Six best practices for integrating chronic disease analytics into population health
Data Science

Six best practices: Integrating analytics into population health

June 8, 2021
We develop, implement and use analytic models for chronic disease.
Natalie Benner

I recently was a guest on the Geneia podcast, Geneia Conversations: How health plans use AI to manage chronic disease. My colleague, Jasmine McCammon, lead principal data scientist, and I were asked,

Ten years from now, how do you think the management of chronic disease will have changed?

My answer was:

  • Better, More Timely Identification: With predictive models and similar tools in hand, we can better identify the right people at the right time, before complications occur, and strive to engage patients to better manage their conditions earlier – before they experience instability and crises that lead to frequent hospitalizations and cost.
  • Personalized Outreach: Our outreach strategies will evolve and become increasingly tailored to each member’s preference. As we live in an ever growing digital world, patients will have better access to healthcare providers and resources; digitally, through telemedicine, apps, online, and of course through good old telephonic and face-to-face avenues for those who prefer it. A test-and-learn approach is critical to outreach and constant monitoring of the data supports engagement optimization.
  • Meet Patients Where They Are: We have to continue to meet people where they are in dealing with their conditions every day, educate and empower them to create achievable, personal goals that are meaningful to them, and provide resources and ongoing support to sustain long term results.

Six core best practices

For my vision to become a reality, one of the many things that needs to happen is effective integration of analytics into the population health initiatives of health plans. Geneia’s work developing, implementing and using analytic models to mitigate the burden of chronic illness has helped to identify six core best practices that can be applied to any population health strategy:

  1. Involve clinicians as subject matter experts in model creation

    At Geneia, data scientists regularly collaborate with in-house clinicians when creating and developing models. Consideration is given not only to the data, but also to how the data can and will be used in clinical practice.
  2. Integrate data and insights so everything is in the care management platform, ideally prioritized by riskiest members

    Simplify the work by integrating data and insights into the care management platform. Rather than ask care managers to wrestle with the data or multiple systems trying to determine who to prioritize, integration enables care managers to spend their time interacting with patients.
  3. Be sure care managers can see and create cohorts as well as drill down to patient-level data and record

    Clinicians need to see patient-level data, in addition to cohorts, to more effectively do their jobs. The more information known and shared about each patient, the higher quality and effectiveness of the interaction, leading to a more individualized plan of care.
  4. Align interventional strategy to risk level

    Healthcare organizations have limited resources so it only makes good sense to target direct clinical human resources to the riskiest patients.
  5. Ensure outreach and engagement scripts reference analytic models in a patient-friendly way

    Since the goal is to activate patients in their health, savvy health plans use motivational interviewing techniques and patient-friendly language, ensuring nurses and population health engagement specialists are well-versed in these techniques and provided scripting. Emphasize rapport, the assessment of current status and the positive impact that self-management and care coordination can have to prevent illness progression and complications.
  6. Use models to determine gaps in care management offerings

    Deploying analytic models can surface gaps in a health plan’s program offerings. Beyond an initial population assessment, which usually includes disease prevalence, cost and utilization, predictive models give a next level or forward look at your population if left without intervention. Program offerings need to be aligned, readily available and communicated to all patients to proactively engage members. Addressing and preventing these complications will, in time, help drive lower cost and utilization and more importantly, improve quality of life.

I encourage you to download the white paper, Chronic Disease Care: Essential AI for Health Plans, to learn more about Geneia’s analytic models to predict the onset of chronic disease and associated complications.