In our headlong rush into electronic medical records, a move fueled by $19 billion in stimulus money, we should heed some cautionary notes that are emerging as the transition from paper to digital takes place.
In our headlong rush into electronic medical records, a move fueled by $19 billion in stimulus money, we should heed some cautionary notes that are emerging as the transition from paper to digital takes place.
One of them comes from Pulse, a U.K. primary care publication run by Diagnostic Imaging's parent company, United Business Media. Last week, Pulse reported that up to 200,000 patients had been placed at risk by failure to properly update medical records in a national summary care records program. The program of the U.K.'s National Health Service aims to digitize health records and patient encounters so that the information is immediately available to members of the system for urgent or other care needs.
The goal is laudable, but the program has been controversial. A finding that 10% of 82,000 patient records under a pilot for the program contained errors in patients' medication or allergies certainly won't help. (The 200,000 figure was an extrapolation based on an enrollment of two million patients in the summary care records program.)
According to the article, the problem seems to have been that staff in the pilot didn't always have access to some of the NHS records they needed to make changes to the summary care records, which means they, in turn, were left out-of-date.
Obviously, in the U.S. we'll be looking at more narrowly targeted EMR programs than the U.K. effort, but even here, the risk of errors remains. Last year, I visited a specialist and we reviewed medication values entered in an electronic medical record by my primary care physician, who was in the same system. Some of the values were wrong. It wasn't a life-threatening mistake, but it was still a good thing we checked.
All of which goes to show that it's going to take a lot of work to make EMRs safe from human errors. Knowing that, caution needs to be the watchword.
Can MRI-Based Deep Learning Improve Risk Stratification in PI-RADS 3 Cases?
January 30th 2025In external validation testing, a deep learning model demonstrated an average AUC of 87.6 percent for detecting clinically significant prostate cancer (csPCA) on prostate MRI for patients with PI-RADS 3 assessments.