Several years ago, I spoke at Baylor College of Medicine in Houston, where Michael DeBakey, the legendary heart surgeon, was master of the universe for nearly half a century.

I heard lots of DeBakey stories during my visit, but one in particular really stuck with me. “A few years back,” someone told me in a voice of hushed reverence usually reserved for descriptions of flawless beach days and single malt scotch, “he performed 16 open heart operations in a single day.” This was clearly intended to impress, but all I could think was, “Boy, I wouldn’t want to be patient #16.”

Lacking any information to help us understand when fatigue trumps even legendary prowess, such monumental tales of endurance can take on Man of Steel proportions. But a recent study in the Journal of the American College of Radiology may be the start of efforts to trim Superman’s cape.

Researchers from the Universities of Arizona and Iowa observed radiologists as they reviewed a handpicked set of 60 bone x-rays, half with fractures and half without. As their eight-hour workdays wore on, the radiologists’ accuracy fell by an average of 4%, with equal drops in sensitivity (missing a true fracture) and specificity (incorrectly calling a fracture when there was none). The degradation in performance was statistically significant.

While this might not seem like a huge hit in accuracy, it does mean that if you trip on a curbstone and end up in the ER with a broken ankle, you’d be better off timing your accident for the morning: nearly one in 20 more radiographs will be called incorrectly at 4 pm than at 9 am. And the study probably underestimated the impact of fatigue and eyestrain for several reasons. Much of radiologists’ time these days is spent in reading CTs and MRIs, each with scores of images to review and lots of scrolling, a much more complex and fatiguing task than reading plain films. The investigators observed only an 8-hour workday – some radiologists actually work far longer days than that, notwithstanding the field’s reputation as a “lifestyle specialty.” Finally, all 40 participating radiologists knew they were being observed, providing ample opportunity for the Hawthorne effect to play out.

More importantly, if this is how performance degrades among radiologists, just think about cardiac surgeons. Eight hours of work? The surgeon is just warming up. Sitting in an Aeron chair sipping a low-fat latte? Nope, erect but stooping uncomfortably, up to his or her elbows in blood and guts. In a quiet, darkened room, just me, my PACS system, and my iPod? Not exactly – the surgeon is meticulously performing an extraordinary array of procedural tasks while choreographic a large supporting cast, in an environment that ranges from tense to Normandy-like. Missing a subtle fracture: minimal clinical consequences, sometimes even none. The surgeon has a beating heart in his or her hands.

Taken together, it seems hard to believe that surgeons (or any physicians, really) can defy the cognitive laws of gravity – the ones that say that humans get tired, and when they do they don’t perform very well.

Until recently, nurses were the only group of caregivers studied regarding the link between fatigue and performance. Penn researchers found clear evidence of patient harm when nursing shifts lasted more than 12 hours. Over the past decade, we’ve seen great interest in one specific group of physicians: residents. Why have we focused on residents and nurses while ignoring the impact of long hours and fatigue on practicing physicians’ performance? I think there are a few reasons. Both nurses and residents are salaried, which means that duty hours are discussed as part of the cornucopia of issues that typically arise in employee/employer negotiations. In contrast, most practicing docs are self-employed, and, absent any standards or regulations, are expected to choose their own hours to optimize work-life balance, income, and (one hopes) performance.

But there are deeper forces at work that help explain why issues surrounding fatigue have been ignored when it comes to practicing physicians. Conditioned by long hours during residency, an “eat what you kill” payment system, and being put on a pedestal by everyone from our patients to our parents, most of us simply assume that we can, and should, work for as long as we can keep our eyes open and bodies erect. We’re simply not very good at recognizing our own fatigue or admitting that it might be compromising our ability to function.

That’s why this radiology study – which hints that this invincibility myth is just that, a myth – is important. I hope that we’ll see more studies addressing this topic, covering questions like: how long can a physician work before he or she begins putting patients at risk? How does this differ by specialty and setting? Are there strategies (such as so-called strategic napping) that can help physicians work long hours without compromising care? Can we make our own judgments about our fitness for practice or do we need someone else to do this to/for us?

If research demonstrates that we are working longer than we should yet we resist self-regulation (always a good bet), will we ultimately have our duty hours regulated, just like pilots and truck drivers… and now residents? I think so. Most of today’s practicing physicians wouldn’t like that.

But the next generation of physicians may be a different breed. Doctors trained in the era of duty hour restrictions – particularly those who go on to be employed by hospitals or health systems – might ultimately welcome such limits, particularly if they’re enacted with some thoughtfulness and flexibility.

More importantly, so might their patients.

Bob Wachter is chair, American Board of Internal Medicine and professor of medicine, University of California, San Francisco. He coined the term “hospitalist” and is one of the nation’s leading experts in health care quality and patient safety. He is author of Understanding Patient Safety, Second Edition, and blogs at Wachter’s World, where this post originally appeared.