In terms of available data to utilize for business intelligence, the sky’s the limit for Group Health Cooperative, which gives it a distinct leg up on most of the competition. But like most of its peers, the organization is still trying to figure out how to put the “intelligence” in BI.

Group Health Cooperative’s competitive edge is its hybrid nature: the Seattle-based, not-for-profit health system operates inpatient facilities and a large physician network as well as health plans. The health plans, Group Health Options and KPS Health Plans, cover 580,000 members in Washington state and Northern Idaho. Nearly two-thirds of members receive care in Group Health-operated medical facilities.

Providers and payers both generate mountains of patient data. Combining the two creates a digital avalanche that can seem overwhelming at times, says Mark Premo, the cooperative’s associate director for enterprise data warehousing and reporting. But corralling that data into a cohesive set of information, while challenging, will be the market differentiator, he predicts.

“Taking a long view, Group Health has understood the value of data for decision support, and now we’re trying to bring our business intelligence and data warehousing closer to the leading edge to do more to leverage [our] unique strategic assets – the data from both the health plan and care delivery perspective.”

Conversation Starts with Complexity

Group Health Cooperative has a lot of experience with data warehousing and reporting. Business intelligence, not so much. “When you look at what we’re trying to do, I think ‘complexity’ is the key word in the conversation. Business intelligence in health care has been lagging because of the complexity of marrying this data. Our data warehousing initiative is an eff ort to get both health plan and provider data mapped into a single data repository – it’s a typical integration, mapping and modeling exercise, but on an extremely complex level.”

Group Health Cooperative is no stranger to data warehousing and reporting. The tipping point for its business intelligence initiatives is the implementation of an electronic health records system (from Epic Systems Corp, Verona, Wis.). The data set driving its deeper foray into business intelligence is the HEDIS performance and quality measurements that are based on its health plan data. HEDIS measurements, developed by the NationalCouncil of Quality Assurance, comprise 71 measures across eight domains of care, based on claims data. NCQA estimates that nearly 90 percent of health plans compile HEDIS measures, which requires complex crunching of large-scale data sets gleaned for claims transactions.

But HEDIS measurements and claims derived analyses tell only part of the story.

A complete picture of patient care requires the clinical and financial data sets from the care provider as well as the claims history from the payer. And those clinical data sets were not reasonably accessible until the advent of mature EHRs.

“Health care has been knocked for not doing much in terms of business intelligence, but the truth of the matter is that until you have a mature EHR, you don’t have the core data that’s prerequisite for real business intelligence,” says Marc Holland, principal consultant at System Research Services. “Before mature EHRs – ones that integrate the clinical and financials and are fed by all the necessary ancillary systems – organizations that tried to do BI had to do chart audits and abstract information as best they could. Now you have the core data available that drives BI.”

Holland also notes that health care business intelligence is getting a lot of play now because the success of federal legislation to spur EHR adoption and massively overhaul the nation’s health care system all hinge on analyzing care delivery to find the optimal intersection of costs and quality.

But to do so will require aligning the incentives for all stakeholders – most notably physician pay-for-performance in terms of clinical quality – which will take years to accomplish on a national level.

Group Health Cooperative and its peers that operate “closed” systems are the real proving ground for health care reform, he contends, because they control both sides of the health care equation and can quickly apply business intelligence to address the broken clinical and financial processes that ail the industry as a whole.

“Group Health employs its physicians and owns its health plans, which means it has everyone’s incentives aligned,” he says.

”Closed systems like Group Health and Kaiser Permanente are going to be leading the way, and have real advantages in the marketplace.”

Laying the Foundation

While Group Health tackles the multiple phases of its enterprise information management strategy, HEDIS measures serve as a foundational element. The measures track health issues such as persistence of beta-blocker administration after a heart attack, breast cancer screening and antidepressant medication management. HEDIS measurements also are “outward” facing in that they are often reported to employers and consumers in health plan report cards.

“What’s nice about HEDIS measures is that they are industry standards that are well-understood and serve as a good foundation for clinical performance reporting,”

Premo says. “With all that information flying around in our systems we have a lot of opportunities to use that data to support our solutions in the  marketplace and drive improvements in care.”

The HEDIS part of equation is pretty much settled. How to combine those standard quality reports with the incoming, real-time clinical data is Group Health’s next challenge. The meta-goal of the enterprise information management strategy is to use industry and internally developed indicators in a near real-time quality improvement cycle.

It’s elegant in theory, but brutally hard in practice. An indication of just how hard the integration becomes can be gleaned from a partial list of data sources Group Health plans to pull together in an enterprisewide repository:

  • Membership and billing system
  • Enterprise master files of consumer and patient demographics
  • The EHR and practice-based practice management suites (both from Epic) including encounter, appointment, ADT and billing data for inpatient and outpatient services.
  • Internal pharmacy claims system
  • External pharmacy claims system
  • Internal laboratory and services and results system
  • Care coordination tracking tool
  • Customer relationship management system, including complaints and appeals data
  • Database of patient experience survey results
  • Cancer screening exclusions database

One of the first goals of the enterprise information strategy is to create outwardfacing capabilities to provide enhanced reporting directly to employer groups and patients via the Web to better explain the quality/cost value Group Health can provide.
“Pure” health plans make claims data accessible, but can’t easily combine that with clinical data to offer a broader set of cost/care information for employers. Nor can they provide patient services like online access to test results and prescription histories, and physician/patient messaging, Premo says.

The second phase of the project calls for the integrated data repository to directly drive patient care, where applying business intelligence to chronic conditions is one the first items on the menu. While Group Health is in an enviable position – having provider/payer core data on chronic populations – it has yet to apply the full power of business intelligence to nail down true costs of treating chronic patients and where the improvement opportunities reside.

Premo believes a significant roadblock in that arena has been the lack of applications and interpretive tools necessary to act on the intelligence sussed out from the data. For example, Group Health Cooperative has had the ability to give providers data on their chronic patient panels, but lacked the ancillary software to help them manage those patients – and that data. If a physician has 60 diabetics in her panel, she needs electronic capabilities to document their longitudinal care – who is overdue for an eye exam, who was recently admitted to the hospital, etc. In addition, Group Health has not yet built its BI capabilities to the point where it can continually analyze the core data being generated by clinical/financial/claims systems, Premo says.

Once Group Health has chronic conditions squared away, it plans to move on to analyzing use cases for more complex situations, such as hip and knee replacements, Premo adds.

Daily Benefits

Having a consolidated repository sets the stage for business intelligence, but it also yields daily benefits at the point of care. Right now, Primary care physicians within the Group Health network often don’t have the information from out-of-network specialists and other providers who treated their patients. As part of its enterprise information strategy, Group Health is rolling out an application that pulls that data from claims submitted to its health plans, so Group Health physicians will have access to information on all episodes of care.

“Complexity can usually be overcome if you have the right approach and expertise. What limited us in the past was not having all the prerequisite components for business intelligence – such as a mature EHR – and not having the clear vision of what we wanted to accomplish,” Premo says. “But with the vision and prerequisite components, you can do almost anything.”

Register or login for access to this item and much more

All Information 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

Don't have an account? Register for Free Unlimited Access