM: RATIONING CARE, HURTING PATIENTS), more than 75 percent of the guidelines developed between 1990 and 1996 needed updating. In addition, they discovered that half the guidelines were outdated in 5.8 years! Therefore, mentioning EBM to some patients is unfortunately enough to make him/her think twice about the physician and his care plan! Although they get obsessed about the 0.1% bad possibility at the hospital; on the contrary, they accept a long list of medications’ side effects!
However, I think more emphasis should be put on lightening patients about EBM. For example, they should know that EBM is practically equal to “Best Practice”, and the legislatures have started adding EBM requirements to health care laws.
Excellent post, excellent points.
But the point seems to be that patients will make the best decision based on the evidence they have at hand. If they are not making the best decision, based on the doctor’s greater command of greater data, then isn’t the role of the doctor to educate and share that evidence?
You can’t blame patients for wanting the best care available based on the evidence they have at hand, while at the same time not sharing all the evidence at hand.
That’s what…self-serving? self-absorbed? condescending?
The problem lies in communication. We often aren’t given the time to explain the “old fashioned” way of sitting down and explaining slowly and if we do, the boss punishes us. It might be better to have a speech ready with reading materials and all. For example, if a patient were given one paper explaining (with references) the ADDITIONAL lifetime risk of cancer with a single CT ( we don’t really know and may not find out due to various factors ) and another that explained the chance of finding something bad that we can DO SOMETHING ABOUT, the patient could make a rational decision. The problem is obvious – we lack information. So what does the parent do, go with their gut which feels much better after a negative CT scan.
Who was our real patient in this case? Which came first the chicken or the egg?
Comments on this entry are closed.