Star Engine

Stars Decision Engine is an AI/ML-powered solution designed to help Medicare Advantage Plans improve CMS Star Ratings through predictive insights and data-driven decision-making. By integrating embedded intelligence, it enables payers to prioritize interventions, optimize performance, and drive sustainable quality improvements.

Client

Healthcare

Industry

US Healthcare

Role

UX Designer + Researcher

Collaborated with

Data Science, SMEs, Product Stakeholders

Challenge

Medicare Star Ratings significantly impact health plans' financial performance and consumer trust. However, payers struggle with:

  • Fragmented insights: Siloed data across contracts, providers, and members.

  • Unclear prioritization: No structured way to allocate efforts for maximum ROI.

  • Lack of predictive intelligence: Difficulty in estimating end-of-year Star scores.

Outcome and Impact

Achieved a 4+ CMS Star Rating

Closed 3× More Gaps through AI-powered prioritization

Enabled Smarter Decision-Making with predictive analytics

Industry Recognition:

"The solution was co-developed with FluidEdge Consulting and featured as a breakthrough AI-powered healthcare tool by CitiusTech."


+4

Improved CMS STAR Rating

3x

Focused efforts aligned with CMS success criteria

3 month

Reduced planning time through smart simulation of scenarios

Process

1. Understanding the Problem Space (SME-Led Discovery)

Goal: Since no clear process for Star Ratings evaluation existed, we relied on domain experts (SMEs) to shape the foundational understanding.

Activities:

  • Workshops with SMEs (Quality Experts, Actuaries, and Payer Executives)

    • Identified pain points in current decision-making.

    • Mapped critical data points needed for Star Ratings improvement.

    • Defined what success looks like for health plans.

  • Shadowing & Observation of Medicare Advantage Payers

    • Understood how decisions were made manually before SDE.

    • Identified inefficiencies in tracking performance gaps.

Key Insights from SMEs:

  • Payers lacked predictive capabilities—they could only react after CMS scores were published.

  • There was no systematic way to prioritize interventions across contracts, measures, and providers.

  • Quality managers spent excessive time in manual analysis without clear ROI guidance.


2. Structuring the Solution (Co-Creation with SMEs & Stakeholders)

Goal: Define an AI-driven approach that integrates SME knowledge into a user-friendly decision-making tool.

Activities:

  • SME + UX Collaboration Workshops

    • Defined key decision points in the workflow.

    • Structured how AI recommendations should be framed for user trust.

  • Persona Development & Journey Mapping

    • VP, Enterprise Quality → Strategic oversight & contract-level impact.

    • Quality Manager → Tactical execution & measure-level improvement.

  • Concept Ideation & Wireframing

    • Explored various AI-powered dashboard layouts.

    • Mapped how data flows into decision-making steps.

Key UX Goals:
Make AI recommendations actionable and explainable.
Align insights with real-world payer workflows.

3. Prototype & Validate (SME Feedback Loop)

Goal: Test usability and validate AI insights with SMEs before deployment.

Activities:

  • Interactive Prototypes in Figma

    • Designed role-based dashboards (contract-level & measure-level insights).

    • Integrated predictive Star Scores and prioritization logic.

  • SME Validation Sessions

    • Ensured AI-driven recommendations matched real-world expertise.

    • Refined data visualizations based on SME feedback on interpretability.

  • Usability Testing with Quality Managers & VPs

    • Validated decision-making speed & efficiency.

    • Adjusted UI for better scanability & quick actions.


4. Deployment & Continuous Learning

Goal: Launch with real-world payers, monitor adoption, and refine AI-driven UX continuously.

Activities:

  • Pilot Rollout & Training for Health Plans

    • Live feedback sessions to monitor how users trust AI recommendations.

  • Performance Tracking & UX Optimization

    • Analyzed how users engage with insights.

    • Iterated on data visibility, tooltips, and next-best-actions.

Conclusion

The Stars Decision Engine transformed how Medicare Advantage Plans approach CMS Star Ratings by replacing manual, reactive processes with AI-driven, predictive decision-making. Through a user-centered design approach, the solution empowered payers with actionable insights, streamlined workflows, and sustainable quality improvements, leading to faster gap closure and higher Star performance.

Design blends creativity with purpose to shape meaningful experiences.