Thursday, December 08, 2005

More problems with CPOE-this time apparently fatal ones

Dr. Andy comments on a tragic situation at a pediatric ICU resulting from the institution of a computerized physician order entry system(CPOE) that apparently significantly impeded proper care with resultant increase in ICU mortality. This experience is described in an article in Pediatrics.(abstract only for non-subscribers)
Previously I had commented on increased adverse reaction rate reported from a VA hospital after they put CPOE into place, but this is worse. According to the proposed explanations of what went wrong, basically the CPOE disrupted the normal flow of care by skilled nurses and doctors and seriously interfered with the procedures,practices,systems and mechanisms that were designed or evolved in that pediatric unit and while they may have not been perfect they were obviously much better than what replaced them.Treatments and diagnostic tests were apparently delayed and the important team work between bedside nurses and physicians was seriously damaged.The ICU is not a forgiving place for delays in treatment and poor communication.I hope this important and frightening article is widely read and appreciated. Systems that alter critical operations at a hospital should not be implemented without considerable input and advice from the people on the ground (the docs and nurses) and need a shakedown cruise or two before it goes live and should run parallel with the old way until folks can determine if it works.If a given IT "solution" will be useful or harmful is an empirical question and to assume that safety or efficiency or whatever positive thing is claimed will automatically occur just because it is IT is foolish. A recent JAMA article on error rate with a CPOE attracted considerable interest.The following quote from an article commenting on that study gives a good description of what it is that computer systems have to deal with in the medical management of patients."Designers of healthcare information technology must be exquisitely sensitive to the non-linear,context dependent, fast communication- dependent,interruption-filled,uncertain and collaborative nature of hospital clinical practice". Unforeseen circumstances will occur with new systems and ways of doing things.We should not be surprised at that.However, in this era of evidence based medicine it is ironic that physicians permit systems to be implanted without evidence of efficacy or safety of a particular system in a particular setting.

1 comment:

Dr. Luke Van Tessel said...

Like the illusion of efficiency lended by the average office cubicle, where dozens of people can now overhear each other's phone conversations and simmer in resentment for tens of years while the floorplan seems a paragon of congruency and industriousness, lazy minds love CPOE. It makes pharmacy inventory easier; it makes everything legible; it makes the dosing less erroneous. I mean, it kills a few people, but at least no one has to read doctor handwriting or get orders barked at them. Except for, um, the NPs who at the places I've been end up transcribing what the docs say into CPOE, turning highly paid practitioners into typists. The thing about lazy minds is that some of them are smart enough at cruise control to get into medical school, and then they become lazy-minded doctors who enjoy administrating hospitals far more than practicing medicine.