A typical workday for today’s radiologist is probably anywhere from 10 to 12 hours a day, and in most settings, nearly all of those hours are spent reading images at a PACS workstation. This may be a good thing from a healthcare enterprise and business perspective, but what happens to a radiologist’s diagnostic performance after even just eight hours of clinical work? Is it the same as when they started first thing in the morning? Are residents impacted more or less by fatigue given their fluctuating work hours and the pressures of learning their craft?
Recent evidence indicates that the answers are worth paying attention to when considering shift times, shift lengths and the need to investigate ways to reduce and perhaps avoid fatigue in the radiology reading room.
Over the past few years, my colleagues and I have been investigating the role of fatigue and its impact on radiologists and radiology residents. The first step was to determine if we could find a physiological marker of fatigue, and the natural place to look was the visual system. We found that two measures, accommodation and dark vergence—both measures of one’s ability to focus, especially at near distances—were significantly compromised after only about 8 hours of clinical work. Subjective measures of fatigue as measured by the Swedish Occupational Fatigue Inventory (SOFI) supported the fact that radiologists lacked energy, felt physically exerted, were sleepy, had physical discomfort, lacked motivation and experienced oculomotor strain after a long work day. These subjective and visual focusing variables were higher (i.e., worse) in residents compared to faculty.
Feeling tired and having trouble focusing is one thing, but does it really matter? Is diagnostic accuracy affected? Unfortunately, yes, it does matter, and diagnostic accuracy is affected.
We have run a series of studies bringing faculty and residents into the lab before and after a long day of clinical reading (about 8 hours on average) and measured their diagnostic performance (using Receiver operating Characteristic (ROC) analysis) interpreting bone images with subtle fractures, plain film chest with nodules and CT chest with nodules. In every case, there was a statistically significant drop in diagnostic accuracy of about 4 percent! We recently also replicated a series of Satisfaction of Search (SOS) studies looking at whether the combination of fatigue and the well-known SOS source of errors results in additional changes in performance. Early results suggest that, again, the answer is yes—being fatigued seems to reduce radiologists’ willingness to report multiple lesions on an exam and may have a significant impact on time to report. And residents, once again, appear to be more negatively impacted than faculty.
We are even finding evidence that visual search patterns are impacted by fatigue, potentially reducing search efficiency and extending the time it takes to interpret individual cases. All of this evidence points towards one inescapable conclusion: radiologists, especially residents, are often experiencing significant fatigue, reducing their ability to focus, and hence their diagnostic accuracy.
So what can we do? There are many possibilities, but two main themes come to mind. The first is to take advantage of the many technological solutions that are becoming available, although some admittedly may temporarily seem more work than they’re worth. Computer-assisted detection (CAD) and other image analysis tools are designed to assist radiologists in a variety of ways by optimizing images and detecting/discriminating image features that the radiologist may be less sensitive too or even unable to perceive in the same way that a computer can. Whenever feasible and appropriate, these tools should be used—as long as they are properly integrated into the clinical reading workflow and are not an impediment to efficiency and accuracy.
Properly optimizing the display and viewing conditions is another critical component that can readily be addressed to help reduce fatigue. One recent study we conducted showed that larger displays may actually make image search more efficient compared to two side-by-side displays.
The second theme centers on the radiologists rather than the technology, giving the individuals direct control of their own situation. Some potential solutions are ergonomic in nature—using chairs that are comfortable and provide adjustable heights, for example, and placing displays at proper heights to avoid neck and shoulder stress injuries. Radiologists should consider height-adjustable workstations, standing on a periodic basis and taking breaks to simply stand up and walk around for five minutes. This gives one a chance to unwind and use muscles that are not used when simply sitting all day. Consider the 20-20-20 rule: every 20 minutes, look 20 feet away for 20 seconds.
Overall, simply being aware of the fact that fatigue is an important issue can help develop good practices to avoid or at least ameliorate its impact. This again may be especially important for residents. Grand rounds lectures on fatigue and ways to avoid it are useful for both faculty and residents.
On an individual basis, radiologists need to recognize signs of fatigue beyond the obvious drooping eyelids. Prolonged staring at an image without really “seeing” anything, straining to avoid double-vision when trying to focus, reduced blink rates, general physical lethargy and discomfort, and feelings of reduced concentration and “mind wandering” are all indicators that it’s time to take a break. And taking a break should not be considered an unnecessary interruption in the workday, but recognized as a necessary and beneficial act.
Elizabeth A. Krupinski, PhD, works in the department of radiology & imaging sciences at Emory University in Atlanta.