There has been a great deal of discussion about the need to exchange patient health information securely across disparate data silos while protecting patient privacy. I agree that this is an essential ingredient for transforming healthcare and have offered a solution in prior posts. But that's only one part of the big picture. We must also focus on the need for (and failure to) funnel comprehensive clinical outcomes measures to researchers who translate them into evidence-based guidelines reflecting cost-effective care. Health IT tools, therefore, should be competent in obtaining these data, delivering them to researchers, and then presenting the resulting guidelines to patients and providers as a means to improve and reward high-value care.
Case in point: The vast majority of "performance metrics" I've seen are based on "process measures" that evaluate care quality based on what was done instead of the outcomes/results of care. Unless this situation changes, care value (cost-effectiveness) is not likely to improve much. I've written about this situation years ago at http://curinghealthcare.blogspot.com/2007/05/knowledge-standards-and-healthcare_12.html. And interestingly enough, a recent article at http://www.modernhealthcare.com/apps/pbcs.dll/article?AID=/20100219/NEWS/302199990/1029# discusses the tension surrounding the use of "core measures" by the Office of the National Coordinator for Health Information Technology (ONC) and how they want to put more emphasis on outcome measures over process measures.
Bottom line: Sharing patient data is important, but using those data to improve care value systematically is a process that is largely overlooked. Today. This is a serious problem that must be addressed seriously.