Rapid-response modernization of Pennsylvania’s legacy Medicaid system
The Problem
Due to H.R 1 Medicaid Community Engagement requirements, Pennsylvania needed to implement new work requirements for Medicaid recipients by January 1, 2027, introducing complex compliance rules that threaten to cause procedural disenrollment for over 750,000 vulnerable Pennsylvanians if implemented poorly. Preparing for these changes would mean costly improvements to legacy benefit systems, and additional manual verifications falling on understaffed County Assistance Offices (CAOs).
MO’s Solution
Using a human-centered design (HCD) approach, we partnered with the Commonwealth of Pennsylvania (COPA) Department of Human Services (DHS) to conduct discovery research, prototype digital solutions, and define a scalable technical architecture for implementing the new requirements. Our rapid-response initiative aimed to support COPA with this transition without adding massive technical debt and system fragility to Pennsylvania’s already strained monolithic legacy system (eCIS). We led an agile, HCD and technical discovery process and bridged silos by facilitating a 60-person cross-functional workshop (spanning policy, operations, and IT) to prioritize implementation targets using a data-driven matrix.
Our team architected and prototyped a modular, API-first verification engine using highly reliable, scalable technology (Python, FastAPI, PostgreSQL). By decoupling the business rules from the core legacy eCIS database, we designed a log-based, event-driven architecture that ingested external data, processed compliance rules, and maintained an immutable, fully auditable history of every determination.
Outcomes
16 hrs
per month projected time saved
We fundamentally de-risked the state’s launch deadline by proving that a standalone rules engine could seamlessly integrate with legacy systems via REST APIs. The prototype we built demonstrated how we could automatically evaluate income compliance for an estimated 361,000 enrollees (48% of the affected population), which would save caseworkers roughly 16 hours per month while preventing system-wide degradation.