Peggy Chou, Medical Director, Medical Management, Atrius Health
As the healthcare industry continues to move toward value-based forms of accountable care, organizations must develop population health tools that will help them coordinate care and make improvements in terms of cost, quality and patient experience. This means that a strong IT infrastructure is needed to help identify patients, analyze data and deploy the proper interventions to improve health outcomes for different populations.
"It is important to be able to meld clinical data from the EMR with data from claims to create a robust picture of a population”
Atrius Health is a non-profit healthcare organization serving 675,000 adult and pediatric patients across eastern Massachusetts. As a CMS Pioneer Accountable Care Organization, our population health tools are crucial for coordinating care across a continuum with accountability for cost, outcomes of care and patient experience. With nearly 80 percent of Atrius Health’s revenue deriving from total cost of care or global budget contracts, population health is a central tenet of our IT strategy.
Melding clinical with financial data
Atrius Health had a long history of delivering and coordinating care across the continuum through commercial alternative payment and value-based contracts prior to when we joined the Pioneer ACO model in 2012. Upon entering the Pioneer ACO model, we gained access to historical and concurrent claims data (from hospital visits, pharmacies, skilled nursing facilities, etc.) for our previously fee-for-service Medicare beneficiaries. By integrating this information with decades of their patient medical histories from EPIC—along with medical and claims data from our existing alternative payment model patients—we were able to identify where patients experienced gaps in quality, safety, experience or cost of care and develop concrete steps to improve.
It is important to be able to meld clinical data from the EMR with data from claims to create a robust picture of a population. In doing so, our analytics team identifies groups at risk for hospitalization who can benefit from earlier interventions in addition to those already diagnosed who require closer management of care. These rosters can range from people with multiple conditions and complex medical needs who would benefit from a coordinated care plan, to those with a single chronic condition (such as diabetes), to those who would benefit from preventive care. We have also developed triggers within our EMR to notify case managers when patients require a certain intervention or screening.
Once these rosters are identified, population health managers or case managers proactively reach out to patients to coordinate appointments, referrals and make sure that patients are receiving the care that they need. For example, we saw through integrated claims and clinical data an opportunity to screen and diagnose patients for chronic obstructive pulmonary disease, ultimately getting patients treated sooner.
We are continuing to build on this progress by using predictive analytics to better understand population health risks. For example, members of our analytics team recently created an Enhanced Care Pathway (ECP) algorithm which helps us identify and proactively care for patients who are clinically at high-risk. This algorithm allows us to identify these patients in real-time through the EMR to ensure quick referrals to our practice nursing staff and a medical appointment on the same day. In blending this algorithm with a prompt intervention, we are able to ensure patients receive timely, necessary care.
Addressing challenges as we look toward the future
While individual organizations such as ours have made great progress thus far in population health, the healthcare industry as a whole still faces a number of challenges.
To coordinate care across the continuum, it is important that health systems are able to analyze data and provide the necessary follow-up care when their patients receive treatment from another organization. Interoperability between health systems can be time-consuming and challenging as it requires investment and energy to create interfaces through which clinicians can pass information. As a standalone ambulatory-only medical group, Atrius Health has collaborated with a preferred network of hospital partners to create interoperable web portals. In doing so, we can seamlessly and securely access patient information at the touch of a fingertip and receive alerts when one of our patients is admitted along with notifications to track their progress. As we move forward, Atrius Health is working to improve interoperability with our outside specialty partners, who are essential for coordinating care for many of our medically complex patients.
In navigating these interoperability barriers and coordinating care, it is important that organizations find ways to make their own EMR systems easier for clinicians to use, reducing administrative burdens and allowing them to spend more time with patients. To address this, Atrius Health is instituting a number of initiatives to automate previously labor-intensive tasks and use new technologies such as voice-activated systems to better navigate EMRs. Another solution may come through developing machine learning capabilities to help clinicians document and code more efficiently and effectively.
As organizations refine their population health strategies and gain access to more and more data, it’s important to make sure this data is meaningful and does not overwhelm clinicians and case managers. Atrius Health has begun to develop methods to better assess what we are seeing and comb out parts that are not useful.
Ultimately the collaboration between clinical, analytic and IT teams is essential in helping organizations to know what their patients need and provide them with the right, timely care. As health systems continue to build their population health capabilities, this team culture of learning and improvement can ensure that patients benefit from a proactive approach to care throughout their lives.