“Just to be sure”

by Walter Olson on January 19, 2009

White Coat begins taking notes on how many times he practices defensive medicine in the course of a day in his emergency room, and concludes that, no, the whole phenomenon isn’t just a figment of his imagination the way so many lawyers say it is. More: Max Kennerly takes an opposing view, and White Coat returns for a followup post.

{ 1 trackback }

More on Defensive Medicine « WhiteCoat’s Call Room
01.21.09 at 7:15 am

{ 24 comments }

1 Matt 01.19.09 at 12:12 pm

Congratulations to him for doing his job. He’s right — “probably” doesn’t cut it — and shouldn’t. It doesn’t cut it in the rest of the world, either, where the improbable consequences are severe. If you took your car in for soft brakes, and your mechanic told you he didn’t actually check them, but they “probably” wouldn’t fail, you wouldn’t accept it. If you’re having electrical issues at your house, you wouldn’t put up with your electrician saying that he really didn’t check all the possibilities, but your house “probably” wouldn’t burn down. You wouldn’t let your lawyer get away with telling you he hadn’t really looked into all the facts or the law, but you “probably” won’t get in too much trouble for whatever you’re concerned about.

When you hire a specialist for something potentially severe, including a doctor, you don’t want a “best guess” without an examination of the actual facts. That’s why you hired the specialist — I guess we’ll call it defensive living.

2 Jason Barney 01.19.09 at 12:32 pm

Ah, but here’s the kicker, Matt. In a civil case all the plaintiff has to prove is the defendant is “probably” responsible for the alleged tort (i.e.: more likely than not). It’s a flimsy standard of evidence that favors recovery and often leaves very much doubt as to the actual culpability of the defendant. I’ve often thought a more reasonable burden would be by “clear and convincing evidence”–greater than the “preponderance” standard but arguably lesser than the “beyond reasonable doubt” standard.

3 Dirk D 01.19.09 at 1:35 pm

” It doesn’t cut it in the rest of the world, either,”

Actually, in terms of medical care, it does.

4 VMS 01.19.09 at 1:47 pm

Matt is 100% correct and White Coat has developed “GOMER (get out of my emergency room” syndrome which he should cure before he gets into trouble.

The physician must eliminate everything on the differential diagnosis by perfoming a risk-benefit analysis. For example, once patient #2 stated that he could not bear weight on his foot, a fracture was a real possibility and an x-ray was required. The cost of the x-ray and the minimal exposure to radiation to conclusively determine whether or not there was a fracture, outweighed the consequences of sending the patient home with an untreated fractured foot.

A patient on Coumadin who has sustained a head injury MUST go for a CAT scan to check for intracranial bleeds even if the chances that such a bleed occured in a particular case is a few percent. The consequeces of the harm caused by missing such a bleed when it is still treatable is death (or worse). The costs that this consequence is balanced against is the price of the test (~$1200), and the VERY high radiation dose to the patient, and the possibility of an anaphylactic reaction to the contrast. This is a no-brainer. Do the CAT scan. It would be malpractice not to do so.

The only case that White Coat presents where furhter test and treatment was not indicated is #5. When there were no signs of a fracture and there were no consequences of not treating a possible hairline fracture, an x-ray was not needed.

5 VMS 01.19.09 at 2:20 pm

Matt is 100% correct and White Coat has developed “GOMER (get out of my emergency room)” syndrome which he should cure before he gets into trouble.

The physician must eliminate everything on the differential diagnosis by perfoming a risk-benefit analysis. For example, once patient #2 stated that he could not bear weight on his foot, a fracture was a real possibility and an x-ray was required. The cost of the x-ray and the minimal exposure to radiation to conclusively determine whether or not there was a fracture, outweighed the consequences of sending the patient home with an untreated fractured foot.

A patient on Coumadin who has sustained a head injury MUST go for a CAT scan to check for intracranial bleeds even if the chances that such a bleed occured in a particular case is a few percent. The consequeces of the harm caused by missing such a bleed when it is still treatable is death (or worse). The costs that this consequence is balanced against is the price of the test (~$1200), and the VERY high radiation dose to the patient, and the possibility of an anaphylactic reaction to the contrast. This is a no-brainer. Do the CAT scan. It would be malpractice not to do so.

