Many hospitals and health systems are looking at ways to improve the quality and efficiency of healthcare delivery. Rather than go it alone, seven Connecticut healthcare organizations have banded together to build a data warehouse combining claims data with electronic health record (EHR) data in order to harness the capabilities of real-time analytics to address gaps in care for effective population health management.
The Value Care Alliance formed in December 2013 and is comprised of St. Vincent’s Medical Center, Lawrence and Memorial Health System, Griffin Hospital, Middlesex Hospital and Western Connecticut Health Network, which is comprised of Danbury Hospital, New Milford Hospital and Norwalk Hospital.
In August, VCA partnered with Burlington, Mass.-based Arcadia Healthcare Solutions and is currently deploying the Arcadia Analytics platform to all member hospitals, unifying claims data from local and national health plans and hospital and ambulatory EHRs, to support real-time analytics.
The analytics work began earlier this month and the VCA anticipates the first results, which will show gaps in care at both individual hospitals and across the VCA, in about 60 days. Within nine months, these gaps in care will include both claims data and the EHR data from individual hospitals. The Arcadia Analytics platform will serve as a central utility and dashboard capabilities will be used to compare cost, quality and efficiency metrics at the aggregate and individual hospital level, allowing VCA members to identify and share best practices between members.
Patrick Charmel, chairman of the VCA and CEO of Griffin Hospital, recently spoke with HCI Assistant Editor Heather Landi about this analytics work. In the first part of his two-part interview, Charmel shared his insights about the challenges of building a data repository with claims and EHR data across seven independent hospitals. In this second part, Charmel discusses the specific quality measures being tracked, what VCA plans to do with the results and how they got physicians on board.
What will the VCA and its members be able to do with the results from this work?
The real-time analytics will be used by VCA and hospital executives for strategic quality and cost improvement planning, and will also be utilized by specific care management teams for patient monitoring, such as coordinating the care of diabetic patients. What we’re hoping to do with the results is to bring about more efficient care and better outcomes, but you can’t do that without having the information in an almost real-time, understandable fashion. You need to have a good user interface, so that people want to access the information, because if they have to hunt for it and if they have to put a lot of effort into getting the answers they are looking for, they are not going to access it. We want our care managers, who are working with physicians and hospital staff to coordinate the care of our patients, to have this information at their fingertips. And, this is particularly important regarding coordinating care for the more vulnerable patients with significant underlying chronic disease and especially those who have been historically high utilizers of service.
What are the specific measures that you are tracking?
There are 35 measures that we are tracking based on the Medicare Shared Savings Program (MSSP) as well as Aetna ACO quality measures. [Editor’s note: VCA joined a co-branded ACO with Aetna and Hartford HealthCare last month.] As an example, with regard to diabetes, we would monitor diabetes hemoglobin A1c test results and we will be able to monitor that through the system and using that medical records interface to look at lab results. So, by looking at lab results for diabetes patients we can monitor if the A1c test was done was in the right frequency and, more importantly, we’re trying to identify whose A1c level indicates that their blood sugar or their diabetes is out of control. If so, then our care managers would contact that patient’s primary care physician or an endocrinologist and say, “We’ve noticed through the latest laboratory test results that there is a problem here and we need to get that patient back in.” And, we can also monitor the preventive exams that patients should have on a regular basis, such as breast cancer and cervical cancer screenings.
How will you use the real-time analytics to share and adopt best practices?
As an example, one of the things we can evaluate, across the Alliance, is all the orthopedic surgeons doing hip replacements and we can ask, “What is the average length of stay in the hospital for patients who have undergone a certain type of hip replacement?” There is variation with hospitals, as with some the length of stay could be two to three days and with other hospitals it could be three to five days. So, we want to identify that variation and ask, what’s going on there? Is there a different technique being used? Is it less aggressive post-surgery therapy that’s happening in the hospital? Is it a difference in the patient population? Is one hospital operating on patients that are frailer or have more underlying problems? And, then beyond that, what happens when the patient leaves the hospital? Some hospitals are sending the patient home with a week or two of home care, while other hospitals are sending patients to rehab facilities or sub-acute facilities and patients could be in that sub-acute facility up to 25 days at a cost of $600 to $800 a day. So what is the difference in the outcomes for the patient that goes home and receives home care for a week after a three-day stay compared to the patient who transitions to a 25-day stay at a sub-acute facility at $800 a day? Is there any difference in the outcome? Or is it just that there is an incentive for the sub-acute facility that, once they get the patient, they get paid per day to keep them as long as possible? So there are some answers to those questions but first you have to do know what’s going on. Having the data allows you to begin to look at that.
Armed with that data, you can then have some conversations about whether those differences are justified, identify whether there is a best practice and then see how all the member hospitals can migrate to that best practice. It’s remarkable what you find. We want to standardize care within a particular hospital, but if you have five different orthopedic surgeons, they are all practicing a little bit differently, so getting them to standardize is challenging. And, it’s an even bigger challenge doing it across hospitals, especially when they are independent. Doctors want to do what is in the best interest for their patients, but sometimes they don’t understand how their practice is deviating or how it varies from other doctors’ practices. If you put doctors in the room and share information, then they are willing to talk about “Is there a way to be doing this better?” Everyone benefits from that – the patient benefits, the people who pay for care benefit and the providers benefit. This is the approach that we’re all pursuing, some more aggressively than others, and I think we are out on the leading edge of that. That’s why we wanted a leading edge information infrastructure partner with Arcadia.
So, how did you get physicians on board with this initiative?
In the absence of the appropriate incentives, it’s a struggle. Most doctors I know are overworked and there’s this unlimited demand for their time, attention and service. This kind of work is usually on top of what else they are doing; it’s not instead of. So doctors have a full schedule and then they are told, “By the way, come by and spend time looking at your care and comparing it to others and look for opportunities to standardize it in a way that brings about more efficient care and better outcomes.” They’ll say, “Yes, I’d love to do that.” But right now, with fee-for-service, the financial incentive is basically the more you do, the more you get paid. So it’s about turning the turn style. If the incentive changes so that we only want to do what the patient needs and when they need it, and when we do provide care we make sure it’s the most efficient and produces the highest quality outcomes, well then, that motivates providers to come to the table to do this work.
Up until two years ago, as a hospital operating under a fee-for-service model, we were motivated to sit back and wait for people to get sick – frankly, from a financial standpoint, the sicker the better. And then our motivation was to do as much as the patient needs, and in some cases what they didn’t need, because that’s how you get paid. Now that’s completely changing and once physicians understand those incentives, they embrace it because they realize that a value-based approach with the right incentives eliminates some of the perverse incentives that used to exist, and still do exist, in healthcare. We’re supposed to be healthcare organizations, but we’re sitting back and waiting for people to get sick. I think policymakers in the federal government and with the Medicare program had this revelation and now it’s moving across healthcare. As a healthcare professional, it is clearly the right approach, but it’s very disruptive. Our whole system was based on fee-for-service, and again, when you are waiting for people to get sick with the idea of the sicker the better, you don’t have to do the advanced analytics work that we’ve been talking about. This transition to a value-based approach all ties into the need for data.