As physicians, we are charged by Hippocrates to "first do no harm." However, when it comes to testing and evaluating disease, many physicians are missing the mark. While there is a multiplicity of reasons for this behavior -- from lack of knowledge to fear of litigation -- many physicians routinely order unnecessary tests, simply because its always been done that way. This week, an article in JAMA Internal Medicine examined the downstream effects of ordering unnecessary tests. The authors of the paper entitled "Prevalence and Cost of Care Cascades After Low-Value Preoperative Electrocardiogram for Cataract Surgery in Fee-for-Service Medicare Beneficiaries" found that in the particular case of preoperative ECG screening for cataract surgery, that the "care cascade" after these tests can be quite costly. As a clinician, I really like this article. It tells us what many of us already know -- ordering indiscriminate low yield tests often produces the need for other unnecessary tests and procedures that can actually result in complications and patient harm -- not to mention the unnecessary costs. When I was a medical student and a resident I can remember my mentors on general medicine asking me and my colleagues, "Why are you ordering that test?" If we could not provide a logical reason for why the test was being ordered and, most importantly, how it would change our management, the test was cancelled. One particular attending physician likened diagnostic testing to a fork in the road. If you order a test, the result should tell you whether or not to take the left side of the fork or the right. If you are going to randomly take either path, then the test is not necessary. Too often, doctors order tests out of fear or ignorance rather than based on science. It's time for doctors to stand up and be doctors: Use their training and their clinical judgment -- in conjunction with appropriate testing -- to decide WHEN and how to test. Bayesian statistics tell us that ordering a diagnostic test (such as an EKG for no reason before cataract surgery) will be very low yield. A good reason to order a test is when you have an intermediate pre-test probability of disease. By that same line of thinking: If you have a low pre-test probability, you do not need the test. If you have a high pre-test probability, you should treat, not test. As you can probably tell, this is a real hot button for me. As a cardiologist, I have seen too many examples of unnecessary stress tests, leading to false positives, leading to un needed heart caths, and in some cases negative outcomes due to complications of the procedure. I have seen indiscriminate use of lab testing in ERs where a patient, regardless of history or physical exam, gets a Troponin sent to the lab (even in a patient with no symptoms of CAD, no risk factors, and no related chief complaint). These, more often than not, come back indeterminate and create the need for another unnecessary and costly work up. So, in conclusion, as physician leaders we must THINK before we TEST. Let's harken back to Shakespeare and Hamlet's "nunnery scene" -- to test or not to test, that is the question ... Source