The only case that White Coat presents where furhter tests and treatment were not indicated is #5. When there were no signs of a fracture and there were no consequences of not treating a possible hairline fracture, an x-ray was not needed.

In response to Janson’s comment, the Plaintiff must prove by a preponderance of the evidence that the defendant physician deviated from the standard of care. A higher burden of proof is not required. Sometimes there are multiple standards of care. For example in microfocal prostate cancer, one can do nothing, do surgery, or treat with radiation. All are acceptable treatments, and each has its benefits, risks and drawbacks. In that situation, it behooves the physician to explain the risks and benefits of each treatment, make a recommendation, but leave the ultimate decision up to the patient after fully informed documented consent. If these steps are followed and the decision was to do nothing, and the cancer spread, there would be no legitimate malpractice case.

6 Todd Rogers 01.19.09 at 3:19 pm

I’m starting to think it would be wise to collect interests in companies which provide goods and services in medical diagnostics. So long as typical MDs feel inclined to resort to such (excessive and/or unnecessary?) paths, maybe there’s a way to make money on this kind of thing (aiside from suing them or their insurers).

7 Griffin3 01.19.09 at 5:23 pm

Matt is not 100% correct. He may be, but he asks the wrong question:

If you took your car in for soft brakes, and your mechanic told you he didn’t actually check them, but they “probably” wouldn’t fail, you wouldn’t accept it.

If you took your car into the shop with soft or squeaky brakes, and the mechanic took a quick $50 look, adjusted the pads, and said they were probably fine, you might be skeptical. But, if the same mechanic offered to rebuild the entire brake system for $1200 (plus IV costs and utility use fees), you might actually tell him no. Didn’t a similar thing happened with you last car, reader, that funny sound under the hood that you never could figure out, but never seemed to bother anything. If it fails because you refused a $1200 rebuild, does that make it your mechanic’s fault?

Of course, if you could make someone else pay for every repair to your car, except for a $20 co-pay, then you too might have every minor problem of your car diagnosed to perfection, no matter what the cost. Because you deserve the best health care, right? Right?

8 throckmorton 01.19.09 at 9:36 pm

VMS:

Why would you use contrast on a head ct if you are looking for blood? You actually dont use contrast for these scans so the chance of an anaphalactic reaction is zero. Also, the radiation dose for a head ct is very low. The cost of the head CT is considerable however. ($1400).

Here is the real debate. We as a society have to decide what degree of sensitivity and specificity we will accept. At present, we want 100% of both but cant pay for it. Do we really want ct angiograms on every asthmatic with SOB? Then we really have high radiation doses and possibility of anaphalaxis. We can use evidence based medicine to define the standards. The issue is can we allow these to hold up in court? There will always be the patient who is really having a migraine who has an asymptomatic GBM that will want to sue.

9 Max Kennerly 01.19.09 at 9:47 pm

Eh, the airplane probably won’t run into geese or need to land in a river, so I’m gonna skip this training session and go grab a beer, let me know how it goes…

10 OBQuiet 01.19.09 at 10:57 pm

I’m not sure what Max’s comment was supposed to mean. IF it was an attempt at linking this sort of issue to something that might be be similar in the airline industry, perhaps he needs a better example.

I’m no pilot but perhaps: the landing gear light might have been showing a false positive or the tires or struts could fail, lets always foam the strip. Just in case. Sure, it cost a lot more, ties up resources that could be better used in other areas and adds to the passengers anxiety level needlessly(WE don’t pay for those medical costs). But if even ONE life is saved(at the added cost of 13.6Billion anually) how can you place a price tag on life!

11 B Rad 01.19.09 at 11:18 pm

VMS

It is not an emergency room physician’s job to eliminate ALL things in a differential diagnosis. Soft tissue injury to a foot can cause someone not to be able to bear weight and this is not detected with xray. Suspicion was very low for fracture. For your case #5, your logic does not hold up. You say there was no signs of fracture so an xray not needed–also low suspicion. But a hairline fracture can become a displaced fracture if not recognized and our patient who thinks he doesn’t have a fracture, continues to bear weight, goes out and plays soccer, or runs a marathon–now he needs reduction, possibly surgery.

What about the coumadin patient. Most if not all of the patients do get a CT but do they need one if they are 100% neurologically intact, not drunk, and do not have any other distracting injuries/issues?

