Thursday, May 26, 2011

A CT Scan foregone

Noting recent research indicating that many cancer screenings are of dubious value, Ezra Klein opines:
as long as doctors are telling scared and uncertain patients that they need to get screened, they’re getting screened. The moment they stop telling patients to get screened, screenings will plummet. In health care, doctors are really the relevant decision-makers. And right now, they don’t have the evidence to make good decisions nor the incentives to make cost-effective decisions.
That reminds me of a tale I heard told recently by a nurse midwife with many decades of experience working with doctors in hospitals -- and often tussling with them to forestall what she regards as unnecessary care, e.g., Caesarians.  She recalled a trip she took with her grandson to the emergency room after he fell out of a (stationary) truck bed, hitting his head on her driveway and getting a good-sized goose egg. She was deeply impressed with the exam that the ER physician conducted to rule out internal bleeding or risk of brain damage: her trained eye could appreciate how thorough it was. This doctor wanted  to avoid an MRI CT Scan* if all indicators were good, as they were.  That's quite unusual.

I asked this person whether she would have pushed back if the doctor had ordered a CT Scan. Again, she is a healthcare professional committed to avoiding unnecessary care, and often willing to be confrontational about it. She said that she would not.  "You don't mess around when the brain is involved."

The point is not that the doctor would have been wrong to order an MRI. It's that he had no incentive not to, and that his eschewing it required an unusual orientation for an American doctor -- a willingness to avoid a large blast of radiation to a young head if there was no benefit to be gained by it. It takes a strong will to ignore the perception of risk and resist what in the U.S. is the default option -- to prescribe any test that may rule out highly unlikely outcomes.

To change American doctors' bias toward overutilization will require major changes in incentives, education and culture.

* Thanks to commenter pmpm: I checked with my wife, who was present when the story was told, and the test in question was indeed a CT Scan.  I guess I've always mentally interchanged the two.


  1. The dose of radiation from an MRI is zero.

    CT scans do use radiation, and are therefore best avoided in young children.

    CT scans are used frequently in acute trauma assessments in ER. MRIs are more often ordered by specialists, and add little or nothing to what is shown by CT images in terms of ruling out fractures or bleeding from a blow to head. The main problem with MRI in pediatrics is that little kids have a hard time holding still in the scanner and so need to be sedated, which means having an anesthetist present. Most MRI scans of the head ordered in small children are done to work up seizures and/or developmental delay.

    Your concept of the incentives at work when doctors order tests that are almost certainly unnecessary is spot-on.

  2. Thanks, pmpm: the storyteller did say "CT Scan," per added footnote above.