5 Essential Steps of Testing Strategy during EMR Data Extraction and Conversion

  • Post category:Blog
5 Essential Steps of Testing Strategy during EMR Data Extraction and Conversion

Few people understand the internal complexities of database systems specially when it comes to EMR data extraction. In the pre-assessment planning, the first step is to identify the key people with the business and their expectations.

The initial source database assessment forms the basis of Santeware’s testing strategy, since this provides complete scoping of data available, count of records, quality of data, data relationships, scanned documents scope and data required in various systems.

Santeware adopts different testing methodologies for data to be migrated to Epic versus data to be archived, this allows the QA team to focus on each data set appropriately.

We partner with your EMR data extraction team to conduct thorough validation and testing of this clinical data.

Having a central location data extraction dedicated team members & access to perform validation makes issue tracking and addressing questions from the validators efficient for all stakeholders.

Santeware experts follows multiple steps during EMR data extraction to ensure appropriate testing and validation broken down in four categories:

  • Unit Testing: Each element of the converted data from the legacy system to the target system is confirmed prior to hand off to the clinical team for validation.
  • Partial Validation: This target common issues by utilizing a small sample set of patients. This ensures that large-scale validation will be maximized to find uncommon issues.
  • Large Dataset Validation: Multiple iterations of this validation step are performed within the Santeware project team, by the Client team, and in concert with both teams. As issues are identified at the element level, they are worked on by the project team and retested until all failures are are identified and resolved. . This round of validation also focuses on workflows, ensuring that each data element is functioning correctly while working through patient charts.
  • Practice Full Load prior to cutover: This encompasses the entire patient population and tests all aspects of the extraction including, but not limited to timing, delivery. Loading of all the live patients in scope to a test environment that closely mimics production. The team will, again, identify and resolve any and all errors before the full final load into production.
A detailed diagram of each step is explained here


Ajith Nair is a Principal Consultant at Santeware

Leave a Reply