We couldn’t do head CTs on coumadin patients with trauma in 1970 because CT didn’t exist. My guess is that doctors really had to use their smarts to determine if someone needed a neurosurgeon right away, observation, or could be sent home. I don’t think that morbidity or mortality due to bleeds was significantly higher than it is today.

12 Max Kennerly 01.19.09 at 11:30 pm

OBQuiet: the price of an airplane ticket includes the extensive costs of making the airplane able to float for a limited time, of training pilots how to deal with air strikes and water landings, of training flight attendants how to response in emergencies, and of countless usability studies for how to design the safety procedures card in each seat.

All for very low probability events. And yet I don’t see you complaining about those costs — only costs for “unnecessary” medical tests which, in truth, were every bit “necessary” to confirm the diagnosis.

Fact is, the author of the linked post, like most authors on “defensive medicine,” is complaining about the differential diagnosis, a standard of clinical medicine. There’s good reason to first rule out the possibility of a severe or life-threatening condition, and good reason not to let doctors use their “judgment” in a haphazard, gut-driven manner.

The GAO did a study on defensive medicine not too long ago and found nothing. If you’ve got something better, I’m all ears, but mostly I hear complaints of the nature “how come my flight is delayed by just a little rain?” Well, to reduce the odds of crashing, that’s why. A 30 minute delay is worth not crashing. So is the extra $10 on your ticket that went towards bird strike and water landing preparations.

13 B.RAD 01.20.09 at 9:44 am

MK

In my personal experience, for every 100 head CTs that are ordered on a patient, 1 had findings that went against the doctors’s clinical impression, changed the course of management. So in other words, doctors judgement, gut instincts are correct the vast majority of the time. But vast majority is not good enough. If this is not defensive medicine, then what is it?

14 Max Kennerly 01.20.09 at 11:29 am

I consider “defensive” medicine to be excessive or unreasonable.

A 1% chance of discovering an abnormality on a CT scan — which, given what a CT scan looks at, generally means a brain injury — is not, in my opinion, unreasonable. Nor is it unreasonable in the opinion of most physicians, which is why I have little trouble finding practicing physicians to testify that a failure to order a CT scan in the face of clinical evidence indicating a possible brain injury is a breach in the standard of care.

Even viewed from a purely economic standpoint it is, excuse the pun, a no-brainer. The sticker price for a CT scan is somewhere around $1000, with a true cost to the hospital being well below $300 (I know this because I can get CT scans for my clients for $300). Using your own example, then, we spend at most $30,000 for each person we catch with a serious brain concern that was previously unknown to the doctor.

$30,000 per patient we save from, e.g., stroke or brain bleed due to coumadin? Is that really excessive or unreasonable? To me, it’s right up there with paying for pilot training regarding bird strikes and water landings.

15 B.RAD 01.20.09 at 12:28 pm

MK

I believe the price tag is much much higher than you state. Brain CTs are only a small piece of the pie. Any hint of trauma–there’s a c-spine CT often done along with the brain–any abdominal pain?-CT of abdomen and pelvis—-slightly elevated d-dimer? Let’s throw in a CT pulmonary angiogram. Many of these are ordered based on no or flimsy clinical data. And remember,
,these studies are not done by themselves and not interpreted by themselves. Human resources are perhaps taken away from someone who is truly sick.

As was stated before, we want 100% specificity and sensitivity in diagnoses in this country but cannot afford it. Until this changes, defensive medicine will continue.

16 Max Kennerly 01.20.09 at 12:49 pm

I simply disagree, and I think both you and the original poster of the article misunderstand how medical malpractice works. It does not matter what “society” thinks or what “we want.” If you attempt to use societal / patient expectations in a medical malpractice suit, your case will be promptly dismissed.

The only thing that matters what licensed, practicing physicians believe is the standard of care. Your position is apparently that the standard of care itself needs to be lower; if so, that’s not that hard to do. Start publishing articles. Start teaching CMEs. Start lobbying hospitals to clamp down on excessive testing, Lord knows the health insurance industry will be on your side.

