As healthcare organizations increasingly take on risk-based contracts, cloud-based analytics are enabling providers to break down data silos and gain visibility into care delivery provided across their enterprises to achieve better clinical outcomes at the best possible value.
By harnessing analytics in the cloud, these organizations are leveraging tools that help deliver actionable insights accessed through the Internet without having to invest in their own costly, on-premise infrastructure. This kind of software-as-a-service ensures that providers are able to store, access and analyze a plethora of clinical, claims, risk stratification and other data, resulting in best practices and strategies for quality improvement.
Gaining shared savings
Accountable Health Partners (AHP), a clinically integrated network of hospitals and physicians in Rochester, N.Y., has successfully moved to new reimbursement models using a cloud-based data repository and analytics platform from Arcadia Healthcare Solutions to support its population health management initiatives.
AHP’s cloud-based population health platform analyzes the data to calculate hundreds of quality measures, cost utilization, risk scores and clinical gaps. As a result, its physicians make better decisions and support stronger collaborations with health plans on risk-based contracts and pay-for-performance quality programs.
According to LaRon Rowe, director of information management at AHP, the accountable care organization includes the University of Rochester Medical Center and more than 40 independent community primary care practices.
While the University of Rochester Medical Center has an Epic electronic health record system, Rowe points out that the more than 40 other practices use 10 different EHRs—which presented some technical challenges. However, AHP taps into the Arcadia platform in the cloud, enabling the ACO to centrally plan and drive clinical outcomes across their heterogeneous, geographically dispersed network of hospitals and rural independent physicians.
“One of the first things that AHP did was we entered into a shared-savings agreement with one of our commercial payers here in town,” says Rowe. “We needed to not just measure cost but quality. We also have risk modeling built into the system that helps us come up with a risk score based on an algorithm, allowing our care managers to figure out who has the potential of being the sickest patients and to coordinate appropriate care.”
Under a three-year Accountable Cost and Quality Arrangement (ACQA) with Excellus BlueCross BlueShield, quality measures are reported via the Arcadia analytics dashboard in an effort to share responsibility for providing coordinated care to patients to improve quality indicators—such as cancer screening rates, hypertension and diabetes control—and reduce unnecessary healthcare costs.
“It’s a population health system that helped us as a network measure truly how we’re doing on quality, aggregating data into a single source of truth that enables success across disparate practices,” adds Rowe. “We’re also able to integrate claims data from those local payers and bring it into a single system with clinical data.”
He says AHP is also focused on optimization for its value-based contracts by improving its workflows to better capture quality measures, adding that when data is shared and presented in a transparent, workflow-relevant format, physicians trust the patient information seen at the point of care.
More than claims data
Beth Israel Deaconess Care Organization (BIDCO), a Massachusetts ACO and value-based hospital and physician network, has also implemented the Arcadia platform to enable real-time, cloud-based analytics aimed at improving population by integrating both clinical and claims data.
“What we’re trying to do is get away from managing risk populations simply through claims data—which, in our case, has lagged anywhere from 90 to 150 days,” says Bill Gillis, BIDCO’s CIO.
BIDCO members, which include eight hospitals and about 2,600 physicians, currently operate more than 40 different EHR systems, and trying to integrate data from all the various hospitals and physician groups across Massachusetts was a daunting challenge. “We’ve got a very heterogeneous network from an EHR and clinical information system perspective,” adds Gillis.
However, using Arcadia’s EHR integration process and technology—called Data Connect—the organization has been able to extract the clinical data needed to meet risk contract requirements for quality outcomes and financial performance.
“The EHRs in our network now provide real-time data for us to actually do something with the analytics,” Gillis says. “We get that data nightly in a batch process, and we pull it into our population health platform, which gets married with claims data as well as scheduling information and [admission, discharge and transfer] to give a real picture to our care teams of what’s going on in the network with our patients.”
