I recently interviewed a man in his late 60s with three advanced chronic diseases—diabetes, congestive heart failure, and COPD. I'll call him "Jim" because he died unexpectedly before I could get his explicit consent to share his story on this blog. After being extremely sick in the ICU last fall, Jim enjoyed several months of  reasonably good health, good enough to tell jokes and golf in Florida and tell me about how well his wife of 38 years took care of him. The article I interviewed him for is, in part, about chronic disease management, but one of the things that struck me as we spoke is how Jim personified the reasons we need more health information technology.

Jim lived in a suburb of a big city, and got his care from physicians who all know each other and share affiliations at a local community hospital with a stellar reputation. He had Medicare and good supplemental insurance, as well as a primary care physician who took the time to coordinate his care and track what all the specialists were prescribing and recommending. But all that data lived in the head and the paper files of the primary care physician. Some of it (not all) lived in an electronic medical record at his usual hospital, which I'll just call St. X. But that wasn't much use when Jim collapsed and was rushed by ambulance to the nearest emergency room—just about 25 or 30 miles away from St. X hospital and the physicians who usually treat him.

Unfortunately, the physicians treating Jim were unable to access his records at St. X. Both hospitals were equipped with computer systems, of course, but they are owned by different health care systems and therefore do not have systems that communicate with one another.

Those 25 miles were an unbridgeable gulf. Jim's wife knew a lot about his medications and conditions, but that's not enough for an ICU team to go on, not with a critically ill patient with three serious diseases. Unable to access the tests and medical records, the new hospital did all of those tests and images all over again. I am not in a position to judge whether the delays in getting all that information hurt his long-term health prospects. And I can't even hazard a guess how much it cost to repeat all those tests—except that it had to be a lot, and all or most of it on Medicare's tab. So it's a quality of care problem. A care coordination problem. And a money problem. The $19 billion in stimulus money for health IT, if spent wisely, could give doctors treating a patient like Jim a far better picture of their health history, and save us money in the long-run.