Computer scientists at Notre Dame have come up with a technology for delivering personalized medical care. What’s noteworthy is that until now, the promise of personalized medical care has been synonymous with genetic testing. Not so with these researchers who are applying Big Data to medical practice.
Big Data has been in the news, of course, because of what Edward Snowden revealed about what the National Security Agency has been doing with our telephone records. In a more pedestrian realm, Big Data is used to populate all of those annoying ads on websites targeting you quite specifically.
Big Data technologies process huge amounts of information in order to arrive at a very specific answer through statistical analysis. In the case of the Notre Dame researchers, their so-called Collaborative Assessment and Recommendation Engine (or CARE) evaluates a patient’s medical records then comes up with likely conditions that person might develop by comparing it to huge databases such as that for Medicare and the electronic medical records that are planned under Obamacare.
With the projections for a person’s health in hand, that person and their physician can take targeted action to prevent medical problems. The technology is based on the International Classification of Diseases, Ninth Revision, Clinical Modification (or ICD-9-CM). Depending on what codes are in your medical records, they can be matched to patterns among large swaths of the population.
The lead researcher describes what we can expect. “What if you could walk out of [your doctor’s] office with a personalized assessment of your health, along with a list of personalized and important lifestyle change recommendations based on your predicted health risks?”
Sounds like marketing, doesn’t it? No surprise here: it’s based on statistical technology developed for marketing. At a more fundamental level, it’s old wine in new bottles. The basis of the analysis is the codes physicians must use in order to, for example, get paid by insurance companies. As we argue in Too Much Medicine, Not Enough Health, understanding your health as a collection of diseases that you currently have and might have in the future is a travesty of any sensible notion of “health.”
What this new entry into “personalized medical care” will do is keep you in the conventional medical system’s chutes and hopefully help control costs. One of the chutes we get driven into is the “personalized and important lifestyle change[s]” that will prevent you from developing another ICD code in your medical history.
Consider another statistical approach.
The Institute for Health Metrics and Evaluation just published The State of US Health. The message of the report is that, as one headline put it, when it comes to your health, your zip code is more important than you genetic code—and presumably your ICD codes.
I happened to hear a piece of an interview about the report. The interviewer asked whether Obamacare would make things better. The answer was “Not much.” Not because it’s ineffective, but because the medical system accounts for about 10% of peoples health. The other 90% is environmental—physical and social.
Unfortunately, like the computer scientists who are flogging their Collaborative Assessment and Recommendation Engine technology, the Institute for Health Metrics and Evaluation report is a statistical analysis of the global burden of disease. The Institute also trots out “lifestyle” as a key solution to preventing disease. Nevertheless, what it demonstrates clearly is that poverty maims and kills as does environmental trauma.
Unlike the Big Data researchers from Notre Dame—who, by the way, have patented their technology—the Institute’s report does not advocate a solution. The Big Data guys want to sell their technology through the medical system and are their using the media to sell it. The Institute is simply fanning (or hope that they’re fanning) outrage at yet another way the less than wealthy are burdened with suffering.
That’s because, even though the Institute clearly believes that social and environmental justice will give us better health and would like to see more of it, it’s not their job to advocate. That’s a matter of public policy and advocacy by public policy makers. It’s their job, not the job of Institute analysts.
On the other hand, the medical system is driven by commerce so that advocacy of a commodity (or in this case, a future commodity such as the Collaborative Assessment and Recommendation Engine) has no such limitation.
But those aren’t the only choices—nor the best. As we know all too well, we wait at our peril for public policy makers to bring us social and environmental justice and with it health worthy of the name. And so you don’t think I’m daft, learning from the experience of others does help keep your health—it just might not involve a diagnosis.
Chawla, N. V., & Davis, D. A. (2013). Bringing Big Data to Personalized Healthcare: A Patient-Centered Framework. http://dx.doi.org/10.1007/s11606-013-2455-8
Institute for Health Metrics and Evaluation. (2013). The State of US Health: Innovations, Insights, and Recommendations from the Global Burden of Disease Study. Retrieved from http://www.healthmetricsandevaluation.org/gbd/publications/policy-report/state-us-health-innovations-insights-and-recommendation
University of Notre Dame. (2013). Researchers develop system that uses a big data approach to personalized healthcare. Science Daily. Retrieved from http://www.sciencedaily.com/releases/2013/07/130725152141.htm
