Electronic medical records

For a number of years now electronic medical records (EMRs) have been promoted by politicians and policy experts who know little about clinical medicine on the ground. Alongside this have appeared publications which purport to show beneficial clinical effects of EMRs. Virtually all of those, however, have reported low level and unvalidated surrogates for quality. The evidence that's out there does not convince that EMRs improve meaningful clinical outcomes. Dr. Robert Centor, the blogger at DB's Medical Rants, recently wrote a post about the unfulfilled promise of EMRs. Much of what he said concerned how they undermine diagnostic accuracy, a point with which I heartily agree and plan to elaborate on in a future post. But he concluded with this:

EHRs could help medical care. Currently it is only making physicians more busy work.

Much of the busy work he was referring to comes from what is known as computerized physician order entry (CPOE). CPOE was touted for various reasons as a process for enhancing patient safety and reducing errors. Again there was only soft and flawed surrogate evidence in support of that claim. But the main thing to understand about CPOE is that it is nothing more than shifting the workload of order processing from the traditional ward secretary to the physician. Unlike the ward clerks, though, doctors lack secretarial skills. Sure the reduction in secretaries saves hospitals money. But whether it improves patient safety by taking a link out of the chain of process or jeopardizes patients by removing a layer of safety (the secretary, usually more skilled in the processing and administrative aspects of implementing orders) is unknown.

Computerization of medicine is great in the abstract. Superficially it's a can't miss idea. That's why people in leadership who are removed from direct patient care can't understand our hesitation and our many objections. The unintended consequences are huge and difficult to navigate. Someday we will learn to use the EMR for all it is worth. Only then can we be confident that patients are better off as a result. It may take another decade.