“How much unnecessary testing goes on in the ER?”

by Walter Olson on November 5, 2009

“Plenty, if you ask the people most familiar with the situation, the emergency physicians themselves.” [KevinMD, Emergency Physicians Monthly, White Coat, WSJ Law Blog] Relatedly: “Just to be sure: an ER slippery slope” [MedRants, WhiteCoat] And yesterday, from the AP.


1 Dr. T 11.05.09 at 7:06 pm

Overtesting in the ER has occurred for decades. Some is due to fears about malpractice (eg: back pain and chest pain workups). Some is done because the ER physicians are so busy that they have too little time per patient, and lab tests and imaging studies are substituted for a more detailed exam. Some overtesting is done because of bad patient transfer policies such as psychiatric units not accepting patients unless drug screens are negative (even when the patient is blatantly psychotic). And, some overtesting is done to reinforce an already suspected diagnosis. Naturally, almost all this overtesting is blamed on malpractice fears. My very rough estimates of overtesting: 40% malpractice fears, 25% too little time, 10% bad policies, 25% reinforce diagnoses.

2 Mark 11.05.09 at 11:58 pm

Dr. T, your recent posts are concerning because they allude to you being a medical doctor. But you can’t even recognize the difference between IBS and IBD?

That actually segways well into my comment: the primary reason for overtesting is physician ignorance (in many cases due to limited intellectual capacity, but most certainly due to deficiencies in our medical education system) and a lack of time spent investigating the issue with the patient. Typically doctor’s make haphazard decisions based on tests with low sensitivity/specificity instead of interviewing the patient to establish a diagnosis.

3 throckmorton 11.06.09 at 10:13 am


IBS and IBD are great examples of defensive medicine in practice. Patients often come in to the Ed in the middle of the night with abdominal pain. Often these patients clain to have IBS or IBD and want narcotics. With the exam you can with a relatively high predictive value determine that they do not have an acute abdomen. Here is the problem. Your predictive value is 99.9%, that means that with your history and exam you can still me 1 in 1000. So you get a ct of the abdomen, since it can also be renal you get a ct ivp at the same time and if female a pelvic US. Now you spent thousands of dollars and are a bit more certain they do not have an acute abdomen.

Oh, here is a little pearl. When you take a patients history, they do not always tell you the truth especially if they have a chronic pain condition. They tend to say whatever they want to get pain medication. IBD patients do this alot. Also, IBD can still have other things go wrong and get appenditis, diverticulitis and ovarian torsion just like anyone else.

As to physian ignornace, could you please help us with your alogrithm that we can use to identify IBD and IBS when patients come in to the ED in the middle of the night claiming to have severe abdominal pain. This way we could post it and use it as the “Standard of Care” so that a long as you followed it you would not be sued. Please be sure to show the sensitivity, specificity, positive and negative predictive values of each of the diagnostic instruments.

4 Mark 11.06.09 at 10:57 am


I agree with the majority of your post, except that you might consider this testing unnecessary :)

The IBS vs. IBD reference was not the segway into my comment… it was the fact that Dr. T cannot discern between the two. Search through the comments on this blog.

As to an IBD patient lying to obtain pain medication, if they already have a confirmed diagnosis of IBD why wouldn’t you administer pain medication in addition to antiinflammatories? (unless of course there was a concern about the medication disturbing motility or a concern about dependency). Maybe the patient wouldn’t have to lie if the doctors weren’t… read my previous post :)

I will concede that the lack of comprehensive, accessible medical records contributes to the problem and does make it difficult to determine, for example, whether or not the patient has a history of substance abuse, inflammatory disease, etc. So I suppose I can’t complain TOO much about doctors until that system is available.

But to be clear, none of that testing is unnecessary. It’s just the system is so broken they can’t help but rinse, repeat over and over again.

5 Dr. T 11.06.09 at 10:51 pm


What are you talking about? I’m a clinical pathologist who wrote a comment about the headline “How much unnecessary testing goes on in the ER,” and you’re ranting about my inability to distinguish among bowel problems. I don’t know who you are. I never read any of your online articles or blogs. I simply posted about ER testing problems. If my post is incorrect, do you have better numbers? Or are you sniping at me because you feel guilty about overtesting?

Your rant about me appears to be linked to a June post about Accutane where I used IBD instead of IBS. The Overlawyered headline mentioned IBD, but the stories I had read about Accutane discussed irritable bowel symptoms. My misuse of the abbreviation was inadvertant and due to fatigue, not ignorance. Besides, there were far more Accutane IBS cases than IBD. Crohn’s disease or ulcerative colitis would be uncommon sequelae to any drug, including Accutane.

So instead of assuming I’m ignorant because I made a posting error three months ago, how about addressing the content of the current post. I’ve looked at testing in three ERs and read many articles on ER testing. My percentage allocations are crude but are based on discussions with full-time ER physicians. At a VA hospital I worked at a few years ago, ER physicians came to me to find a way out of the “bad policy” based testing and to document the “too little time” testing. (We had already established those numbers. At the VA the “bad policy” testing was higher than average, but the malpractice fear testing was lower.) The ER physicians were trying to show that the costs of adding staff would be partly offset by reduced costs of testing.

6 throckmorton 11.06.09 at 10:54 pm


How do you confirm their diagnosis of IBD. This is for the most part a diagnosis of exclusion. Further, do you have any idea how many people come in to the ED with a history of “kidney stones”, IBD and the “worst fibromyalgia” just to get thier fix of narcotics? If they truly have IBD, then they should be followed by their doctor and should not be the ones coming into the ED in the middle of the night.

Also, Matt, say a pateint has a history of IBD and you have records that prove it. How do you know that this pain is not from appendicitis or a ruptured tick? Just because they have IBD does not mean that something else could be wrong.

7 Mark 11.21.09 at 10:48 pm

Dr T:

You got it! Yes, I was referring to your post regarding Accutane. Your follow up comments here are less than reassuring, as you are once again confusing IBS and IBD. IBS is the diagnosis of exclusion. I’m not sure that fatigue is the culprit here.

What’s even more disturbing is that you have AGAIN

8 Mark 11.21.09 at 10:57 pm

*hit send accidentally*

… confused IBS and IBD. A quick google/pubmed search will illustrate that virtually all emphasis is on IBD rather than IBS. Where have you heard of Accutane causing rampant IBS (or at least moreso than IBD)? My point is that you have no idea what you’re talking about, and with your help, I’ve substantiated that claim. I can only imagine the number of lives you’ve ruined during your career while “fatigued” :)


My previous comment was actually directed more at you than Dr. T, but alas I hit send too soon. IBS is the diagnosis of exclusion. Histophatology showing cryptitis confirms an IBD diagnosis. As to patients coming in begging for pain killers, I refer you to my previous comments regarding comprehensive medical records.

If the patient had IBD, you would run the test regardless!! How would that ever constitute over testing?

Comments on this entry are closed.