“If you’re a physician or hospital that’s in a value-based contract, those insights are important to ensuring that health plans are measuring you in a timely manner and that all those important insights around quality are captured in their analytics so that your outcomes are not understated or misstated,” says Eric Sullivan, senior vice president of innovation and data strategies at Inovalon, a cloud-based data analytics vendor focused on healthcare.
Gillis notes that BIDCO “has no actual internal infrastructure, aside from the desktops, PCs and network components in the closet—there are no servers here running databases; everything we have is up in the cloud.” He asks: “Why would I own this infrastructure, build it and have to maintain it when I can go to the cloud?”
“As demand goes up and needs increase, being able to scale the analytics quickly is definitely something a cloud-based solution has helped us with versus an on-premise system,” adds Rowe.
Sullivan agrees that scalability and configurability are major benefits of the cloud. “You as a provider don’t have to download software and get updates—that automatically occurs,” he says.
When it comes to cybersecurity, Gillis doesn’t worry about data stored in the cloud. He recommends that providers go with a reputable cloud vendor that is a “known and trusted entity” in the industry.
Data become insights
In many ways, cloud computing is being transformed through artificial intelligence. It’s no surprise that cloud giants like Amazon, Google, IBM and Microsoft are all making AI tools a part of their service offerings.
When it comes to AI, Rowe sees tremendous potential for machine learning and natural language processing to provide real-time analysis of clinical data and claims data for quicker insights. In particular, he observes, NLP holds great promise for unlocking the value of vast troves of unstructured data hidden within EHRs.
Cloud-based analytics vendor Inovalon is collaborating with the University of Maryland’s Center for Health Information and Decision Systems as part of its ongoing development of NLP, machine learning and deep learning solutions. One goal is to advance the ability to perform ultrahigh-speed analysis of unstructured data contained within raw clinical documentation, such as that found within EHRs.
“More than two-thirds of what is really clinically relevant is in unstructured text and uncodified fields,” says Sullivan. “If you can have a machine scroll through a 300-page medical record in a fraction of a second and identify five potential places where the physician is indicating that they may have some diabetes complications, that is incredibly more efficient than having a human try to identify those areas.”
“Artificial intelligence can be used to help with disease prediction, identify high-risk patients and preventative therapies,” contends Véronique Grenon, vice president of risk analytics of The Risk Authority Stanford, as well as the director of risk analytics for Stanford Health Care and Stanford Children’s Health. “Risk management can also help with automating and optimizing hospital operations.”
The Risk Authority Stanford, created from the hospital risk management department serving the Stanford University School of Medicine, Stanford Health Care and Stanford Children’s Health, has developed a platform for hospitals that leverages machine learning and NLP algorithms to classify millions of data points into categories that can identify key areas of risk.
TRA Stanford’s platform, called Innovence Pulse, provides a suite of tools that deliver evidence-based data on demand to the hospital industry. The risk management software analyzes disparate data sets such as incident reports, loss runs, patient complaints and net patient revenue so users can manage expectations and claims in a given area, while taking actions to address issues and prevent future losses.
“It uses machine learning and natural language processing to read the unstructured free text of an event and then categorizes that event through our Stanford Risk Lexicon, which provides accurate risk classifications, descriptions and reports that make information actionable,” says Randall Smith, product manager for Innovence Pulse. For instance, in the case of an infection that goes through a certain unit of a hospital, Smith says the platform can identify that emerging risk and enable the provider organization to intervene before it potentially affects more patients.
Beazley, an underwriter of hospital professional liability insurance, is using Innovence Pulse to identify trends in past claims and generate actionable data in real time to increase patient safety. Its claims database is one of the largest in the insurance industry and includes nearly 900,000 unique loss records dating back two decades.
Register or login for access to this item and much more
All Health Data Management content is archived after seven days.
Community members receive:
- All recent and archived articles
- Conference offers and updates
- A full menu of enewsletter options
- Web seminars, white papers, ebooks
Already have an account? Log In
Don't have an account? Register for Free Unlimited Access