Fact is, doctors are encouraged to run every test they believe will help the differential diagnosis proceed because (1) it rules out severe or life-threatening conditions and (2) it adds to the doctor’s / hospital’s billing. It’s half economics, half prudence. I don’t see how we can attribute to doctors ample economic reasoning relating to hypothetical malpractice suits while simultaneously ignoring the direct financial incentive they have to order additional tests.

17 B.RAD 01.20.09 at 1:31 pm

Yeah, doctors are encouraged to run every test but you forgot the important reason—CYA—Sure a lot of CTs have to be done to pay for that $2M state of the art scanner, but someone is going to have to pay for all of this down the road. I see medical care becoming more unavailable if this trend continues. I’ve never practiced in Europe but I can’t imagine that physicians there order test after test after test like they do here. Do Europeans suffer greater than we do? Just the contrary, we’re a less healthier bunch overall.

18 B.RAD 01.20.09 at 2:10 pm

About the standard of care–It’s what one person says about one thing at one time. It changes like the wind as you can get 10 people to say 10 different things.

19 Dan 01.20.09 at 5:33 pm

A 1% chance of discovering an abnormality on a CT scan — which, given what a CT scan looks at, generally means a brain injury — is not, in my opinion, unreasonable.

I don’t want to be one of the 99 people subjected to an unnecessary CT scan, especially when you haven’t even told me the false positive rate.

Americans are really screwed up about basic math. We’re also screwed up because we want someone else to tell us our lives are risk-free, when they definitely aren’t in medical care. We will all die of something, and trying to get tests to stop this will waste a lot of money and make people sicker.

20 OBQuiet 01.20.09 at 6:43 pm

MK: The price of medical treatment also includes the cost of training doctors to deal with low probability events. When evidence suggests that these are likely enough, I have no issue with ordering the tests to find out.

As for the GAO, my memory is that many considered the definitions used by the GAO as rather dubious. Or were you referring to the study last summer that listed Defensive Medicine as one of the reasons that the amount spend on imaging has more than doubled recently to about $14B?

I was recently told to get an MRI when I was suffering from Bell’s Palsy. Since I would have had to pay about $300 of the bill, I asked if it was needed and what it was for. The Dr was amazed that I would be concerned about the cost, I had insurance didn’t I? When I explained the costs involved, she admitted that it was really just to rule out some things that were really rather remote and we could wait a few days to see how things progressed.

21 WhiteCoat 01.20.09 at 8:40 pm

Many thanks again to Walter for the link. It is a continued honor to be included in his posts.
I especially think that this debate is good for everyone.

A response to several of the comments thus far:

Matt’s analogy holds no water. The implication that, like the mechanic and the “soft brakes,” I “didn’t actually check” the patients is silly. A correct analogy would be “I checked your brakes and there is still a half inch of pad left on the brakes. They’re ‘probably’ good for another 12 months. I can run them through a $2000 machine to give us a better idea of when they’ll fail, if you’d like. Oh, you demand the test or your lawyer will sue me for ‘$7.24 million’ if the brakes don’t last 12 months? In that case, sure.”
Let’s look at the electrician example. Is it the standard of electrical care for the electrician to rip out all of the wiring in your house “just to make sure” that there won’t be a fire? That’s what you’re implying. It might cost you or your home insurer tens of thousands of dollars, but otherwise, how can you be “sure” that a mouse didn’t eat through the wire right next to some insulation?
Oh, and I love your example about lawyers. Is it the standard of legal care to research each issue at law in each of the fifty states before creating a brief for a client? Perhaps laws in other states don’t have binding authority, but the added persuasive authority *might* be the difference between winning and losing a case. How much legal malpractice are you committing by failing to include other state rulings in writing *your* briefs?
Not so funny when you have to live by your own illogical assertions, is it?

