2024 Health Meeting VIRTUAL - Session 7D: Leveraging Predictive Models to Estimate Stop Loss and Large Claim Risk 7.25.24 Recording

As health insurance self-funding becomes more popular among smaller group sizes, risk is evolving for the MGUs, stop loss writers, and reinsurers that cover them. Assessing more small groups at lower attachment points means rethinking traditional stop loss underwriting techniques. Predictive models and machine learning have been adopted across the industry and can now be used to assess group-level claims risk in the stop loss and level-funded markets. This session will explore large claims estimation used by underwriters today, emerging modeling techniques for specific and aggregate stop loss and how they can be combined with clinical expertise to optimize business outcomes. We’ll assess the need to consider future treatments, emerging costs, risk variance, and tail-risk metrics—in addition to expected claims—to make definitive underwriting decisions. In this session, attendees will be able to: Better understand modeling stop loss claims considerations vs. other claim estimation techniques Compare clinical underwriting and machine predictions Assess considerations when underwriting and binding stop loss risk