Dr Bobby Green on How Flatiron Is Working to Improve Care Delivery

Bobby Green, MD, MSCE, senior vice president of clinical oncology at Flatiron Health, discusses analyzing patient data to improve the overall care delivery and its challenges involved with being so early in the process.

Transcript (slightly modified)

What does Flatiron Health do with patient data that its clients capture to help improve care delivery?

Ultimately, what we’re building is what’s called a learning healthcare system. So, taking data from the point of-care, as doctors are taking care of patients, using that data to generate knowledge, and then taking at that knowledge and feeding it back to the point-of-care to help inform better care. That’s a learning healthcare system. We’re doing it predominately in 2 areas—1 is using de-identified patient data to generate real-world evidence to help go back and inform that care. So looking, for example, at how something called tumor mutational burden—which is something you can find in lung cancer patients—how that interacts with the use of certain drugs that are being used for drug cancer. And, whether you are actually able to use a biomarker like that to predict whether or not patients are going to respond to drug.

Ultimately, that’s incredibly useful information as a clinician to be able to share with your patients. There are questions that we think can be answered with real-world evidence. At the same time, there are always going to be a need for prospective interventional clinical trials. We believe that the world is changing in this direction. Groups like the FDA, legislation from 21st Century Cures, the Cancer Moonshot, are all pushing us in a direction and in many ways almost mandating that we rethink how we do clinical trials, how we think about using technology, as well as a network of practices to really enable the clinical trials of the future; that’s another thing that we think is critical and that we’re building the data in the infrastructure to do.

How far have we come in making data usable at the point of care? What challenges remain?

I think we’re very, very early. We’ve luckily been able to take some of these data in publications that we’ve had—the example that I just gave you of the tumor mutational burden—where we’ve published that data and think that it’s just one more thing in the tool kit that a clinician can use to help inform care and we’re continuing to publish. So, that’s one way the information gets back through clinicians—through publications. We’re also working on ways to deliver that information immediately, at the point-of-care through a portal to clinicians. We’re very, very early in this but we think a big component of it is first having the network, and second having the data infrastructure that allows you to do this. We’ve been actively working at building both.