VMS creates the novel concept that a physician must “eliminate everything the differential diagnosis” using a risk/benefit analysis. Please provide all of us physicians suffering from GOMER syndrome with a prospective legal and/or medical formula that we should use to escape liability. It’s fine to stand in front of a jury and retrospectively say that if “X” was only done, my client wouldn’t have had an injury. Funny how I never hear any lawyers prospectively saying “do Y and you won’t be sued.”
In the patient with foot pain, there was no bony point tenderness. He had a little red mark. He said that he couldn’t bear weight on his foot, but he walked into the ED. What is the likelihood of a fracture? Pretty much nil. What does an x-ray contribute to the ultimate diagnosis? Nothing but increased costs and the false security that I can convince a jury that nothing was there on the x-ray if the patient does happen to have a bad outcome. Let’s not stop there, though, Let’s assume there is a real risk of a fracture. Does x-ray “rule out” every fracture? Not at all. Probably need a bone scan to do that. Are you now saying that every patient with foot pain who alleges he can’t walk needs a bone scan? Just because we’ve ruled out a bony injury doesn’t mean that a ligamentous injury might not exist – that is on the differential and a ligamentous injury of the foot could cause lifelong pain. Now we really need an x-ray, a bone scan, and an MRI … on every injury, is that right?
That CT scan on the little old lady on Coumadin didn’t find a bleed, but we didn’t rule out everything else on the differential diagnosis for a headache. I must have been remiss. I probably should have gotten an MRI and MRA, to rule out vascular causes, right? How about a lumbar puncture to rule out pseudotumor cerebrii? I have probably caused more injury in radiating patients than I have saved finding needles in haystacks. Where’s the risk/benefit there?
Where do I get to stop working up my “differential diagnosis”
Your example of prostate cancer is particularly appropos to this discussion. Look up “Daniel Merenstein” on the internet and see how far your “no legitimate malpractice case” gets you with informed consent.

Max, I agree that “defensive medicine” is excessive or unreasonable. Now define those terms in the setting of medical care. Why is it OK to use a “more likely than not” (i.e. 51%) standard when expert witnesses testify about a “reasonable degree of medical certainty” but you state that we must be 99+ percent positive of a diagnosis when practicing in the emergency department?
You know as well as everyone that reads Walter’s blog that there is a list of “usual suspects” for every specialty who will say whatever they are told to say for a price. The “standard of care” is an unreproducible legal fallacy.

B.Rad is absolutely correct. We are already starting to see gaps in basic health coverage because so much money is going toward paying for testing that is both excessive and unreasonable that states can no longer adequately fund their Medicare and Medicaid programs.

It is unfortunate that so many people spread the misconception that just because some type of testing was done that medical care was somehow “better.” In many cases, additional testing adds nothing to the diagnostic “certainty” or “exclusion” of a condition and often contributes to real harm.
Where are doctors allowed to stop testing so that they can “be sure” that they are free from liability if they miss a diagnosis?
None of you that stands in line to throw stones at me has even tried to answer that question.

22 trinlayk 01.21.09 at 12:39 am

Years ago, I showed up at the ER at 2am Saturday morning with my young child in tow…

I was dehydrated from both ends, and had abdominal pain. I had driven myself there, (no choice) and wasn’t convulsed in pain… so they said “something viral” and sent me home. no tests, just something for the nausea, rehydration.

I had a ruptured appendix, and a few days later was in a different hospital (having collapsed at work) for over a week. When my own doctor saw me, he went to look at my ER records, fever, the cause of dehydration (violently ill in both directions), etc and was agast that they hadn’t done a white cell count. It would have shown the massive infection that had put me in Emergency surgery the night before. (Someone he talked to people at the first hospital so I never was billed for the ER visit…)

Apparently, my pain threshold is not “normal” PERHAPS in part because I’ve had chronic rather intense pain most of my life… and had to function anyway. Perhaps because I’m somehow “odd”…

so yeah, run the extra test, _I_ can’t tell you it might be my appendix, I’m not a doctor.

23 Max Kennerly 01.21.09 at 10:01 am

WhiteCoat,

I’d like to give you a snappy comeback to your remarks about “reasonable degree of medical certainty,” but I can’t. It’s a terribly ambiguous doctrine. Your analysis is a little bit off of the way the law is — a doctor can’t just say “more than 50%” to satisfy RDMC, there was recent case here in Pennsylvania holding just that — but I won’t deny that juries might see it that way sometimes.

To your other argument, “The “standard of care” is an unreproducible legal fallacy,” well, then what should we do? Total immunity for doctors? Specialized health courts? Keep in mind that post hoc reviews by physicians of medical malpractice verdicts have found that juries by and large get it right and, if anything, have a slight bias in favor of defense verdicts.

24 Max Kennerly 01.21.09 at 10:16 am

And, if of interest, I’ve replied (post duplicated on my blog, click my name).

Comments on this entry are closed.