Correct Data Interpretation Is Invaluable, Says Dr Ken Cohen

Big data has valuable implications for analyzing datasets and in day-to-day clinical practice, explained Ken Cohen, MD, director of translational research for Optum Care.

Transcript

In general, how well is data currently being used to decide clinical care?

There are 2 answers to that question. One of the roles that I have now is director of translational research for Optum Health and it gives me access to very large databases. For example, we have a database that has 130 million patients. We just used that database to do a comparative analysis of patients with chronic low back pain to see if spinal cord stimulators are effective. We were able to demonstrate, with high reliability that we couldn't measure any positive effect of spinal cord stimulators. So, that's a synthetic randomized controlled trial that uses large data sets.

The reason that that's important is that a lot of trials aren't going to be done, spinal cord stimulators as an example. This gives you a way that you can duplicate with almost the same degree of rigor, a prospective randomized controlled trial just using big data.

The second way is how can you use data in day-to-day practice? There, it's really incumbent on the medical directors of the large medical groups and of the health plans, to work with physicians to have them understand how to interpret that data. People can get very fixated on the accuracy of the data, and the data will never be 100% accurate, but it can be equally inaccurate across all providers. If it is highly accurate, and where there are flaws, they're distributed equally, then that data becomes highly reliable data.

Once providers get over that hurdle and understand that that's highly reliable data, when it's presented to them in a clear and concise fashion with actionable results attached to it, then it becomes valuable.