Population Health’s Mandate for Better Data

By Karen Knecht, CPHIMS
Chief Clinical Officer

Value-based care is healthcare’s mantra today, and it goes hand-in-hand with population health management (PHM). Though there are many definitions of PHM, the term simply means managing patient populations for the best outcomes and the best economics. PHM is hard to wrap your arms around because it involves changes in process, mindset and workflow, as well as the implementation of diverse technologies.

It’s also neither a practice nor a methodology but a way of delivering healthcare to avoid high-cost, emergent situations. PHM requires a systematic approach to delivering care, revolving around a mix of technology solutions such as electronic health records (EHRs), disease management programs and decision-making tools. Since PHM has many moving parts and complexities, it’s important to have a roadmap to guide healthcare organizations.

At emids, we’ve developed the following four-part framework for population health management IT:

  • EHRs, which serve as the foundation for providers to deliver care across the continuum and the hub for data collection;
  • Infrastructure, which enables safe data sharing and engagement across providers and with patients. This includes technologies such as telehealth, patient portals, security and the systems integration platform;
  • Care Management, consisting of tools for provider/patient communications and to ensure patients follow care plans and adhere to medications and other therapies;
  • Analytics, which involves leveraging data to assess population health and risk. This segment of the plan is where the magic happens.

Data Is the Heart of PHM

Having the right technologies is foundational, but we need to take a step back and focus on the data. The future of healthcare depends upon the ability of organizations to do an excellent job of collecting data from many sources, cleaning it up so it makes sense to all parties, securing it, and leveraging it for decision-making and continual improvement. That’s quite a lot to accomplish. Here’s how I see the challenges:

1. The state of data is not sound. From my view, the 2009 HITECH Act was only partly successful because it dictated the installation of systems but said nothing about standardizing data. We have much work to do in the areas of data governance, data exchange, privacy and security, and workflow.

2. Better data quality requires standardized workflows. Taking the length-of-stay (LOS) metric as an example, even within one hospital, caregivers might calculate admissions and discharge times in several different ways. Adding to the challenge is caregiver mobility. Admissions and discharge times depend on when an individual has time to input data into the system. These challenges make it nearly impossible to find a usable metric for LOS—a problem as it’s a critical measure for analyzing cost.

3. We can’t ask caregivers to do more. Consider the process of an ER admission. Before a patient sees the doctor, she undergoes visits from a steady stream of data collectors: nurses, financial people, pharmacists and technicians. There’s simply too much administrative burden on the care team and on patients, which impedes quality of care. There’s got to be a better way.

4. Data overload. Here’s the good news: We finally have systems in place to collect all the data needed to transform healthcare. The hitch: We must know how to clean it up, integrate it and analyze it so that it has meaning.

5. Interoperability is still nascent. As organizations have acquired IT systems, they’ve also undergone mergers and acquisitions and partnerships. It’s not uncommon for a single health system to run several versions of a single vendor’s EHR, which makes data exchange problematic. The challenge is exacerbated when sharing data outside of an organization’s walls between different vendor solutions. Solving the interoperability issue is not just a technology problem. There are cultural and competitive issues around sharing data, along with regulatory issues around how organizations can share data.

Those large organizations ahead of the PHM curve, such as Kaiser, Providence St. Joseph Health and Geisinger Medical Center, share a few tenets: They have strong data governance programs and minimal silos in place, and they have engaged their stakeholders. As always, people institute change, not technology. As long as we keep our attention on data quality, we can make steady progress toward value-based care.

For more of our perspective on IT strategy for population health initiatives, check out our white paper.

Karen Knecht is Chief Clinical Officer for emids, bringing more than 30 years of expertise in healthcare operations and healthcare IT. Karen has extensive experience helping providers implement strategies and clinical computing technologies to improve the quality and costs of care and respond to federal and national eMeasure initiatives. She has provided consulting for major medical centers nationwide as well as the National Institutes of Health Clinical Center and leading health organizations in the U.K. and Singapore. She previously held leadership positions at Encore Health Resources, IBM, Healthlink, Inc., and UTSW Medical Center. Be sure to connect with Karen on LinkedIn.