With payment reform at the gate, emergency medicine needs to find ways to promote the more judicious use of advanced imaging tests like CT scans, or other groups will make the rules for us.
Over the past decade, one of the most notable changes to the practice of emergency medicine in the United States has been the boom in CT imaging. Published papers have documented the trend using national data in abdominal pain, trauma, chest pain, and headache. A 2011 study in the Annals of Emergency Medicine documented a 330% increase in CT use across all indications from 3.2% in 1996 to 13.9% in 2007.1
Understanding this trend in our specialty is important today but will become much more important in the new future world of payment reform, where accountable care organizations, payment bundling, and pay-for-performance will soon become the new norm. While it is not entirely clear how these new models will impact how emergency physicians get paid, what is clear is that new scrutiny will be placed on expensive decisions such as ordering CT scans.
The real question is whether the increase in CT scanning has improved medical care or just contributes unjustifiably to higher costs. Within our specialty, arguments fall into two distinct camps.
The first cites clinical anecdotes touting the benefits of CT imaging through improved diagnostic accuracy, the prevention of costly hospital admissions through definitive risk-stratification, and the reduction of invasive diagnostic procedures where CT serves as a substitute.
Any of us, in fact, can probably recount stories of patients who seemed to have one condition clinically, but the CT result told a very different story. Remember that slam-dunk case of appendicitis which turned out to be just a simple gastro or even an important alternative diagnosis like terminal ileitis? Or how about that vague abdominal pain patient where you were on the fence about getting a scan, and ended up having a small duodenal perforation? Now think of how many people you’ve seen where a definitive work-up in the ED that included a CT and other tests could be discharged where even 10 years ago they would have been admitted for the same work-up. It is easy to argue that CT has revolutionized the practice of emergency medicine.
Another group in our specialty views increases in CT scanning as just plain excessive. They argue that the CT has replaced a good physical exam. People, they say, are often CT’d because we don’t want to spend enough time to make a nuanced clinical decision. A CT is the easy way out of more complex medical decision making.
Just as any of us can recount CT saves, most of us can also tell unfortunate stories of CTs that should have never happened in the first place. It is easy to remember many “low-value” CTs, where several hundred to thousands of dollars were added to someone’s ED bill for a test they may not have needed in the first place. How many times has this happened: a minor head-injured patient who was already over at radiology getting a head CT, and you or a colleague threw in a C-spine CT just to make absolutely sure to neck was OK, even though these wasn’t much C-spine tenderness? CT is not a benign test: it imparts a potentially carcinogenic dose of radiation, and when intravenous contrast is used, serious nephrotoxicity can result, and some patients can have serious complications from extravasation or allergic reactions.
Like EPs, radiologists are schizophrenic on the issue of CT. Guidelines from the American College of Radiology recommend against CTing low-risk patients for pulmonary embolism,2 while many hospitals’ radiology directors have been known to complain when the ED isn’t “giving” them enough CTs to read.
The broader public also falls in the two camps of CT-lovers and CT-haters. Health policy professionals – the haters – talk about the perverse incentive that encourages over-use of expensive tests such as CT scans and suggests that we are increasing CT as a way to pad payments. On the other hand, patients often themselves come to the ED wanting a CT: studies show that they are happier getting a CT than not getting one, regardless of whether they needed it or not.3
With two camps formed in our specialty and beyond, it is often difficult to know who’s right and who’s wrong. Clues to the potential right answer can come from close consideration of why we might be CTing everyone these days. Here are several potential reasons:
Defensive Medicine: Maybe we are scanning more because of the fear that we will be sued or that someone later will discover a finding on a radiographic test that will come back to haunt us. Missed diagnoses that never end up in the legal system become the discussion of M&M conferences, and even local sanctions against physicians. “Remember that patient” are three words that can cause even the most seasoned ED physician to go weak in the knees.
