Radiologists Must Take Care of Their Vision, Study Shows

New research is raising awareness about the visual accuracy of radiologists as well as questions about the need to create vision requirements within the specialty.

Nabile Safdar, M.D. University of Maryland School of Medicine

Elizabeth Krupinski, Ph.D. University of  Arizona

Research from the University of Maryland School of Medicine (UMSM) in Baltimore shows that 16.7 percent of the radiologists surveyed don't recall ever having their vision tested and 35 percent said it had been 24 months or longer since their last vision exam. The study was published in the June 2009 issue of the American Journal of Roentgenology.

Considering that good vision is critical to a radiologist's ability to make accurate interpretations, the results proved somewhat surprising to lead author Nabile Safdar, M.D., an assistant professor in the Department of Diagnostic Radiology and nuclear medicine at UMSM.

"Most radiologists have 20/20 vision or better, which we expected, but we didn't expect to see such variability in terms of how radiologists are taking care of their own vision," he said. "While so much attention and research has been dedicated to imaging technology, we hadn't looked at whether or not the radiologist can actually see the fine level of detail that is being shown, which is important because physicians and patients depend on the ability of radiologists to accurately discriminate findings."

Visual Acuity Tested

In the study conducted in 2005 and 2006, 48 radiologists from four institutions completed a brief survey before undergoing visual acuity testing at three separate times during the day. Subjects wore corrective lenses if routinely used. Testing was performed with modified versions of a U.S. Federal Aviation Administration visual acuity test instrument.

"We wanted there to be reliability to the vision test and also to see if their vision changed due to eye strain, stress, et cetera," Dr. Safdar said.

Results showed that eight of the 48 participating radiologists could not recall ever having their vision examined and 14 of the participants reported that their last eye examinations were 24 months or more in the past.

Results also showed a statistically significant difference between a radiologist's visual acuity in the morning versus other parts of the day, though the difference was relatively modest and within previously published ranges of variability for similar visual acuity tests.

The study also found that not every radiologist had 20/20 vision, a few needed visual correction, and more than a few had not undergone a thorough eye examination in as long as 15 years.

Vision Guidelines for Radiology Needed?

While the study's goal was not to determine whether results affected patient outcomes, Dr. Safdar said it is reasonable to think they could.

New research showing that 16.7 percent of the 48 radiologists surveyed don't recall ever having their vision tested used a modified version of a U.S. Federal Aviation Administration visual actuity test instrument (left). A scatterplot (above) shows the correlation between age and logarithm of minimum angle of resolution (logMAR) visual acuity. Reprinted with permission from the American Roentgen Ray Society

 

 

 

 

 

 

 

 

 

 

 

 

 

 

"There are instances when the smallest detail might make a difference depending on what the radiologist is reading," he said. "With technology like PACS, radiologists have the ability to enlarge or enhance the image, so even those with less than perfect vision can do a fine job, but if they don't know their vision is not up to par, they might not think about it."

For that reason Dr. Safdar said he believes it makes sense for the radiology community to raise awareness of the importance of their own visual health.

"Although the variation we found in visual acuity among radiologists is unlikely to have an influence on clinical diagnostic performance, I think it would be reasonable to discuss whether radiologists should undergo a baseline level of visual maintenance since that is so critical to their work," he said. "That said, it would be important to find out if this actually does make a difference in outcomes before any kind of firm regulations or guidelines were entertained."

At least one radiologist who is not in favor of creating guidelines said he believes the study should be taken with a grain of salt.

"There is absolutely no study that says a radiologist's vision has to be 20/20 in order to render accurate interpretations," according to Leonard Berlin, M.D., vice-chair of the Department of Radiology at NorthShore University HealthSystem, Skokie Hospital in Skokie, Ill., and chair of the RSNA Professionalism Committee.

"I admit it would be difficult for a blind radiologist to do an adequate job, but various limited impairments of a radiologist's eyes do not necessarily make him or her inadequate as a radiologist," said Dr. Berlin.

New Study Targets Reader Fatigue

In another study currently under way, Elizabeth Krupinski, Ph.D., a professor of radiology at the University of Arizona in Tucson and president-elect of the Society for Imaging Informatics in Medicine, and Kevin Berbaum, Ph.D., a professor of radiology at the University of Iowa in Iowa City, are examining the impact of visual fatigue on diagnostic accuracy.

Researchers are acquiring an objective measure of fatigue by using a refractometer to record visual accommodation (the ability to maintain focus) before and after a day of reading images. Drs. Krupinski and Berbaum are also surveying radiologists on fatigue in the workplace. To measure diagnostic accuracy, subjects are reading a set of 60 bone cases—with and without fractures—once early in the day and once after spending time reading images.

"The results have not yet been published, but I can say that we have demonstrated there is more fatigue toward the end of the day and diagnostic accuracy is impacted negatively, so radiologists should be more careful and maybe take a bit longer to look at things during that time period," she said.

Dr. Krupinski said she believes the results of her study of 40 radiologists and Dr. Safdar's study of 48 subjects will be mirrored in a larger group of radiologists. She said she does not, however, feel the outcome of these studies should lead to regulation.

"I don't really think so—I can't think of a medical discipline that requires that," Dr. Krupinski said. "I just think radiologists should be aware of their vision and we are hoping this study helps with that."