Medical records stored on computers instead of paper are touted as the modern way to cut health costs while improving quality. But the validity of the proposition is called into question by a study appearing in the current issue of Archives of Internal Medicine.
On 14 of 17 quality indicators, physician practices using electronic health records, or EHRs, performed no better than those that didn’t. And on one, the wired offices actually fared worse: patients with high cholesterol were less likely to get statins.
On the plus side, medical practices using EHRs were significantly better at appropriately avoiding anti-anxiety medicines called benzodiazepines for patients with depression. They also skipped unnecessary urine analysis for general medical examinations.
Still, the authors of the report wrote that they “found no consistent association between EHR use and the quality of ambulatory care.” They are Harvard Medical School’s Jeffrey Linder, David Bates and Blackford Middleton; and Stanford University’s Jun Ma and Randall Stafford. The researchers acknowledge that their findings conflicts with those at a few “benchmark” facilities, but notes that those hospitals developed their own electronic-records systems. The point of the newly published study was to learn whether electronic records in the real world did much good.
The study notes that correlation doesn’t mean causality. Lower-quality clinics could have been using the technology to try to improve. It’s also worth noting that the study covered the years 2003 and 2004, and companies have improved software since then.
In the end, the researchers weren’t altogether bearish on electronic records. There was “substantial room for improvement” at all the clinics, and they suggested that improved electronic records “almost certainly represent part of the solution.”
