All Articles Leadership Using risk modeling to improve patient outcomes


Using risk modeling to improve patient outcomes

This post is sponsored by LexisNexis Risk Solutions. Health plans are looking to integrate risk modeling and other stratification methodologies.

4 min read



This post is sponsored by LexisNexis Risk Solutions.

Kathy Mosbaugh


Health plans are looking to integrate risk modeling and other stratification methodologies to enable better care coordination between members and providers improve health outcomes and reduce costs.

In this post, we hear from Kathy Mosbaugh, VP and General Manager of Clinical Analytics for LexisNexis Health Care. Mosbaugh leads strategy and business operations for the clinical business, which focuses on population health management and provider performance analytics.

Question: What is driving the need for various risk modeling methodologies, and to which populations do they apply?

Kathy Mosbaugh: Risk modeling is key to proactively managing the known as well as the unforeseen health risks of member populations. In light of where the health care industry sits today, employing risk modeling or stratification methodologies has become a top priority for all at-risk organizations. We see three main drivers for this push: 1) The need and desire to improve quality both in terms of care coordination and delivery by identifying avoidable and unknown risks and engaging the right members at the right time; 2) Containing costs by deploying care management resources effectively and ensuring the right treatment plans are developed and the appropriate level of member engagement is achieved; and 3) Uncovering opportunities to enhance revenue by driving efficiencies across the health care ecosystem.

Q: Why don’t revenue enhancement approaches work for cost containment?

KM: Before you deploy a methodology, it’s important to understand the purpose for which it was created. Risk models built on the premise of revenue enhancement likely stratify at the group level, ignoring individual member risk and co-morbidities. Methodologies that are built for the purpose of identifying clinical outcomes, and in turn containing costs through early intervention, take those pieces into account. Risk modeling is not one-size–fits-all. Applying the wrong model can lead to the inaccurate selection of members for outreach, wasted resources, frustrated care managers and, more importantly, a less–than-optimal outcome for the member.

Q: What do I need to do to ensure I address cost containment appropriately to maximize business performance?

KM: The first step is being clear on your organization’s goal. If the goal is cost containment through better member risk identification and stratification, then you can move on to other areas. These include:

  • Understanding the current gaps and strengths across your organization as they relate to obtaining and managing data, analytics skills sets (e.g., predictive modelers or informatics/analysts), and system limitations (e.g., software/hardware).
  • Developing a strategy appropriate with where your organization finds itself in terms of clinical analytic sophistication. If you currently employ a small team without predictive modeling expertise, the best approach would be to partner with a vendor who can provide robust stratification capabilities including data and predictive science. If your organization is further along in this area, then your strategy should include complementing existing teams, models and analytics with non-traditional data sources that capture risk outside of what you currently have access to and/or the integration of unique models that help stratify populations from various perspectives.

Q: How should organizations translate insights from risk modeling into action and what benefits can they expect to see?

KM: We recommend organizations start by defining clear goals and objectives for the use of these analytics and align care management actions accordingly. Leveraging advanced risk models enables organizations to specifically target members with avoidable risk due to non-compliance with evidence-based care protocols and prospective clinical risk drivers. With this level of understanding, care managers are empowered to tailor their interventions, leading to improved member engagement. Ultimately, the benefits translate into more effective and more efficient care management.