A study by a medical malpractice insurer has concluded that misinterpretation of scans and films, most notably CT scans, were the leading cause of patient injury involving diagnostic radiologists.
The study, which was published by The Doctors Company, analyzed closed malpractice claims against both diagnostic and interventional radiologists in an effort to understand what led to patient injuries.
Researchers found that with diagnostic radiology, misinterpretation of diagnostic studies was the key contributing factor to injury in 78% of cases, with the most common injury resulting from such mistakes was an undiagnosed malignancy.
The study also found that CT scans were a factor in 34% of the cases involving the misinterpretation of diagnostic studies.
Meanwhile, in claims against interventional radiologists, physician experts found that the top contributing factor to patient injury was technical performance, which came up with 76% of such claims. Most of these involved patients suffering poor outcomes after invasive procedures. Just 11% of claims were due to poor technique or incorrect body site.
In 65% of such cases, the patient saw an undesirable outcome despite correct procedures being performed appropriately.
In theory, if your healthcare organization has done everything it can to create appropriate processes, and physicians are performing their duties appropriately, you may have done everything you can to keep mistakes to a minimum. However, emerging artificial intelligence technologies might be able to help you to make improvements in ways you hadn’t expected.
Before discussing AI options, it’s worth pointing out up front that no one expects to see AI doctors take over the jobs of human physicians. There is still much that trained human clinicians do which AI algorithms can’t do, or at least can’t do accurately and well, and for a number of technical reasons things are likely to stay that way.
On the other hand, it’s entirely possible that AI technology will become a valuable “member” of the care team over time. Radiologists are already using AI to do computer-aided cancer discovery, which in turn frees them up to spend more time on less-routine cases.
Such tools will eventually be available for virtually any specialty you can name, with the AI playing an important but supporting role in the process of patient care. They will also help primary care providers do the best possible job of empowering specialists to whom they refer. For example, the FDA has already approved an AI-based technology for use allowing trained primary care providers to diagnose diabetic retinopathy independently.
In other words, it’s probably a good idea to keep your eye on the progress of AI tools designed to support clinicians. While it’s early in the game, eventually they’re likely to offer big benefits.