Directory Image
This website uses cookies to improve user experience. By using our website you consent to all cookies in accordance with our Privacy Policy.

Why Connecting Epic to Your Genetics Lab Does Not Fix Bad Orders

Author: Sam Here
by Sam Here
Posted: Oct 23, 2026

A genetics lab can have a working Epic or Cerner interface and still lose hours every day to missing requisition fields, double orders and re-keyed patient details. The interface is doing its job. The orders are the problem.

When a clinician orders a hereditary cancer panel, the lab needs far more than a patient name and a test code. It needs the correct test version, the clinical indication, specimen details, consent status, family history where relevant, and some way to know whether that patient already has an open or recently completed order. If any of that is missing, a faster interface just delivers the gap to the accessioning bench sooner.

Duplicates rarely look identical

Labs tend to picture a duplicate as the same order arriving twice. In practice it shows up in less obvious ways:

  • The same test ordered through two different channels
  • An EHR order and a faxed requisition for the same patient
  • A reorder placed under an updated test name
  • Two related panels ordered before the first one is reviewed
  • An order re-entered because the clinic assumed the interface message failed
  • A specimen that arrives with no visible matching order

Not every repeat is a mistake, either. Replacement specimens, reflex testing and new test versions are all legitimate. So a good rule does not reject on sight. It hands the reviewer context: is there an active order, was this test finished recently, is this a new version, is the indication different, and has the clinic been told?

Validate while the clinic can still fix it

The cheapest moment to find a missing indication or consent field is before the order reaches lab operations. Once the specimen has been received, the same gap turns into phone calls, held samples and scattered notes. Structured intake tools such as SmartReq requisition intake software extract and validate requisition data against lab-defined rules, log the outcome, and push only confirmed information downstream. Incomplete orders go to a visible exception queue with an owner, instead of an inbox.

Decide which system owns each field

Most reconciliation pain comes from unclear ownership. A workable split looks something like this: patient demographics and ordering provider come from the EHR, the test definition comes from a governed catalogue, the accession number and specimen status come from the LIS or LIMS, and the final report comes from the reporting system. The exact split varies by lab. What matters is that it was decided on purpose, along with what happens when a message fails, a patient cannot be matched, or a test is retired.

This is where purpose-built HL7 and FHIR order validation for US genetics labs earns its keep. Intergenix validates EHR orders, flags duplicates, and routes failures into worklists so someone owns each exception.

Close the loop on results

The integration is not finished when the order lands. The report has to return to the right chart, encounter and ordering provider, and the receiving side should be able to tell preliminary, amended and final reports apart. Programs that go further and return discrete genomic data rather than a PDF attachment get far more use out of it. The PennChart Genomics Initiative is a good public example of what that looks like at a health system. For the order side, HL7 describes the request itself in its FHIR ServiceRequest specification.

A ten minute self test
  • Do we have one approved source of truth for the test catalogue?
  • Can we spot a possible duplicate before accessioning?
  • Does every incomplete order sit in a queue with a named owner?
  • Can we trace a final report back to the original order and its documents?

If two or more of those are a shrug, the lab does not need another interface. It needs a defined order management process around the interface it already has.

NonStop has written a longer walkthrough of this, covering the seven-step implementation sequence, the build versus buy options and an Epic and Cerner checklist: how to integrate genetic test orders with Epic and Cerner without creating duplicate orders.

Rate this Article
Leave a Comment
Author Thumbnail
I Agree:
Comment 
Pictures
Author: Sam Here

Sam Here

Member since: Oct 20, 2026
Published articles: 1

Related Articles