The title of this article is the very question my current client and I have been trying to solve for many months now.  This health organization had previously done a data migration from their Misys EHR system to Allscripts TouchWorks™, but at that time were unable to migrate the EKGs and Holters from Midmark because of the difficulty of rendering those files in the target EHR.  Recently, we began working on completely archiving their Misys EHR data into the Galen VitalCenter Online Archival (VCO-A) solution and we hit the same snag with the EKGs/Holters.

Ultimately, the challenge was that Midmark has an integration with the legacy EHR (in this case Misys) where the actual result images are stored in the Informix database in a “.car (compressed archive report)” format.  This format is troublesome because most EHRs and archival applications require the file type to be in a more standard format – ie: Tiff, PDF, doc, etc.  Since the data migration had already been completed years ago, we were now solely focused on how to convert these files into our VCO-A environment as PDFs.

We searched far and wide for other folks who encountered and solved this same issue and didn’t find any great leads.  After some internal discussions and trials, we were able to come up with a solution to convert the .car files to PDF format and then treat them just like any other image file.

1. First step was to extract the .car files from the Informix database to a file on disk.
   • In order to associate the patient to the converted PDF, It was important for us to also maintain the metadata.
2. We were then able to leverage the IQManager application to view .car files.
   • This same program can open and save the file to PDF.
   • The system had over 200k .car files and doing this process manually one at a time was not feasible.
3. The next step was to write a script that allowed us to automate the process of the IQManager application opening and saving the .car files as PDFs.
4. We then needed to make sure that we kept a standard naming convention of the files so the PDF matched the name of the .car file and could be associated to the correct patient.
5. Another key piece of the process was that we needed to have a dedicated server running this solution, as it was crucial for the conversion process to run uninterrupted. The entire process to convert all of the .car files to PDF took roughly one week, after which, we were able to proceed with the rest of the archival process.

As with all of our data migrations, we leveraged a plugin-based itinerary pipeline architecture within our GalenETL platform to ensure accuracy, repeatability, audit-ability and efficiency. Data migration best practice requires a robust audit trail to provide a reliable method to track data back to the source. Leveraging our platform with established libraries allowed for ease in automation, intelligent matching, and ultimately, file type conversion while maintaining fidelity.

This is just one example of many that we have encountered when migrating data from one EHR to another and archiving  data into VCO-A. It serves as a reminder that many systems store data in different ways, and it might not always be a straightforward process to back up the data.