The Lemming Hypothesis: Maybe we are CT’ing more because we are just lemmings, following the lead of our respected hospital-based colleagues. Take trauma surgery as an example. Some call it the “pan-man-scan”, others call it a “trauma-gram”: a thorough, high-priced radiographic search for any occult diagnosis has become standard fare in our nation’s trauma bays. Hospitalists are another example. Admitting a patient with abdominal pain without a CT scan (or one done in the past week) has become heresy. The lemming hypothesis is that we too have drunk the Kool-Aid that this is good medicine and apply it to our ED practice.
Patient and referring physician expectations: On average, a CT’d patient is happier than a non-CT’d patient: a cozy CT scanner makes some patients feel like something was done to address their concerns. With more reliance on patient satisfaction scores to drive business decisions in emergency medicine, maybe we are bending to patient desires and ordering scans that we wouldn’t otherwise have done to boost our scores, and make sure we keep our contract. The same goes for the patient referred in by a PCP for an “emergent” CT with questionable indications. Just CT’ing them, even though you know it’s going to be negative, is easier than trying to justify avoiding the test.
Reading research reports: As more research is conducted surrounding the physical examination, and the reliance on clinical judgment, the more we see how imperfect it really is. Pneumonia is a great example of this. Research shows that individual clinical findings, such as crackles or rales, don’t really appreciably increase the likelihood a patient will have a radiographically-confirmed infiltrate.4 Research also reports concerning “miss rates” for serious radiographically-detectable diagnoses such as appendicitis and subarachnoid hemorrhage.
Bizarre, scary case reports: Just this June in Annals of Emergency Medicine, a case was reported of a patient who was thought to be low-risk and fell outside of all the validated clinical decision rules who accidentally received a D-Dimer, and got PE-protocol CT that found a large saddle embolus.5 This might be enough for the particularly risk-averse among us to lower our threshold for ordering the poorly sensitive D-Dimer for marginal patients and poo-poo PE decision rules, resulting in higher CT rates.
The Clinical Judgment Demise Theory: Here, the thinking goes: why do a careful physical exam if you are going to get a CT anyway? Because you’re not doing careful physical examinations, you lose the skill and need to rely more on CT. It’s a vicious cycle. ED residents today are not taught the same physical examination skills that they were 10 years ago because CT has replaced the need for it. With skills gone and the new generation not being taught, it will be next to impossible to resurrect the ability to differentiate people who really need CTs.
With all these potential forces driving us to order more and more CTs, what will (or can) we do when the era of cost-consciousness finally trickles its way into our bedside decisions?
One of the keys may lie in looking beyond anecdotes and believe in evidence-based clinical decision rules. The PERC is a great example of this, although it is not 100% perfect. With the PERC criteria, it is possible to differentiate patients who don’t need a D-Dimer in the first place to rule out PE, and so we don’t fall into the trap of needing to CT ultra low-risk people.
Also, relying on decision rules that are more specific will be important. Take the NEXUS criteria v. the Canadian C-Spine Rule for using cervical spine radiography. Both are sensitive, but the Canadian C-Spine Rule is considerably more specific (45% v. 37%).6 Commonly people fail NEXUS because of midline tenderness. In the Canadian C-Spine criteria, imaging can be deferred in people with midline tenderness who also have other low-risk criteria.
Making local or national guidelines surrounding the use and non-use of clinical decision rules may also provide the clinical justification needed to push back against others who are compelling us to order these tests. It may also serve as some protection or reassurance to providers who still feel compelled to order scans for the one-in-a-thousand chance that the patient may have something serious. It may just make us less defensive if we have a hospital policy or national policy telling us that it’s OK to not order a CT.
Even better, systems that integrate these decision rules at the point of care when ED physicians are ordering tests may be most effective in ensuring that people remember the right rules and implement them appropriately. People tend to mis-remember clinical decision rules, but may be less likely to make a mistake if rules are embedded into our electronic systems. But it is also important to consider the potential for unintended consequences. In a study of the implementation of the Canadian Head CT rule, imaging rates were paradoxically higher after the rule was disseminated.7
Another potential solution may be to give feedback to physicians on their imaging rates. With the advent of electronic health records, many systems allow easy retrieval of a physician’s ordering rates. If we knew how our ordering rates stacked up against our peers, maybe the outliers on the high end would think carefully about whether they needed to reconsider their approach to diagnostic imaging.
