Claiming the Future of Preventative Health

In benefits consulting, the right data can make all the difference. Maddy Compogiannis, a third-generation consultant at USI, sees how predictive analytics powered by AI and machine learning is transforming the way consultants and employers navigate today’s healthcare challenges. Echoing her perspective, HR leader Shelley Zajic highlighted the hurdles employers face when managing complex healthcare needs. Together, their insights show how predictive analytics brings clarity to an ever-changing landscape.

Outgrowing Yesterday’s Data

Traditional underwriting tools rely heavily on actuarial benchmarks and only a subset of member data, offering a limited view of what’s to come. This is especially problematic since only 30% of employees have a primary care provider (PCP).1 As Maddy describes, the other 70% are “ticking time bombs”—individuals whose unpredictable health trajectories present both challenges and opportunities for consultants and employers to address through advanced analytics.

This lack of visibility often leads to costly surprises for employers, particularly those who are self-funded and bear the financial impact of every claim. Shelley reinforces Maddy’s point from an employer’s viewpoint, explaining that self-funded employers find it challenging to anticipate high-cost claims arising from hidden health risks without predictive insights. She describes self-funding as “attractive and scary all at the same time.” Reflecting on the shock of COVID-19, Shelley shared, “When COVID hit, it was a big shock—healthcare needs spiked, costs rose, and we weren’t prepared. It highlighted the need for a better way to manage unpredictable health events.”

The High Cost of Only Looking Backwards

Traditional data analysis tools primarily use two forms of predictive analytics: actuarial projections based on historical claims, often called “experience,” and generalized industry benchmarks, known as “manual.” While these methods have long been the standard, their limitations become apparent when viewed through the lens of advanced analytics with a platform like Merit Predict:

1. Patient Behavior:

Conventional claims analysis often focuses on isolated data points, missing patterns in how patients adhere to care plans, vary in provider access, or are influenced by socioeconomic factors. Without comparing these nuances across a broader dataset, consultants are left with an incomplete view of potential risks.

2. Blind Spots:

Claims data alone offers a static, retrospective snapshot of healthcare events but lacks the dynamic context of where a patient is on their longitudinal care journey. Without additional data sources that show how similar patients navigate their unique treatment journeys, anticipating escalating risks or opportunities for effective intervention becomes more challenging.

3. Health & Wellness:

Traditional analysis rarely integrates broader determinants of health, such as behavioral, lifestyle, and preventive care gaps. By incorporating these variables, advanced models offer a more holistic understanding of long-term healthcare costs and identify actionable opportunities for improvement.

For Maddy, these technical limitations create fundamental barriers to adequate healthcare risk management. Without insights that can stratify risk at the individual level, consultants and self-funded employers are left vulnerable to unexpected costs and unable to implement preventive strategies effectively. Predictive analytics could bridge this gap by providing a clearer path forward in a member’s healthcare journey, enabling innovative interventions that offer quantifiable value and delivering real results

From Guesswork to Guidance

Unlike traditional underwriting approaches, Merit Predict, combined with vast external datasets, leverages machine learning (ML) and artificial intelligence (AI) for multidimensional predictive analytics. These models can more precisely forecast deidentified individual healthcare trajectories by analyzing demographic, socioeconomic, lifestyle, and clinical data. Shelley describes it as “a tool that takes the guesswork out— something transparent and easy to understand… a game-changer,” underscoring the value of solutions designed to solve real-world challenges rather than just selling another product.

Leveraging ML and AI-driven insights provides consultants and employers with a roadmap for identifying and managing high-cost claimants while enhancing health outcomes. In practice, predictive analytics delivers a detailed view of a population’s health profile, empowering consultants to design customized health plans that proactively address current and future employee needs. Specific applications might include:

• Risk-Based Plan Customization: Creating targeted plan options and incentives focusing on high-risk individuals before high-cost claims materialize.

  
  

• Enhanced Stop-Loss Strategy: Using predictive models to determine the right attachment point needed for stop-loss policies, making this critical coverage more accessible and affordable for self-funded employers.

  
  

• Preventive Health Programs: Allocating wellness funds toward programs that engage at-risk populations, especially those who lack a primary care provider or avoid routine medical care.

These proactive measures enable consultants to go beyond traditional retrospective cost management, and advance toward a model that identifies and mitigates health risks before they escalate.

As the healthcare landscape grows more complex, the shift to ML-driven predictive modeling has evolved from a competitive advantage to a necessity. Merit Predict is transforming healthcare cost management, giving consultants and employers a comprehensive, proactive view of future health risks. “If we had a tool that could give us better visibility into high-risk claimants before they hit that threshold, it would be a gamechanger,” Maddy notes. “We could take action earlier, put the right resources in place, and manage costs more effectively.” By embracing predictive tools, consultants and self-funded employers are positioned to move beyond reactive responses, building a sustainable, proactive approach to healthcare.

REFERENCE:

1. Bazemore A, Wilkinson E, Petterson S, Green LA. Proportional Erosion of the Primary Care Physician Workforce Has Continued Since 2010. Am Fam Physician. 2019;100(4):211-212.