In a multisite randomized trial reported in the Journal of the American Medical Informatics Association, completeness of the patient problem list, a pivotal section of the electronic health record (EHR), was improved with automated disease surveillance and suggestions for clinicians to consider adding specific problems that appeared to be missing from the list.
Adam Wright, PhD, Dean Sittig, PhD, and colleagues created electronic algorithms to identify 12 significant heart, lung and blood diseases in the EHR. The algorithms generated 288,832 suggestions for clinicians during clinical encounters at health systems in Massachusetts, Pennsylvania, Oregon and Tennessee. With clinicians accepting 22% of these suggestions, 4.6 times as many total problems were added in the trial’s intervention arm as in its control arm.
While no doubt posing benefits for all concerned, the improvement in record-keeping and communication didn’t appear to translate into improved clinical quality, at least as comprehended by the set of quality measures promulgated by the National Committee for Quality Assurance.
Others on the study from Vanderbilt University Medical Center include Shari Just, Brian Koh, Trent Rosenbloom, MD, Elise Russo, MPH, PMP, and Asli Weitkamp, PhD. Sittig (formerly at VUMC) is a professor of Biomedical Informatics at the University of Texas Health Science Center in Houston. Joining the study were researchers at Brigham and Women’s Hospital and Massachusetts General Hospital in Boston, Penn State Health Holy Spirit Medical Center in Camp Hill, Pennsylvania, and Oregon Health and Science University in Portland.
The study was supported by the National Institutes of Health (HL122225).