In the absence of reviving the physical examination, maybe another approach is to work at creating alternatives. The specificity of D-Dimer is only about 50%, but maybe there’s some other more specific test for PE just waiting to be discovered. Also, if we make other imaging tests the standard (such as ultrasound), this may be ultimately safer as patients are exposed to less radiation. A great example is the push towards using ultrasound in pediatric abdominal pain. Finally, our specialty needs more research to develop new clinical decision rules, and improve the specificity of the ones that exist.
We should use the recent flap over OP-15, Medicare’s head imaging measure, as a cautionary tale. OP-15 uses Medicare claims data to determine the appropriateness of head CTs for patients with a final diagnosis of non-specific headache based on whether they had specific exclusions. The measure may likely still be included in the 2012 outpatient prospective payment system on Hospital Compare, despite evidence from chart audits of ED patients that the measure was only reliable 1/3 of the time.8
With payment reform at our doorstep, emergency medicine needs to figure out real ways to promote the more judicious use of advanced imaging tests like CT scans, or other groups will make the rules for us. We need to reconcile the two camps – those who think CT is our panacea and those who think we are unjustifiably CTing the world – through open discussions, local and national guideline development, and additional evidence generation.
1. Kocher KE, Meurer WJ, Fazel R, Scott PA, Krumholz HM, Nallamothu BK.
National trends in use of computed tomography in the emergency
department. Ann Emerg Med. 2011 Nov;58(5):452-62.e3
2. Bettmann MA, White RD, Woodard PK, Abbara S, Atalay MK, Dorbala S, Haramati LB, Hendel RC, Martin ET 3rd, Ryan T, Steiner RM.ACR Appropriateness Criteria® acute chest pain--suspected pulmonary embolism. J Thorac Imaging. 2012 Mar;27(2):W28-31.
3. Baumann BM, Chen EH, Mills AM, Glaspey L, Thompson NM, Jones MK, Farner MC. Patient perceptions of computed tomographic imaging and their understanding of radiation risk and exposure. Ann Emerg Med. 2011 Jul;58(1):1-7.e2
4. Metlay JP, Kapoor WN, Fine MJ. Does this patient have community-acquired pneumonia? Diagnosing pneumonia by history and physical examination. JAMA 1997; 278:1440–5.
5. Hennessey A, Setyono DA, Lau WB, Fields JM. A patient with a large pulmonary saddle embolus eluding both clinical gestalt and validated decision rules. Ann Emerg Med. 2012 Jun;59(6):521-3
6. Stiell IG, Clement CM, McKnight RD, Brison R, Schull MJ, Rowe BH, Worthington JR, Eisenhauer MA, Cass D, Greenberg G, MacPhail I, Dreyer J, Lee JS, Bandiera G, Reardon M, Holroyd B, Lesiuk H, Wells GA The Canadian C-spine rule versus the NEXUS low-risk criteria in patients with trauma. N Engl J Med. 2003 Dec 25;349(26):2510-8.
7. Stiell IG, Clement CM, Grimshaw JM, Brison RJ, Rowe BH, Lee JS, Shah A, Brehaut J, Holroyd BR, Schull MJ, McKnight RD, Eisenhauer MA, Dreyer J, Letovsky E, Rutledge T, Macphail I, Ross S, Perry JJ, Ip U, Lesiuk H, Bennett C, Wells GA A prospective cluster-randomized trial to implement the Canadian CT Head Rule in emergency departments. CMAJ. 2010 Oct 5;182(14):1527-32.
8. Schuur JD, Brown MD, Cheung DS, Graff L 4th, Griffey RT, Hamedani AG, Kelly JJ, Klauer K, Phelan M, Sierzenski PR, Raja AS. Assessment of Medicare’s Imaging Efficiency Measure for Emergency Department Patients With Atraumatic Headache. Ann Emerg Med. 2012 Feb 23. [Epub ahead of print]