No sooner did I post a column praising the wonders of computerizing health-care records than the Journal of the American Medical Association weighed in with a study saying that the computerization of health-care records is creating problems of its own:
Hospital computerized physician order entry (CPOE)systems are widely regarded as the technical solution to medicationordering errors, the largest identified source of preventablehospital medical error. Published studies report that CPOE reducesmedication errors up to 81%. … [But] [w]e found that a widely used CPOE system facilitated22 types of medication error risks.
This is hardly surprising. During the last two decades, the rest of us learned that new computer systems are always a disaster when they first go online, creating unforeseen problems of various kinds. Nearly all of these problems, however, can be fixed over time, and what can’t be fixed can be overridden manually. As best I can make out, none of the problems mentioned in this survey were inherent to computerization itself. For example, at the hospital where this study was conducted, the terminals the doctors use don’t display information about dangerous drug interactions that appear on the terminals that the pharmacists use. You don’t have to be a computer engineer to grasp that the solution to this problem is to make the information available on the doctors’ terminals. Duh!
Whether hospitals find that providing such crucial followup is beneficial to their bottom line is another story. But that just reinforces Phillip Longman and Arnold Relman’s point that medicine should not be run according to a market model.