One of the best things about medicine is that there is so much data to give us information about the health of patients. For the most part, though, the data is underutilized and underanalyzed – it sits on a server somewhere hoping that someone will parse through it and find potentially life-saving trends and correlations. We are now entering the era of Big Data in medicine, where more and more, people will sift through these giant datasets and discover previously unknown relationships between diseases, environmental factors, demographics, treatments, and so much more.
And to kick off this new era, Practice Fusion has partnered with Kaggle to create a challenge for data scientists of any age, in any profession, at any location in the world, to show them what this data can do for them and for healthcare. This idea is exciting for three reasons:
(1) Datasets like this are usually proprietary – hospitals, large groups, and others want to keep a tight handle on this sort of information because there is so much money at stake – data means the potential for publishing interesting findings, for uncovering errors in diagnosis and treatment, or for discovering significant new tools that might give a healthcare provider an edge over the competition. They’d be crazy to hand information like this over to just anyone.
(2) Practice Fusion has a huge dataset because it is the EMR of choice for practitioners operating outside of the big systems – it requires no upfront cost and is essentially free to operate, so the little guys without economies of scale can use it. So this dataset will provide excellent insights into the decisions made by doctors operating small partnerships or private practices around the country.
(3) Practice Fusion is essentially crowdsourcing innovative ideas about how to use a lot of data well. No need to hire the brightest minds in the business – just give prizes and see what interesting ideas emerge.
I wonder whether healthcare providers that use the Practice Fusion EMR understood that their patient data could be accessed by anonymous data nerds when they signed on to this system. If you were a healthcare provider, would you be comfortable with having your patient data handed over to anonymous data scientists to compare your health outcomes with ~40,000 other (possibly competitor) practices using the same EMR?
Regardless, I’m excited to see what exciting ideas come out of competitions like this.