The challenge
Legacy clinical and resident datasets contained history, diagnoses, transfers, admissions, and fields that did not map one-to-one to the new system. Exceptions needed to be separated from records suitable for migration.
My contribution
- Reviewed source-to-target clinical and demographic mapping rules for the AWARDS to WellSky RCS workstream.
- Applied a structured diagnosis validation workflow using ICD-10 reference definitions, creating categories rather than silently accepting questionable codes.
- Documented rules for admission dates, service categories, event sequences, and review ownership.
- Supported cross-functional design, test scenarios, and UAT-readiness discussions with implementation stakeholders.
Results and scope
In an August 2026 diagnosis audit of 10,894 rows, 8,813 were categorized accurate, 1,981 needed verification, and 100 were classified wrong. That made 2,081 rows actionable exceptions for follow-up. These figures describe review output, not a claim that every exception has since been corrected.
What made the work challenging
Fields that appear simple on paper can carry operational meaning: an admission date may depend on historical transitions; a resident category varies by house; and a diagnosis code should be validated rather than only reformatted. The key deliverable was a traceable set of rules and exceptions.
Public portfolio summary only. Identifying details, clinical content, production screenshots, and confidential employer records are excluded. All diagrams and demo visuals are illustrative.