The increase of data from the digitization of patient records has undoubtedly sparked a big data revolution within the healthcare industry. However, there remains a significant challenge of turning that data into knowledge that improves patient quality of care and reduces costs. When this happens, and the data becomes actionable, healthcare organizations can then work on expanding their care to patient populations.
To this end, leaders at Granite Health Network (GHN), a five-hospital healthcare network based in Concord, N.H., have set ambitious goals regarding big data and population health management. In fact, according to Bob Kay, director of population health at GHN, the network’s board of chief medical officers (CMOs) has set the objective for GHN to be the premier health system in the state within five years.
“Successfully managing data and moving forward with population health management is a key to that,” says Kay, who will be part the Health IT Summit in Boston, hosted by the Institute for Health Technology (iHT2 – a sister organization to Healthcare Informatics under the corporate parent, the Vendome Group LLC), on May 19-20, 2015 to discuss the big data revolution. Serving as a preview to that panel discussion, Kay recently spoke with HCI Associate Editor Rajiv Leventhal on how GHN is progressing with population health management, and how the health network is leveraging data for better value-based care. Below are excerpts from that interview.
Which value-based programs are GHN currently a part of?
We are a venture of five different independent health systems in New Hampshire, having started that process about three years ago. We have our employees, an accountable care organization (ACO)-type arrangement with Cigna (CAC, collaborative accountable care), and four of the five health systems are also involved in the Medicare Shared Savings Program (MSSP) So we have about 85,000 lives at risk at the moment. With the MSSP, I would say there has been mixed results—the quality has been really good, but it’s been hard to meet the savings targets. We haven’t been losing money though, at least as far as I’m aware.
How are your providers taking to this shifting way of delivering care?
Honestly, I think they’ve all described it as a great learning experience, as it’s been a way to crystallize all the learning that needs to get done and make it data-driven as well. Being data-driven has really been a huge help. All of the health systems have had an electronic medical record (EMR) for a decade or more, so they’re used to looking at data and managing it. When we started with the population health initiatives, it was a natural fit for what a lot of them have already been exposed to anyway, so that was very helpful. Our board set the goal to be the premier health system in the state within five years, and in order to do that, we have to show that we’re meeting benchmarks. There is also the physician “ I don’t want to be last” aspect to it as well.
You mention being data-driven. Can you expand on that?
Taking data out of silos, well that’s hard at the moment, I don’t know if we have what you’d call “big data” right now. We use claims data, publicly reported data, and right now we are just starting to move clinical data into our data vendor. Once the clinical data is in, we’ll have more of what’s considered “big data” to work with. But we have been using the claims data to look at variations and to drive clinical initiatives where we’ve seen some of that variation come out to be true. We have had four clinical initiatives across GHN from which there have been recommended practice guidelines derived from that.
Right now we have claims data, and that’s adequate I would say. Claims has a breadth of services that you don’t get in clinical data, but not the depth of clinical information that you would need for it to be really actionable. So we use the claims data more directionally, and then we work with experts in the health system to say, “We found something in the claims data, do you see this in your own health system in your own EMRs, and can you verify this for us?”
Can you elaborate on these four clinical initiatives?
They are asthma, low back pain, depression screening, and ED throughput. Expert panels drove those initiatives. We looked at data for asthma, for instance, and we saw some high rates of drug utilization that some of our CMO group didn’t think was appropriate. Then a panel of experts across GHN started to go though the data, using NHLBI (National Institutes of Health) guidelines as recommendations to start from, and then planned from that. At GHN, the five independent health systems all operate a little differently, so that’s why we call them recommendations so the health systems can implement them to fit their own structure. For asthma, we found that all chronic persistent asthmatics should be having a spirometry once a year rather than peak flow meters.
For low back pain, we found high rates of narcotic use for some of the patients. Also one of the elements that was missing was early treatment with physical therapy, and no one was using that in the other guidelines that we found. One of our health systems piloted that, with good results, so that was rolled out across the network.
For depression screening, we saw obviously much higher rates of utilization for patients with depression, so catching it early results in much better outcomes for both the patient and financially.
And for ED throughput, we had a workgroup, but that is really the entire system from admission to discharge, and how quickly we could get patients through the system. For that, we used more publicly reported data, and we used benchmarks within that data that we created to look out our own health systems and how we compared. We were probably in the middle for a lot of those measurers, so our CMO group and ED throughput workgroup thought we could do much better with that. We have a group of CMOs from each of the health systems who are extremely engaged and drive a lot of this process. Our board members, the CEOs of all of the health systems, are very quality driven, so you see that drive from the top down.
You recently partnered with athenahealth for its cloud-based population health management program. How is that working for you?
We were looking for two things—access to the back-end data warehouse, and also to move in our clinical data. athenahealth fit those needs very nicely, and they also have a great front end to their product with a robust data warehouse, so that was a natural fit for us going into that. Things have been working out very well with them so far.
As you continue to move forward with value-based care, what are the biggest challenges for GHN?
From a business perspective, it’s not having multi-payer alignment, so we don’t have risk-based contracts with all the payers, and that makes it difficult for it to go 100 percent into these risk-based initiatives—we’re not getting reimbursed for them. However, it’s hard for the payers to figure out the landscape they’re in and how they’ll fit in to that landscape. The pace and change in healthcare is just creating some of this uncertainty with everyone, quite frankly. So the payers have to do what’s best for them as well. Everyone will get there eventually, but it’s taking longer in some parts of the country. The key is to work collaboratively, and not force anything.
What’s next for GHN, in terms of getting the data to be used to the best of its ability?
We have to get the clinical data in first and foremost. That would be a huge benefit for us, though data quality is an issue there. If we could get more data sharing amongst health systems and payers, that would be a huge benefit and open up the data sets the way that Data.gov is doing right now.
If we are really serious about making this change in healthcare, making the data easier to use and having more of it would be a great help. There is resistance from a lot of people; from the payer side, they may look at it as giving away an asset. From the provider side, it’s giving away clinical data and data privacy implications around that. If we can solve these things and anonymize the data, it would be a jumpstart to even more innovative approaches to meeting some of the healthcare challenges that are in front of us.
To see Bob Kay and others in Boston at the iHT2 Health IT Summit, register here.