How Reliable Are Those Test Results?

When you go to your doctor with one complaint or another, the most common response is, “Let’s run some tests.” You and your physician both expect that with those results in hand, you’ll have a better idea of what’s going on and can work toward a solution.

Of course, anyone who’s seen an episode of House knows it’s not always that simple. But even when your symptoms aren’t due to some obscure condition, there’s a complicating factor: the test results may not give the right answer.

How Things Can Go Wrong

We need to get into statistics again, if only briefly. There are four possible outcomes of a test:

  1. You have the condition, and the test says that you do. This is known as the sensitivity of the test.
  2. You have the condition, but the test says that you don’t. This is called a false negative (in statistics lingo a “Type 2” error).
  3. You don’t have the condition, but the test says that you do. This is a false positive (a “Type 1” error).
  4. You don’t have the condition, and the test says you don’t. This is the specificity of the test.

A Type 2 error leads to not getting treated for something you have, while a Type 1 error leads to unnecessary treatment for a condition you don’t have—or, more importantly, instead of the condition you do have. (If you’d like to wander into the weeds of statistical methods, here’s a good explanation of the errors from Mark Liberman, a professor at the University of Pennsylvania.)

Tests can go wrong for any number of reasons. They can be misread or misinterpreted, for example on an x-ray or sonogram. Or they can be based on uncertain data, as in PSA testing. What caught my eye was some recent news about kidney damage.

Good Tests Can Lead to Bad Treatments

Many medications can cause damage to your kidneys—pain relievers, antibiotics, and drugs for heartburn among them. Certain health conditions can injure your kidneys as well, chiefly high blood pressure and diabetes. Kidney injury is usually diagnosed by measuring levels of a protein called creatinine; more particularly, changes in the level of creatinine.

A study published in last week’s Clinical Journal of the American Society of Nephrology showed that blood tests for levels of creatinine can result in a false diagnosis of kidney damage (specifically, acute kidney injury) an average of 8% of the time. That may not sound like much of an error, but in people with higher levels of creatinine to begin with, the error rate rose to 30%.

The implication is that your physician may be making treatment decisions based on these test results. The most drastic decision is to begin dialysis, but he/she may also withhold an effective treatment because of known effects on the kidneys, such as the pain relievers I mentioned earlier.

For you, here’s what this means. Whenever you see test results that are out of the normal range, dig further. Before beginning any treatment based on those results, ask your doc what the error rate is and how that rate affects the decision. (They may be surprised to hear that question, and in fact may not even know.) It’s part of being a smart consumer.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s