Recently, the makers of the Oncotype DX prostate cancer test presented results of a large study demonstrating that their test can help men decide whether their prostate cancer carries a low enough risk of progression to forgo surgical or radiation therapy, two treatments that typically eradicate prostate cancers but also cause most men to experience impotence and incontinence.

Lacking such a test, many men have felt compelled to receive these aggressive treatments even though they know that most men in their position—with low grade cancer localized to the prostate—will not experience aggressive, metastatic disease.  Low grade tumors—what are called Gleason 6 tumors based on how they look under a microscope—do not usually cause fatal illness.

But there are a couple problems with our current staging system, at least in the minds of most patients.  It’s phrases like “don’t usually cause fatal illness.”  Those are troublingly vague words for someone who has just found out he has a cancer diagnosis.  It must mean that some of those tumors turn nasty.

Enter the Oncotype test.  If the test is as good as experts hope it to be (warning: the results have not passed peer review muster yet), if the test better identifies safe tumors that have almost no chance of spreading, then men should be able to avoid those nasty treatments.  And they should also be able to avoid the costs of being monitored every six months with prostate blood tests and biopsies.

But will human psychology interfere with optimal use of the Oncotype test?  Research that I have conducted with Angie Fagerlin has demonstrated that cancer diagnoses often create an action imperative, whereby people are willing to experience net harm in order to rid themselves of malignant tumors.  Here is how I described that line of research in my book Critical Decisions.

I asked people to imagine that [they had a slow-growing cancer, and that a strategy of] watchful waiting led to only a 5% chance of death from the cancer whereas surgery, which would cure the cancer, carried a 10% risk of death.  In this case, a substantial majority of people said they wanted the surgery, preferring death from activity to death from inactivity.  “Get it out of me,” they said. “Better to go out fighting than to wait for bad things to happen.”  The thought of untreated cancer bothered these people so much that they preferred taking action, even when that action was more likely to harm them.

Based on this psychological imperative to take action, we face the possibility that men will receive the Oncotype test—at almost $4,000 per patient—and they will still choose either active treatments (surgery or radiation) or active surveillance (with all those follow-up tests and biopsies).

Here is a potential solution, assuming that the Oncotype or some other test is proven to strongly predict tumor behavior.

1.  If patients choose to receive the Oncotype test

2. And the test shows them to be in the lowest risk group (whatever that means)

3. Then insurance will not pay for aggressive follow-up

4. And insurance will not pay for surgical or radiation treatment

If we want to control healthcare costs, we need to hold patients financially accountable for unnecessary medical expenses.  If we allow men and their physicians to ignore expensive new predictive tests, we should not be forced to pick up the tab.

Peter Ubel is a physician and behavioral scientist who blogs at his self-titled site, Peter Ubel and can be reached on Twitter @PeterUbel.  He is the author of Critical Decisions: How You and Your Doctor Can Make the Right Medical Choices Together.