In The News

Tale of the costly test: An Ottawa doctor's startling discovery has led him to question whether expensive diagnostic tests lead to better patient care

By Tom Spears
The Ottawa Citizen
Friday, August 5, 2005

Medicine sees new diagnostic tests every week, bringing more lab work, more costs, but not always more help to the patient. Now an Ottawa doctor and researcher is asking: Who's testing the tests?

Shawn Aaron is a lung specialist at the Ottawa Hospital Research Institute, and his tale of one expensive test that led nowhere is being rushed into publication tomorrow in the British medical journal The Lancet.

Dr. Aaron led a team of 20 that developed a test for diagnosing lung infections that were plaguing people with cystic fibrosis.

It seemed like a great idea on paper. These patients develop "really horrible" lung infections, resistant to many antibiotics, and treatable with different antibiotics than the ones that may work in another person's case. Usually a blend of two or more drugs works best.

So Dr. Aaron reasoned: Why not test each bacterial sample against several drugs at once, instead of one drug at a time?

Normally, a hospital lab might test the bacteria from a patient against seven or eight single drugs, he says. In his study, he had labs running 96 tests on each bacterial sample, and each of the 96 involved at least two drugs, sometimes three.

This costs $300 per patient -- about 10 times the cost of a conventional test.

And it did no good at all.

The complex tests did give more information to doctors treating a cystic fibrosis patient, Dr. Aaron said. But in a double-blind study -- the kind where neither doctors, nor patients, know what test has been used -- there was no difference in the patients' health.

New drugs must always be evaluated in this double-blind method, he notes. Why not diagnostic tests?

"There are hundreds of new tests that are adopted every year in clinical practice, but nobody knows if these tests actually change patient outcomes" -- in other words, whether they improve anyone's health.

"The tests may simply be better than the existing tests in giving more information to the clinician (doctor), but the question is: Does adopting the test really change outcomes? And ours is one of the only trials in the last 20 years to actually look at that."

His 4 1/2-year study followed 132 patients in Canada and Australia, working with the universities of Ottawa and Sydney and the Ottawa Hospital. It checked patients for shortness of breath, for lung function, for the length of time between bad attacks of the infection, and other "very hard" data.

"And we didn't see a difference."

In this case, the complex tests used up money but did no harm to the patients. But he says there are other tests that may be needlessly invasive -- opening up patients and injecting dyes or probes into them, for instance.

"We have a lot of invasive tests that we use in medicine, that nobody has ever evaluated rigorously.

"Whereas if these were drugs, they would have been evaluated in clinical trials. It is something that I think has been neglected. ... New tests come along and supplant older tests every week.

"They're always more expensive and more intricate, but the question is, do they improve patients?"

Note: Reprinted with permission of the Ottawa Citizen

Back to news