Diagnosing OSA: Keep it simple

Tuesday, February 28, 2006

Is it any wonder that the number of beds in America's sleep labs grew by 21% in 2005 and should jump another 31%, according to a Wachovia Securities study released in January? It shouldn't be, not after CMS balked at common sense last year and declined to reimburse suppliers for sleep-disordered breathing screening in the home.
If you asked CMS why they declined, they'd tell you that the science had yet to support the viability of in-home diagnostics. Of course, the real answer is that the vested interests of lobbying groups for sleep physicians and sleep labs prevailed over the less well-heeled advocates of common sense. As a result, the Medicare program -- and healthcare costs in general -- will continue to boom in the sleep sector as people with sleep-disordered breathing undergo unnecessary PSG tests to diagnose their ailment.
Many sleep physicians will tell you that several simple indicators can tell whether someone ought to be on CPAP. Is the person overweight? Does he snore? Using inexpensive screening technologies, like an oximeter, in the home could greatly reduce the costs associated with identifying someone with sleep disordered breathing. But advocates of reducing costs and getting more people on CPAP lost that battle last year.
The American Academy of Sleep Medicine argued that there's "insufficient evidence to support the use of portable home monitoring." Insufficient evidence. It's too bad the Academy didn't read Blink by Malcolm Gladwell.
In Blink, Gladwell describes how a small but telltale body of evidence saves more people from dying of heart attacks than more evidence. At the Cook County Hospital in Illinois, doctors were filling too many beds with too many people who came to the ER complaining of chest pains. They ran costly batteries of tests to judge whether the patients were indeed in jeopardy. But then a new chairman of the hospital's department of medicine implemented a screening process in which doctors combined an ECG with three urgent risk factors that had to do with angina, fluid in the lungs and systolic blood pressure. Very easy. Very simple. Instead of putting patients through a battery of tests, the hospital put this simple algorithm to work and watched as less information--not more--improved patient outcomes by a startling 70%.
"What screws up doctors when they are trying to predict heart attacks is that they take too much information into account," Gladwell writes.
The Cook County algorithm worked, but there was widespread resistance to it, just as there's widespread resistance to letting suppliers screen people for CPAP in the home.
One reason for the resistance, suggests Blink, is a doctor's ego. The doc doesn't want to believe an algorithm might trump his considered opinion.
"There is a tendency to say, 'Well certainly I can do better,'" a colleague of the doctor who put the algorithm to work told Gladwell. "'It can't be this simple and efficient; otherwise, why are they paying me so much money?'"
If a simple algorithm works to screen heart attacks, isn't it probable that an algorithm could screen for people who need CPAP? Probably, but don't tell that to the people funding the constructions of more and more sleep labs.