What does a technology platform for accountable care look like? What tools do organizations need to allow them to integrate clinical, claims, patient-reported and public health data into a patient-centric data model?
Healthcare Informatics recently interviewed several chief information officers (CIOs), chief medical information officers (CMIOs), and consultants involved in creating accountable care organizations (ACOs), and a consensus seems to be developing about the areas of infrastructure they need to bolster.
“The nice thing is that every consulting firm white paper I see is saying the same thing about the tools you need, so I feel like we are marching down the right path,” says David Lundal, vice president and regional CIO for St. Louis, Mo.-based SSM Integrated Health Technologies, Dean Clinic and Wisconsin Integrated Information Technology & Telemedicine Systems (WIITTS), the technology organization that supports SSM and Dean Clinic in Wisconsin. “We have formed an ACO, but we’ve had our own health plan for years, so the Medicare ACO is in many ways just an extension of what we have been doing,” he adds. “However, this movement to ac- countable care does open up many new avenues for us.”
Premier Research Institute, a division of the Charlotte, N.C.-based Premier health alliance, recently created a framework to assess an organization’s progress toward meeting ACO model requirements. The project is described in a report, “Measuring Progress Toward Accountable Care,” (http://www.commonwealthfund.org/Publications/Fund-Reports/2012/Dec/Measuring-Accountable-Care.aspx), which was released last December. After studying 59 organizations, Premier found that beyond the basics of a sophisticated electronic health record (EHR) and health information exchange (HIE) capability, ACOs must have “the ability to integrate disparate data, analyze data across a population of patients, stratify financial and clinical risk in the population, and measure the impact of targeted interventions.” In the organizations it studied, Premier actually found a low level of IT development, but its report stressed that “a robust capacity for population health analytics will become more important over time due to the heavy reliance on data sharing and analytics that will be required in mature ACOs.”
But even if they understand that sophisticated analytics and HIE capabilities are necessary, the real challenge is that IT leaders are “trying to hit a moving target,” says Bob Schwyn, principal with Pittsburgh-based consulting firm Aspen Advisors. CIOs need to understand the business case and determine how fast they have to be moving, he adds. “Many realize they need to build an analytics environment, but what are the specific requirements needed? Often, there is not a clear answer.”
Bob Schwyn
The challenge for CIOs and other healthcare IT leaders is thinking deeply enough about what they need now and what they will need in the future in terms of things like secure messaging. Are they walking then running, or jumping right in? “CIOs already have a lot on their plate between meaningful use and ICD-10. They are stretched with budget and resource challenges,” Schwyn notes. “They have to hit the pause button, and really understand ACOs, before jumping in.”
Lundal says the biggest areas of growth for WIITTS would be business intelligence and telehealth. “We have created a shared analytics platform that pulls claims data and clinical data from its Epic (Verona, Wis.) system. It was a hard thing to create,” he stresses. “We are using data analysis and predictive modeling tools from CPM Healthgrades. It is a long road. We have to work through defining new terms involving the cost of care. If you ask the health plan, they will have one definition while the provider side has another.”
As provider organizations are paid more for value than volume, Lundal adds, telehealth services that can improve access make more financial sense. “Giving patients more convenient access and the ability to speak to clinicians may stop their health from degrading. We are already doing teleradiology, telepharmacy, and teleconsults.”
HIE: PUBLIC OR PRIVATE?
If one of the foundational tools for care coordination is health information exchange, the question CIOs may face is: public or private HIE—or both? Dean Clinic and SSM Health Care in Wisconsin have created a single platform based on Epic Systems technology across six hospitals and 900 physicians. Using Epic’s Care Everywhere tool allows the organization to share data across all those providers as it waits for public HIEs in Wisconsin to develop so it can reach even more care settings.
But the University of Michigan Health System’s (UMHS’s) ACO is turning to a public HIE, the Great Lakes HIE (GLHIE), in East Lansing, Mich., to share information with those outside the health system. (UMHS is partnering with IHA Health Services Corp. and Huron Valley Physicians Association practices, both in Ann Arbor, Mich., in a pioneer ACO.)
