DataBridges in Action: Multi-System Migration to Epic

How a large academic health system migrated data from three legacy platforms on time and with confidence.

At a Glance

3

Source Systems Migrated 

Epic, Cerner, Athena

4

Clinical Validation Rounds

1 Small + 1 Large + 2 Full Scale

Tier 3

Data Tier

Epic’s Highest Classification

Dec 2025

First Go-Live

On-schedule delivery

79 / 26,000

Tickets Attributed to Conversions

< 0.3 % issue rate

5

Post-Live Delta Loads

Ensuring full continuity
The Challenge

A large academic health system on the West Coast was consolidating onto a single, unified Epic instance, a transformation that required migrating three distinct source systems simultaneously:

  • A legacy Epic instance from a shared platform with another health system
  • Cerner, inherited through the acquisition of several hospitals from a national health system
  • Athena, used across a collection of independent clinics and providers

This was a Tier 3 migration, Epic’s highest classification, meaning the full breadth of supported data types needed to be converted: discrete clinical records, patient photos, allergy and medication histories, immunizations, eye exam data, FYI flags, notes, labs, and more.

The stakes were especially high on the Epic-to-Epic side. When providers move from one Epic instance to another, they expect their data to follow them completely. Gaps aren’t acceptable, and they’re not always easy to explain.

The Solution: DataBridges

Health Data Movers brought DataBridges, its structured, four-phase delivery framework, to guide the migration from initial scoping through post-go-live stabilization. The framework provided the discipline, checkpoints, and stakeholder alignment needed to manage a project of this scale without chaos.

Phase 1: Define  |  Eliminate Assumptions Early
  • Assembled a conversions workgroup with clinical, operational, and IT leaders to co-own scope decisions
  • Defined data requirements across all three source systems, with special focus on the Epic-to-Epic conversion
  • Conducted granular scoping on note types, encounter categories, and document types — going well beyond high-level classification
  • Applied data-specific historical lookback recommendations: standard 3-year windows for most data, extended to 5 years for certain note types, and lifetime lookback for labs like genetic testing where historical validity is permanent
  • Documented the convert vs. archive vs. abstract strategy for each data type, ensuring no false assumptions about what would land in the new system
Phase 2: Prove  |  Build Confidence Before Scale
  • Began with small-scale validation internally — surfacing early issues before clinical teams were involved
  • Conducted four progressive rounds of clinical validation: one small-scale, one large-scale, and two full-scale
  • Each round provided over 220 clinical and operational SMEs a list of 50–100 patient charts to review; their feedback was triaged line by line
  • Performed a facilitated validation session with highly engaged clinicians who walked through real patient records to confirm day-one care readiness
  • Executed formal Confidence Gate sign-offs at each stage — from IT-only at small scale, to full clinical, operational, and IT approval at large-scale and full-scale testing and validation
Phase 3: Execute  |  Deliver a Controlled Cutover
  • Managed a detailed dependency map tracking every data type across bulk load, Delta 1, Delta 2, Delta 3, and beyond — with real-time status per item
  • Deployed a live lab interface from the legacy Epic system to the new environment, eliminating delays for bringing time-critical results into the new system
  • Obtained production sign-off before moving any data type forward — layering test environment approval with real-world validation
  • Presented a clear cutover tracker to the workgroup each week, surfacing blockers and progress with full transparency
  • Maintained close communication with stakeholders on what data would be available, and when — preventing confusion at go-live
Phase 4: Stabilize  |  Maintain Continuity After Go-Live
  • Completed approximately five post-live delta loads for Legacy Epic and Athena — each communicated clearly with a dated data availability schedule
  • Continued managing Cerner data finalization, navigating Oracle’s extraction timelines and extended Cerner go-live dates
  • Stood up a dedicated issue triage queue using the client’s own ticketing system, tagged specifically for Epic Conversion, with a designated captain managing inbound tickets
  • Of approximately 26,000 total post-live tickets, fewer than 79 were attributed to the conversions team — a sub-0.3% issue rate
  • Provided a clean, structured handover as delta loads completed and the migration reached full closure
Results

The migration went live in December 2025 — on schedule, across all three source systems. The outcomes reflected the discipline the DataBridges framework was built to deliver:

Outcome Detail
On-time go-live First go-live in December 2025 and 3 more in February & March 2026
Tier 3 data delivered in full Labs, imaging, notes, documents, problem list/allergies/medications/immunisations
Low post-live issue rate Fewer than 79 of 26,000 total tickets were conversion-related (<0.3%)
Clinical readiness confirmed Facilitated validation sessions verified providers could deliver care on day one
Transparent stakeholder alignment   Weekly workgroup reviews kept clinical, operational, and IT leaders informed throughout
Structured post-live continuity ~5 delta loads completed with clear date-specific data availability communicated at each step
Why It Worked

The DataBridges framework transformed what could have been a chaotic, high-risk project into a disciplined, predictable process. A few things made the difference:

  • Governance was established first. By building a cross-functional workgroup before any data moved, the team ensured alignment, not assumption, drove every scoping decision.
  • Validation was progressive. Starting small and scaling up meant issues were caught early, not at go-live. Clinical SMEs were engaged early and often.
  • Communication was continuous. Stakeholders always knew what was loaded, what was pending, and what was coming. This reduced anxiety and last-minute surprises.
  • The team stayed lean where possible. Downtime windows were minimized; real-time interfaces handled time-sensitive data. Scope discipline kept the project manageable.
  • Issues were contained. A dedicated triage structure ensured post-live issues were captured, tracked, and resolved quickly, without burdening the broader IT team.
Ready to move your data with confidence?

Learn how DataBridges can bring structure, speed, and reliability to your next migration.

healthdatamovers.com/databridges