Monday, February 13, 2006

The two gaps between evidence and clinical practice

Dr. M.R. Tonelli in his 2001 article entitled " The Limits of Evidence Based Medicine" speaks of two gaps or two aspects of one gap between what is referred to by the body of evidence based medicine as "empirical" evidence and the clinical practice of medicine.

The EMB spokesmen acknowledge one gap, the gap requiring considerations of the patient and the professional values.

The second gap exists due to the fact that the evidence is not directly applicable to the individual patient.The data do not "directly answer the primary clinical question of what is best for the patient at hand."

That atorvastatin might in a clinical trial reduce the risk of stroke over a given period of observation by half does not mean Mr.Jones should necessarily take atorvastatin even if his clinical profile as defined by the relatively few parameters which characterized the study group seem to fit fairly closely with his.

The observations from that trial, for example, are only one of several imputs that should be considered in a clinical decision. Tonelli lists the following;

"empirical evidence,experiential evidence, physiologic principles, patient and professional values, and system features. The relative weight given to each of these areas is not predetermined, but varies from case to case "

He does not believe that empirical evidence (which is basically various types of clinical research) should automatically or necessarily supersede clinical experience and physiologic rationale.

Tonelli asserts that EBM has made a conceptual error by grouping knowledge that we derive from clinical experience and pathophysiologic reasoning under the heading of "evidence" and to make matters worse we developed a hierarchy that assigned those forms of knowledge to the bottom tier. Those forms of knowledge differ in kind from empirical data and should be used as complementary to empirical knowledge and are part of the tool set we need to use to bridge the gap.

Since I grew up in a medical era during which pathophysiologic reasoning was emphasized, I am sympathetic to the author's view but I am still processing what he said and my mental jury is still deliberating. The broad definition of EBM is to integrate the best available evidence with clinical expertise and patient values and I think clinical experience and physiologic rational could come to play in the "clinical expertise" category.

However,in the one version of the EBM bible (Evidence -Based Medicine,second edition, by Sackett et al,2000) we find a table of "level of evidence" in which the lowest level is " expert opinion ...based on physiology,bench research or first principles." This quote from scripture tends to support Tonelli's characterization of EBM hierarchy as denigrating physiologic reasoning and not my ad hoc attempt to salvage it under the heading of clinical expertise thus ending my short lived attempt to consider his thoughts as a straw man argument.

1 comment:

Anonymous said...

Dr. Gaulte,
It was a pleasure to read your comments on "Evidence Based Medicine"... The evidence is only as good as the reporting.. How many statin patients actually suffer rhabdomyolysis??? Well, the resident ER doc (most inexperienced = easier to "pick on") has to first be able to spell it, pronounce it, and explain it to their patient, before they can Dx and Tx it. Will a report be sent to the patient's PCP? Will a report be sent to the FDA? When healthcare providers are relying on 22 a y/o lab tech to tell them their patient has a UTI, the art of medicine is being lost. A bunch of labs are ordered before the patient is ever interveiwed or examined by a M.D., N.P., or P.A. Now, the quality of care for an E.R. visit is only as good as the triage. The treating healthcare provider would need to order and understand that calf tenderness on palpation, Hx, and a simple U/A dipstick AND microscopic U/A could yield a "definitive" Dx of rhabdomyolysis before a urine myoglobin or CK-MM ever came back from the lab. Hmm, if the dipstick was + for "blood" but, the micro was - for RBCs, it's not to hard to figure out. Missing it can have drastic results. Medical training no longer empasizes understanding pathophysiology as the "golden key" to Dx and Tx patients. Having graduated a Physician Assistant program in 2000, I was very fortunate to receive an excellent education. But, you get out of it what you put in. Learning how to pass a written multiple choice test with a "C" and "earning" a diploma, does not equate to understanding the body and medicine. Why are news articles just now warning of the CV risks of first generation sulfonylureas when the mechanism of action has been known since the 1970s? How does it effect the pancreas? How would we expect it to effect the heart? Sometimes the information is plain as day if we read the PDR pamplet prior to prescribing. Look at the PDR report for 2001 on Vioxx, and read the numbers. As a healthcare provider and a patient, I have seen kidney infections Dx as "anxiety" because the urine results were not looked at,....Tx for migraine when the "C.C." was abd pain and the KUB found a stone,..and "C.C." was the nurses medical "diagnosis" instead of the patient's "words". If we cannot perform an accurate and thorough Hx and PE, how can we accurately Dx and provide appropriate Tx? What happens to our patient????