Hospital computerization doesn't save money
2009-12-02 04:10:05.431251+00 by
Dan Lyke
2 comments
Harvard Medical School Study finds computerization does not save hospitals money (PDF):
RESULTS: More computerized hospitals had higher total costs in bivariate analyses (r ϭ 0.06, P ϭ .001) but not multivariate analyses (P ϭ .69). Neither overall computerization scores nor subscores were consistently related to administrative costs, but hospitals that increased computerization faster had more rapid administrative cost increases (P ϭ .0001). Higher overall computerization scores correlated weakly with better quality scores for acute myocardial infarction (r ϭ 0.07, P ϭ .003), but not for heart failure, pneumonia, or the 3 conditions combined. In multivariate analyses, more computerized hospitals had slightly better quality. Hospitals on the Most Wired list performed no better than others on quality, costs, or administrative costs.
The conclusion is that computerization might lead to modest improvements in "process measures of quality". From this ComputerWorld article via MeFi, Wired and /..
[ related topics:
Health Software Engineering Economics
]
comments in descending chronological order (reverse):
#Comment Re: made: 2009-12-02 19:14:01.636147+00 by:
m
Whole health care systems tend to be a disaster because they are either designed by an ignorant committee, or the decisions are made by an ignorant but egotistical individual who will never use the system. Too many such systems make a truly lousy clerk out of a physician or medical staff. Many of the ones that I have seen make no attempt to be ergonomic in their use. Little or no study is made of how people will use this system, how it will be implemented, how to cut keystrokes, or even support a reasonable approximation of the required workflow.
But, that describes a lot of non-medical systems as well.
#Comment Re: made: 2009-12-02 18:28:05.226253+00 by:
Dan Lyke
I need to read through a couple of these things more closely, and the evaluation probably doesn't get very fine-grained, but I'm guessing that the projects which didn't help much were the big top-down re-engineer everything systems. Software, and IT, works best when it picks small problems and figures out how to do that one thing better, and then iterates to the next problem.