“We are an Epic site and Epic has some pretty good tools, but the challenge for ACOs like ours is that one-third of patients see providers outside our system,” says Andrew Rosenberg, M.D., CMIO for the University of Michigan Health System in Ann Arbor. We need a platform to exchange data with providers in the ACO, IHA and Huron Valley, because we all work in the same area and we might be seeing each other’s patients and not know it. This sub-state HIE is the way to go.”
Rosenberg says UMHS needs more types of information to be shared in GLHIE. “We can share medication lists, allergy lists, and problem lists. That’s good, but we also need electronic referrals, and we have to start sending CCDs [Continuity of Care Documents], which we can do through our Epic system.”
THE NEXT STEP: MELDING BUSINESS AND CLINICAL DATA
UMHS is an academic development center that developed many homegrown tools during an ACO demonstration project phase in order to track elements about patients with chronic conditions such as diabetes and heart disease and feed that information to physicians to help prevent readmissions.
“What we are still looking for and what vexes us and a lot of people are tools that meld business data from revenue cycle systems with clinical data,” Rosenberg explains. “There have been dozens of applications created trying to answer that one question: How much cost can be attributed to this one patient’s care? It takes true integration of business and clinical systems and tightly tying it together for a given patient’s total cost, as well as splitting up reimbursement.” Current EHRs can’t do innovative analytics to better meld clinical, quality and cost data, he says. “That is where we will see opportunities for new firms,” Rosenberg says. “Some academic medical centers like us have built these analytic tools in the past, but I think if we rely on home-grown solutions we will find it difficult and costly. As vendor systems and applications become better, and as there are more requirements such as the quality measures and thresholds in ACOs, the onus on us will be to execute, not to develop systems.”
Cynthia Burghard, who tracks ACO trends for Framingham, Mass.-based IDC Health Insights, says that some ACOs are using payer-based analytics tools and figuring out how to fine-tune them to use in a provider setting. Some solutions providers such as Eden Prairie, Minn.-based Optum have tools for both providers and payers and are plucking what they need from both to create tools for ACOs. Others are turning to EHR and HIE vendors that are starting to add analytics tools. “The challenge there is that they are not used to bringing in claims data or doing predictive modeling,” she says.
Cynthia Burghard
Burghard says ACOs also need tools with workflow capabilities to manage patients across settings of care. Some are relying on patient registries; others are looking at care management tools that payers have been using for several years. Once these are in place, they will also turn to better consumer engagement to work on medication adherence and behavioral changes with them, she adds. “That is why you need the analytics first: to help identify people most likely to make changes and devote resources to them.”
POPULATION RISK MANAGEMENT TOOLS
MissionPoint Health Partners ACO in Nashville, Tenn., was formed as an alliance of hospitals, technology firms, wellness providers, and more than 1,200 physicians who work as a care team supporting the health of area enrollees. It is using population risk management tools from the Washington, D.C.-based The Advisory Group’s Crimson division. “Identifying which patients are in the 5 percent that make up 60 percent of your cost is one thing, but discerning who is moving toward being in that 5 percent is more difficult,” says MissionPoint CEO Jason Dinger. The organization is also using what Dinger calls an “ACO-centric” registry, a Web-based tool that allows health partners and physician offices to share data and track patients, and a care continuum tool to engage physicians in clinical performance assessment.
Paul Roscoe, CEO of the Crimson division of the Advisory Group, notes that some industry efforts to engage physicians in quality improvement haven’t worked. “You really have to have good data sets to engage physicians,” he says, “and have it be as real-time as possible. We have a new initiative to get caregivers data feedback while the patient is still in the hospital.”
MissionPoint works in partnership with the IT teams of five-hospital Saint Thomas Health in Tennessee, Dinger says, but intentionally has its own separate IT group because they have different perspectives. “If you try to just leverage hospital IT resources for population health management, you end up with a suboptimal situation or a platform that is overbuilt,” he explains. “You see the world from where you sit, and from the hospital perspective, population management might mean their patient population, whereas only a small percentage of people actually end up in the hospital, so I think it is a fool’s journey to think that clinical integration is sufficient.”