Gombuters
Thursday, February 5 2009
Hendrik Hertzberg wonders whether there aren’t further advantages to computerized health records:
If everybody’s medical records were put into what Dr. Strangelove called “a chigentic gomplex of gombuters,” wouldn’t it be possible to use data-mining techniques to tease out patterns that would add immeasurably to the health of the public? [...] What if the computerization allowed millions of individuals, sick and healthy alike, to supplement their medical records with a continuing record of their eating, drinking, drug-taking, sleeping, exercise, and other relevant habits, along with any physical and psychological symptoms and/or improvements they experience? Might not something like this yield all sorts of correspondences that the necessarily small-scale “studies” on which we now rely miss?
The picture is of Peter Sellers as “Dr. Strangelove”, of course – and not of Hertzberg.
But to answer his question: I should think so, with a couple of caveats. First, the scope of these correspondences would probably be much smaller. I would not expect patients to be too diligent in inputting their dietary or exercise habits into these records (unless paid to do so); even if they do, there’s all sorts of potential for bias. But there is a high likelihood that the medical information in such records could add significantly to our knowledge of medications, side effects, et cetera.
Second, this could only happen in a system where the gombuters are government-owned, which implies significant changes to our health care system. We can’t do this in our current system, because each provider considers this data proprietary – to the extent they collect it – and refuses to share it. Right now, if I have a serious side effect from a drug (let’s say prednisone), there’s nowhere that information is supposed to go; my doctor doesn’t pass it on, my insurer doesn’t care, and the government isn’t terribly interested, either. Moreover, there’s a lot of information I don’t want my insurer (and thus employer) to know, that I would expect the government to protect.
I have often thought that one of the advantages of a single-payer system (or at least a system with a viable public option) is that we could put this kind of information into gombuters and get a much better understanding of the long-term consequences of drugs, the extent of their side effects, the rate of patient adherence to prescribed regimens, and the efficacy of new drugs compared to similar drugs for similar conditions. I also suspect it would be easier to recruit, monitor, and track patients for clinical trials – which is presently a major obstacle to new drug approval. Hertzberg’s wonderment goes a bit further than that, but he is still within the realm of the possible.
I guess the other danger is the perceived commercial value of such data!
Hi Andy – by “proprietary”, I meant to imply that the information might have some commercial value. But, yes – you’re right: concerns that Health IT will exploit patients’ records for commercial gain are a problem for this process.