Down To Business: Debate Over: Let's Get On With Health E-Records

IT rivals national security as the most important and intractably complex IT challenge of our time--namely, getting medical practitioners, insurers, employers, and individual patients on something that resembles a digital health care grid. As costs soar and care goes wanting, the physiological and economic health of a nation is on the line. Electronic medical records are by no means the drop-in solution to our nation's health care woes, but they're a step toward taking some of the inefficiency, complexity, and mystery out of health care decision-making. Yet for every step forward, we seem to take one backward.

We reported last December that Wal-Mart, Intel, and a handful of other blue chip companies were putting their considerable clout behind a coalition, called Dossia, to build a federated data warehouse that would help more than 2.5 million employees, retirees, and dependents take more control of their health and health care purchases. The goals of the Dossia information grid, to be built and run by a nonprofit group called Omnimedix Institute, were to reduce medical errors, show employees how to lead healthier lives and make better health care purchasing decisions, and cut employer health care costs in the process. Finally, companies spending billions of dollars on health care coverage for their employees would be exerting some concerted leverage over how services are delivered.

Or so was the vision. As senior writer Marianne Kolbasuk McGee reports (see story, "Another E-Health Project In Disarray"), the initiative is starting to unravel as Omnimedix and the Dossia principals squabble over money and deliverables. A spokeswoman says Dossia remains committed to offering e-health records to at least some employees by the end of this year. We'll see. The health care industry is littered with e-records initiatives that went nowhere because of financial, competitive, liability, privacy, technical, and other concerns.

Now, if there already weren't enough fits and starts in digital health care, along comes a journal study that questions whether electronic medical records do patients any good in the first place. In the national study of doctors visits in 2003 and 2004, results of which were published last week in the Archives Of Internal Medicine, researchers affiliated with Stanford and Harvard universities found no significant difference between care associated with using e-records and not using them for 14 of 17 quality indicators.

Among the 14 categories where no difference was found: management of common diseases, diet and exercise counseling for high-risk adults, screening tests, and avoiding inappropriate prescriptions for elderly patients. Electronic medical records were found to yield "significantly better performance" in two areas: avoiding tranquilizer prescriptions for patients with depression and avoiding routine urinalysis during general medical exams. For one quality indicator, statin prescriptions for patients with high cholesterol, e-records were found to yield "significantly worse quality."

In an interview with Science Daily, Dr. Randall Stafford, the senior author of the article, says doctors in 2003 and 2004 may have been using rudimentary systems that didn't recommend medications or tests best suited to a patient's condition. Furthermore, he said most doctors, especially older ones, are "still on a learning curve" in relating to electronic media.

It's almost embarrassing to still have to argue that medical e-records are critical. They eliminate prescription errors caused by illegible handwriting; they make it easier to follow a patient's care over time; they alert doctors and pharmacists to dangerous drug combinations; they keep doctors apprised of new tests, treatments, and drugs; they put more information at the fingertips of patients; and they reduce costs, just as the conversion of any supply chain process from paper to digital does. Get on with it.

Rob Preston,
VP/Editor in Chief
[email protected]

To find out more about Rob Preston, please visit his page