Clinical Data Integration - The Key to Payer Data Strategy

As the U.S. continues to shift from fee-for-service payment to value-based care, there is increased focus on improving business, operational, and technical alignment between payers and providers. For payers, this creates a demand for a more member-centric view of patient information. To achieve better alignment, data – all data – must flow more freely between value-based partners, so health plans need to include clinical data integration as part of their successful path forward. It’s an important part of the journey in the evolution to value-based payment and it requires a digital strategy.

The most recent survey from the Healthcare Payment Learning and Action Network¹ showed that 34% of all healthcare dollars are now spent on some form of risk-based payment model. Increasingly, both government and employer groups are demanding that more healthcare dollars shift to payment arrangements with both upside and downside risk. Add to that stiff 2019 ONC and CMS interoperability proposals, and now has become time to have a holistic strategy for the bi-directional exchange of clinical and administrative data.

A recent HFMA study³ (see chart) showed that interoperability was the most important capability for value-base care. Leveraging an interoperable system creates value for payers and providers. The top uses for effective information sharing between payers and providers all target removing administrative costs and streamlining cumbersome processes. At the top of this list is reducing the “chart chase” overhead associated with HEDIS and STARS measurement. Direct access to clinical data also enhances the overall quality of the data for the hybrid measures. A corollary to that same process, closing gaps in care to improve provider performance on the measures affecting them, also rates high on the list of priorities.

Payers can use real-time alerts based on Admission, Discharge, and Transfer messages (ADTs) and other data sources to help close those gaps in care and to improve outcomes for their members through better care coordination. For example, knowing that a member, especially one that has not been to a provider in a while, has been admitted to the hospital, enables care managers to intervene. Access to lab results allows payers to know whether their diabetic members have their HbA1C in control.

Other high value use case opportunities include automating prior authorization, identifying coding gaps to support risk adjustment, and medication reconciliation. National initiatives such as the DaVinci Project are helping to accelerate the transparent data sharing required by these kinds of use cases and to ease the burden of processes that historically cause friction between payers and providers.

Many payers wonder where to start once deciding to implement a clinical data strategy.

• First, identify the use cases considered highest organizational priorities such as HEDIS/STARS improvement or gaps in care closure.
• Based on the high value use cases targeted, then determine what data elements are required and the best sources for that data. Connections to national labs and state immunization registries, and connections to a state or regional health information network (HIN) both provide clinical data from a large number of provider organizations. However, if the largest providers in your network are not participants in the HIN, then the best strategy may be to connect directly with your network partner.
• Determine where clinical data belongs in your data architecture.

Historically, payer data warehouses and data lakes have not been developed to manage the complexity of clinical data. Often those databases are topic-centric (i.e. a HEDIS warehouse) vs. membercentric, do not support near-real time data, and typically need to be re-indexed every time a new data element is added. Health plans need a strategy for an enterprise data platform that can combine a member-centric view of the clinical data with claims data, as needed. This data home allows payers to bring in clinical data from across the network once, and then reuse that data for multiple purposes such as feeding a case management system, as input to a HEDIS tool or as augmentation to risk adjustment.

Siloed data acquisition can create redundancy in your information gathering. Many topic-centric point solutions (e.g., HEDIS submission), only bring in the data needed to support that specific application. A data feed might contain the entire contents of a CCDA, yet some of that data may get “dropped” since it’s not needed to support the specific application or use case. A strategy that aggregates and stores all the clinical data from each data source allows a multi-purpose approach and that can provide value to other parts of the business. For example, a care plan might not be important for risk adjustment, but would be important for case management. Moreover, you might need that same information to inform another application at another time.

Finding a vendor that supports a holistic Clinical Data Integration (CDI) strategy is a strong recommendation from Gartner.⁴ A vendor like InterSystems can support the clinical data value chain by integrating, normalizing, deduplicating, and processing data so it’s always available for upstream use cases.

Claims data is relatively error free. The standard itself is highly constrained and the purpose of payment assures a high degree of accuracy. Unlike claims data, clinical data is messy. There are multiple healthcare data standards, and each has implementation variations. A standard ADT message can vary greatly in terms of what information gets inserted into the data stream and what gets added to the outgoing message based on what has been agreed to by the recipient and the sender. EHR vendors, and the ways in which provider organizations deploy those EHRs, can also impact the variations in clinical data.

In the Gartner report cited above, payer analyst Mandi Bishop states, “CIOs are responsible for delivering business outcomes through Clinical Data Integration (CDI), such as risk score or quality measure improvement. Yet, payer IT departments lack institutional knowledge and individual experience in the handling and usage of clinical data relative to their expertise with structured administrative data. IT downplays the size of this knowledge and experience gap, which makes CDI project planning particularly difficult, and often leads to disappointing results.”

The report indicates that many health plans have failed at executing a clinical data strategy, and that they need a trusted partner like InterSystems, which excels at interoperability and integration, at the core of their enterprise platform to provide a member-centric view of their data.

---

^{1 - Healthcare Payment Learning and Action Network LAN APM Measurement Effort, October 30, 2017
2 - https://www.healthcarefinancenews.com/news/value-based-care-requires-payer-and-provider-collaboration
3 - HFMA Executive Survey http://www.hfma.org/ValueBasedPaymentReadiness
4 - Healthcare Payer CIOs, Leverage Vendor Partners to Succeed at Clinical Data Integration Published 20 August 2018 - ID G00366638 - By Analysts Mandi Bishop}