When doctors do harm: understanding pre-test probability

What doctors don’t tell you: we’re scared. More than just scared of being sued, most of us are scared we’re going to miss something that will harm you. Scared this is going to be the one time out of a thousand that we miss something bad. We miss things all the time, but it’s usually things that don’t cause any damage. So, we test.

We over-order tests because the systems tells us to. You, our patients, tell us to. To over test, but not miss something bad, is rewarded. No one says, “good job not getting a CT scan on that person. By the way, she had appendicitis, which you missed, but at least you tried to not give her a big dose of radiation.” So we test.

Overlooked by patients, the best “test” your doctor has available is talking to you and examining you. Let’s introduce a medical term. Pre-test probability. Let’s say you see your doctor for chest pain. After she talks to you and examines you, but before she orders a test, she determines how likely it is that you are actually having a heart attack. We call that pre-test probability. After doing her assessment, maybe she decides that there is at most a 1% chance this is your heart. That means that you have a LOW pre-test probability. Then, she gets an EKG, which is normal.

Now, after examining you and doing an EKG, your doctor thinks you really are not having a heart attack. A low pre-test probability. This is where this concept becomes important: if she orders more tests, and one of those is positive, it is more likely that positive test represents a false positive. Tests are like doctors. They’re not perfect. Nearly every test can be falsely positive, or falsely negative. If you test a bunch of people for a disease they don’t have, some of them will test positive. But this is false. That’s just the nature of testing.

In the above example, if your doctor orders further testing, she is ordering it on someone who was very unlikely to have the disease to begin with. If you test positive, you may have tested positive for a condition you don’t really have. The test is positive, but it’s WRONG. It should be negative. The test is out to get us.  What’s the solution? Stop doing a bunch of tests on people who are very unlikely to have the condition you’re testing them for.

The problem with false-positive tests? They cause the ordering of more tests and the prescribing of more medications. Sometimes false positive tests result in referrals to specialists who have their own tool bag full of goodies for you. Talk about escalating health care expense! This whole series of unfortunate events could have been curbed by the doctor understanding, in the right setting, you’re ok with not testing. A little dialogue that says, “If you don’t think I need the test at this time, I’m fine with that.”

I have had patients basically demand tests. Like I said, I don’t want to miss anything. So, if pushed, I find myself obliging, to a degree. If I really think a test is harmful, I won’t do it. But, the art of medicine is splattered with gray. Most of the time, the best test I had was talking to my patient. The extra testing a patient requests, that I thought was unnecessary to begin with, rarely contributes to the care. It does, however, contribute to the patient’s bill.

Save some exposure to testing and hard-earned dough. Try having a discussion with your doctor about pre-test probability and such. That might just floor her. It might just save you some agony and money.

What unnecessary trips have you taken down the road of follow-up tests?