As an epidemiologist with interests in many clinical questions, I have naturally gotten drawn into the realm of evidence-based medicine. If you ask almost any clinician, he or she will state that they practice evidence-based medicine. But I wonder if this is actually true? In other words, what is the evidence that clinicians actually practice evidence-based medicine? I am doing a project on this now in the contexts of infectious disease practice, and in pain medicine and related specialties. To begin with, though, I have observed a dearth of research on this topic.

“In My Experience Medicine”

I’ve observed that in the surgical specialties especially, it’s very common to hear “in my experience I’ve found…” type reasoning. This has both strengths and weaknesses that are too extensive to note here. Consider the following scenario, though.

An acquaintance at the gym had hip replacement surgery 3 weeks ago. There are two major approaches to this surgery: the traditional approach, which involves a long incision laterally, and the newer anterior approach, which involves a short incision on the front of the hip area. This does not involve much muscle destruction, and the recovery times are far shorter than with the lateral approach. When I asked this person about his experience, I found that he had had the lateral approach done, not because of any clinical indication, but because his surgeon said “in my experience, it works.” The evidence is clear, though. In the hands of skilled clinicians, the outcomes from these two approaches after 1 year are identical, yet the anterior approach–the newer approach–involves less pain, less time off of normal activities, and shorter rehab. This is an example of “in my experience” medicine. To be fair, since the anterior approach is newer, there is a learning curve, and some retraining that is involved. Yet by saying that this is how he does things, this surgeon was impacting the patient’s recovery. This “in my experience medicine” cuts both ways. On the one hand, it may have sub-optimal outcomes, yet we don’t want to completely discount the importance of personal experience and the patient-physician relationship.

There are reasons to suspect “in my experience medicine”. Cognitively, there is selective memory, so that clinicians remember their patients who have done well with a given practice. This is backed up firmly by multiple studies. This introduces bias. Just as significant is the fact that that there is a dearth of data on whether or not clinicians actually practice EBM. This is what is getting me into much intriguing territory.

Consider, too, the case of spine surgery. There are a multiplicity of approaches to and indications for surgery. We can take the case of spinal stenosis, or narrowing of the spinal canal due to osteoarthritic changes, or of the foramina–areas through which spinal nerve roots exit the canal to the periphery.

Decompression surgery involves removing some of the disc that may exert pressure on the nerve roots, and “shaving” or removing some of the areas affected by the osteoarthritis, thereby minimizing nerve root compression, and therefore pain and neurologic deficit. Another approach that is sometimes followed is decompression plus fusion (which can be accomplished in a host of different ways.) Despite some definitive studies–some randomized clinical trials–that show that in the absence of instability, the outcomes from the less extensive decompression surgery are equal to or superior to fusion surgery. The latter involves much greater risk of post infection, and involves greater pain, more blood loss, more anesthesia time,  and longer recovery. Despite this, fusions are among the fastest growing surgical procedures in the US. Why?

The cynical view holds that surgeons like to operate, and that medicine in general and surgery in particular in the US is a business, with psychological and economic incentives to enhance revenue. Surgeons (neurosurgeons, orthopedic surgeons, with or without spine fellowships) are paid far more for the more procedure. In addition, fellowship trained surgeons (1-2 years post neurosurgery or orthopaedic surgery residencies) are well trained in “complex spine operations” which include fusion. Thus, this begins to invoke the “in my experience” scenario, which may operate to the detriment of patients if they are not among those few for whom fusion is really indicated.

So…we don’t really know how frequently surgeons and interventionists actually practice evidence based medicine. Yes, experience counts for something–I cannot and will not deny this. However, there are many other things at work here.

I think–I argue–that medicine is undergoing some growing pains moving from a fairly subjective base of evidence that is based upon experience and case series, to the 30 year old evidence based medicine paradigm. And, on top of that is the fact (ironically based upon “in my experience” on my part) that clinicians do not have sufficient training in critically reading and understanding clinical research and clinical trials, with a strong ability to identify a study’s strengths and weaknesses, which include biases. Put this together with the tremendous time demands of physicians, and we find that much knowledge is based upon what others say about the existing studies in various formats–and in reading abstracts, and the proliferation of publications that summarize the findings of original research, with varying accuracy.

So back to my gym acquaintance. He is doing remarkably well 3 weeks post op. I asked him if he had consulted additional surgeons before making his decision about he operation, and the answer was the all too common “no” because of his trust in the surgeon, and in his primary care physician who was the source of the referral. Yet we are in Seattle, where there is no shortage of excellent orthopedic surgeons, including many world leaders in hip replacement surgery at the University of Washington. Without going off on a tangent here at the end, was it not the responsibility of the referring physician to suggest speaking to–“interviewing” as I call it, at least one other surgeon prior to the decision of how to proceed? Was this person served best by speaking to only one surgeon–the one who performed the more extensive surgery?

As I said, he is doing well and recovering nicely. In his case, the lateral approach went fine. But epidemiology, and all clinical studies as well, are concerned with populations. Would selecting the less invasive approach make a difference in a population of 1,000 patients and surgeries? Probably yes–though the evidence may be lacking to back up this contention.

In the end, we need a solid base of evidence on whether evidence-based guidelines are followed, and whether they make a difference to patient outcomes.