Disposition Driven Test Ordering/A Change in Paradigm

The most common complaints about Emergency Care is cost and inappropriate utilization.

The Sensible Test Ordering Practice in an Emergency Department report was published by a pathology magazine, and provides the STOP methodology to reduce the # of coagulation and glucose testing most of which do not affect outcome and or treatment. A major problem seen from the viewpoint of these authors in the lab (pathologists) is the classic rationale for test ordering, which consists of ordering packages of tests that (if negative) do not affect outcome. Ordinarily this happens for the sake of consultants, completeness, diagnostic uncertainty, usual practice, training dogma, and liability concerns.

Analyses of this sort don’t get the principles that truly apply in emergency medicine, and those principles are based fundamentally in clinical judgment.  Is it the responsibility of the ED doc to save money, or potentially save the life and health of the patient?  When standing in the pathology lab, any negative test appears to have been unnecessary. 

On initial presentation, the ED doc is confronted with a unique chance to impact the care of any patient.  If he/she has seen a hundred patients and cast the “wide net” of a “routine” set of lab tests, receiving, say one or two with a positive result that impacted life, but 98% or 99% are unnecessary, is this type of pathologist’s perspective really of value?  Ask the patients who were impacted.  No wonder that getting out of the “routine order mindset” is a daunting task. The article shows how difficult and complex getting rid of a “low hanging fruit” test is. Low hanging fruit is fruit because sometimes, just enough times maybe, it makes a difference. But is testing practice really a result of mindless habit?  Or could it be experience and concern for the patient driving such testing?

It would be “politically correct” to suggest that the Emergency Department Provider change focus and order only tests that directly affect patient safety and disposition. How realistic is that?  In truth the ED doc, when first seeing a patient stands at the gate of a large city, looking for a place to live.  Since real estate is specific, finding the one that fits requires casting a larger net than other providers do.  Of course, understanding red flags and high-risk situations can deliver the provider to a quick risk assessment of most problems. But not every important flag is red, nor are all situations high-risk today, that will become high-risk tomorrow. Ask any plaintiffs lawyer. Then factor in the true cost of limiting testing from the pathology suite.

While the provider might be counseled to focus on the 1 or 2 tests that clinch the diagnosis, determine the disposition, and send the patient down the proper pathway to health, this paradigm ignores entirely the patient whose disease process would be found through other testes.  And certainly, limiting tests to the “money tests” will decrease costs on the set of patients for whom they are limited, but not for the system in which misses will naturally increase. 

An example of a restricted test protocol would be:

The provider would be given:

1. the vital signs with pulse oximetry
2. EKG if indicated
3. The option to order 1 or 2 tests maximum
4. Upon receiving the results of those tests the provider can discharge, admit, start treatment protocols.
5. If the patient is critically ill or immediately falls into a specific treatment protocol, these can be initiated. The treatment protocols include all the tests anybody would want.
6. Time hopefully would be saved due to rapid disposition of the majority of patients who were concerned about very specific problems.
7. Likely however, time will be lost in total because there will be significant number of doctor-patient combinations where sufficiently clarified status is not obtained by limited testing, and repeat, after repeat, after repeat keeps the patient there for hours (and hours, and hours)
8. On the one hand, when successful, down-stream costs with endless consultations may be avoided. On the other, the costs will HUGELY accelerated in those cases missed.
9. The ED is a unique opportunity, especially for the many who have no other access to care.

Examples:

1. Patient has chest pain. EKG + for Acute MI. Disposition would cath lab and MI protocol. (The authors would not check the INR, if the patient is on warfarin, and if the INR=50, that patient will likely not exit the Cath lab).
2. Patient has severe abdominal pain. CT-neg. Clinical reevaluation then discharge or admit. That would be fine if inter-observer reliability was present. It often is not (Br J Radiol. 2012 Sep; 85(1017):e596-602. doi: 10.1259/bjr/95400367.)
3. Patient has head injury. CT-neg.  Clinical reevaluation and disposition. (but you missed the leaking aneurysm)
4. Patient has sore throat.   Treat.  Reevaluate if no improvement. (Did you catch the epiglottitis?)
5. Patient has severe cough and fever. CXR-right lower lobe infiltrate. Reevaluate clinically and discharge with appropriate prescription or admit with pneumonia protocol. (ah…yes, forgot the CT scan, and the patient returns dead after multiple pulmonary emboli)
6. Patient with weakness. EKG shows u-waves. Get electrolyte package and treat. (ah sure, treat for hypokalemia, but did you remember that u-waves can be sign in subarachnoid or intra-cerebral hemorrhage)
7. Patient with low-risk chest pain. Troponin-neg, EKG normal. Repeat troponin 2 hours. Make disposition. (that’s the one whose pericarditis you missed because you did not do a CXR)
8. Patient with possible stroke. Depending on distribution of symptoms get MRI or CT. Start treatment protocols if indicated (don’t forget the coag profile!)
9. One of the most important clinical tools one can develop happens when one makes oneself record (write-down) the results anticipated for any test. This was taught to one of us by an illustrious grandfather of medicine shortly before he died.  The lesson was that by doing so throughout a career, dependency on tests would decrease. One’s clinical insight would gradually increase, and the path to a correct diagnosis would be much quicker.  It works. And it is a much better approach than simply STOPing the test net.
10. The approaches suggested from the pathologist Labe are intended to save a lot of upfront costs, admissions, and hopefully downstream costs. Do they really save time???  Save lives, save morbidity? Theoretically they should. But the misses are simply left out of the picture. And those have extreme costs to both the patient, the doctor, and the system.  Bottom line: take the time to hit a “home run” rather than run the count to 3 balls, 2 strikes on each patient before making a disposition. Home runs require hitting all the bases, you know.