Four health systems, Armonk, N.Y.-based IBM, and the Charlotte, N.C.-based Premier healthcare alliance have come together to launch the Data Alliance Collaborative (DAC), with the goal to ensure healthcare providers get the data analytics and business intelligence they need to improve population health.
Through this first-of-its-kind initiative, DAC members are co-developing and sharing knowledge, data and resources to address unmet healthcare needs. The members include:
- Carolinas HealthCare System (Charlotte, N.C.)
- Catholic Health Partners (Cincinnati)
- Fairview Health Services (Minneapolis)
- Texas Health Resources (Arlington, Tx.)
- Premier healthcare alliance
Healthcare is rapidly moving toward becoming more integrated and accountable, but its IT fails to connect and interpret the data sets needed to effectively manage population health.
For instance, legacy electronic medical records (EMRs) cannot integrate clinical, financial and operational data across individual hospitals, health systems or the continuum of care. As a result, providers are making major investments in separate business intelligence and analytic solutions, which are often more complex to implement and administer, affecting their ability to quickly attain value.
In addition, providers are developing and acquiring the same or similar analytics to those being used by their peers, but the nature of today’s technology makes sharing them a challenge. This keeps data —and providers—locked in their individual silos.
The outcome of all this is waste: providers reinvent the wheel, make technology investments without a clear return, and encounter patients who do not benefit from innovative care.
To address these challenges, DAC members are sharing their experiences and intelligence to co-develop solutions that integrate data across the continuum. They’re building data analytics designed by them, for them, in a collaborative format that accelerates efficiencies and cost savings while avoiding duplication of effort.
“Instead of investing in and developing multiple, fragmented solutions that address the same problem, we’re pooling resources to develop single solutions we all can use,” Terry Carroll, senior vice president of transformation and chief information officer (CIO) for Fairview Health Services, and DAC chair, said in a statement. “We’re using big data, as opposed to local or siloed data, and will get richer insights as a result. Sharing assets and testing new and innovative ways to use analytics will help us achieve system-wide change that positively impacts quality, cost and the care experience.”
Among the first DAC co-development projects is a first-of-its kind model designed to quickly notify providers of groups of patients who have not filled prescriptions within 24 hours of discharge, and to immediately intervene, DAC officials said. Today, getting such accurate, timely data requires manual analysis that can be done on individual patients only; real-time tools to review large groups of patients and surface outliers have previously not been available.
DAC members are also co-developing an all-cause predictive readmissions model that analyzes both EMR and administrative data to identify patients who are most likely to be readmitted before they are discharged. Current readmissions models can’t analyze all conditions while accessing both EMR and the administrative data used by payors. The DAC model will also identify risk factors leading to readmissions, tying patients to appropriate evidence-based checklists based on their condition.
“Such data integration can help ensure members develop and deploy best practices for population health, while accounting for their unique care delivery processes and cultures,” Allen Naidoo, vice president of data analytics at Carolinas HealthCare System, said in a statement. “What will change in upcoming years as we continue to build this level of analytics capability is the ability for providers to make more informed decisions for patient care. This collaborative will result in the ability to leverage the same data model, and that’s a tremendous win from a data aggregation perspective.”