Home / Insights / Connectivity
Interface engine vs lab middleware: do you need Mirth to connect your lab?
Ask an integration consultant how to connect analysers to an EHR and the answer is often "put an interface engine in the middle". Ask a lab and the answer is often "buy middleware". Both can work. This guide explains what each one is, what you would still have to build on an interface engine, and how to choose.
Written and reviewed by the Catenix team. How these guides are written and checked.
In brief
- An interface engine (for example Mirth Connect, Rhapsody, Cloverleaf or Iguana) is a general tool for routing and transforming healthcare messages between systems.
- Lab middleware is built for analysers: it ships with drivers, specimen matching, QC handling, rules and a lab audit trail.
- An interface engine can connect a lab, but your team builds and maintains the analyser link, mappings, QC logic, reruns and corrections.
- Interface engines shine where many systems exchange many message types; lab middleware shines where analysers must reach a LIS or EHR safely and quickly.
- Catenix is lab middleware with analyser drivers, QC and lockout built in, and it delivers HL7 v2 or FHIR to any interface engine, LIS or EHR you already run.
Take it with you: POCT connectivity buyer's guide · POCT readiness checklist (PDF, no form).
What an interface engine is
An interface engine receives messages from one system, transforms them and sends them to others. It is the switchboard of a hospital's integration: admissions from the patient administration system to dozens of departmental systems, orders from the EPR to the LIS and radiology, results back to the EPR, billing feeds and so on. Well-known products include Mirth Connect, Rhapsody, InterSystems products, Cloverleaf and Iguana. They are flexible, they support HL7 v2, FHIR, files, databases and web services, and an experienced integration engineer can build almost anything with them.
Mirth Connect deserves a note because it is the one small labs hear about most. For many years it was available as open source, which made it a popular free starting point. Licence and support terms for current versions are set by its owner and can change, so check them directly with the vendor before you plan a project around it.
What lab middleware is
Lab middleware is purpose built for the space between analysers and the system of record. Out of the box it includes:
- drivers for named analyser models in ASTM E1394, HL7 v2, POCT1-A2 or proprietary formats, including the low-level link handling
- specimen and order matching, including host query so an analyser can ask which tests to run for a barcode
- QC recognition and filing, with Levey-Jennings charts and Westgard rules
- rules that hold results when QC fails, an operator is not authorised, or a flag needs review
- a lab audit trail showing what the analyser sent and what was stored and released
- delivery to the LIS or EHR in HL7 v2 or FHIR
What you would build yourself on an interface engine
| Capability | Lab middleware | Interface engine |
|---|---|---|
| Analyser link layer (ASTM E1381 framing, checksums, acknowledgements) | Included per analyser | Build or buy a plugin, per analyser |
| Message parsing per analyser model | Included | Build per model |
| Test code mapping | Configured in screens | Build in transformer scripts |
| QC recognition and filing | Included | Build, plus somewhere to store and chart QC |
| Westgard rules and lockout | Included | Build |
| Operator competency checks | Included | Build, plus a competency register |
| Reruns and corrected results | Included | Build and test |
| Store and forward when the network drops | Included in the gateway | Configure queues and alerts |
| Lab audit trail for inspectors | Included | Build from message logs |
None of this is impossible on an interface engine. It is simply work that someone must design, validate against your quality system, document and maintain whenever an analyser or EHR changes. For a hospital integration team that already runs an engine, that may be fine. For a clinic or small lab, it usually is not.
When an interface engine is the right choice
- You already run one, with engineers who maintain it.
- You exchange many message types across many systems, not just lab results.
- You need to route the same results to several destinations with different formats.
- Your lab middleware or LIS will deliver results to the engine, which handles the rest of the estate.
In this last case the two work together: middleware handles the analysers and quality rules, and the engine handles distribution. Catenix often sits exactly there in larger organisations. The boundary should be written down: which system owns test code mapping, which holds QC, and which alerts a person when delivery fails.
When lab middleware is the right choice
- You need analysers connected in days, not a development project.
- Inspectors will ask for QC, competency and an audit trail linked to every result.
- You have no integration engineers, or they are busy with other work.
- You expect to add analysers and want each one to be configuration, not code.
A worked example: one ASTM analyser, two ways
Take one serial immunoassay analyser that sends ASTM E1394 and supports host query. On an interface engine, your team will typically:
- Get the analyser's interface manual from the manufacturer and work through its framing rules: ENQ, frame numbers, checksums, ACK and NAK, EOT and timeouts.
- Build or buy a serial or TCP connector that handles that framing, including retransmission.
- Write a parser for the analyser's record layout, including how it uses components in the test ID field.
- Build a query responder that finds the order for a barcode and replies within the analyser's timeout.
- Write transformers that map test codes, units and flags to the receiving system's format.
- Decide how QC runs are recognised, where they are stored and how they are charted.
- Add alerts, retries and an audit store, then validate everything and document it for your quality system.
On lab middleware with an existing driver, the same job is usually: install the gateway, set the analyser's host settings, map its test codes to your test catalogue in a screen, run QC and a few patient samples, and validate. Validation is the same in both cases. The build steps are what differ. To see whether a driver or connection guide already exists for your analyser, check our analyser connection guides or the open connectivity dataset.
Common mistakes
- Counting the licence and not the engineering. The licence is often the smallest cost of an engine-based lab interface.
- One engineer who understands it. Custom channels need documentation and a second person who can support them.
- QC as an afterthought. If QC is not handled in the interface, it ends up in a spreadsheet with no link to patient results.
- No plan for analyser software updates. A changed record layout can break a parser silently, so retest after every analyser update.
- Choosing by category rather than by job. List the jobs, then see which tool does each one with the least custom work.
Questions to ask before you decide
- Which of our analysers have a ready driver, by model and protocol?
- Who maintains the interface when an analyser or EHR changes, and how is that change validated?
- Where are QC results stored and charted, and can a QC failure hold patient results?
- Can we see the raw analyser message behind any result?
- What happens to results when the network or the receiving system is down?
- What does it cost to add an analyser next year?
Comparing costs honestly
A free or low-cost engine licence can look cheaper than middleware, but the comparison should include engineering time to build each analyser interface, validation effort, the cost of maintaining custom code, and the risk of a key engineer leaving. Middleware priced per site with drivers included turns those unknowns into a fixed monthly line. See how much does a LIS cost for a three-year comparison worksheet.
Where Catenix fits
Catenix is lab and point-of-care middleware that grew into a full platform. It connects analysers through an on-site edge gateway in ASTM E1394, HL7 v2 and POCT1-A2, with drivers, specimen matching, QC with Westgard rules, lockout, competency checks, corrected-result handling, store and forward, and a wire-level audit trail. It sends HL7 v2 or FHIR to your LIS, EHR or interface engine. Analyser not on the list? We build the driver, usually within a couple of days. It is priced per site with no per-test fees and is used in the UK and the US. Catenix is connectivity, workflow and record-keeping software; it does not interpret results and is not marketed as a medical device.
Questions people ask
Is Mirth Connect free?
Older releases were published as open source. Licence terms for current versions are set by its owner and can change, so check them with the vendor before planning a project around it.
Can Mirth connect directly to a lab analyser?
It can receive HL7 over TCP, and with extra work it can handle ASTM. You would still need to build the analyser link layer, parsing, mappings, QC handling and audit trail yourself.
Do I need both an interface engine and lab middleware?
Hospitals often run both: middleware for analysers and quality rules, and an engine to distribute results across many systems. A clinic or small lab usually needs only the middleware.
Can Catenix send results to our interface engine?
Yes. Catenix delivers HL7 v2 or FHIR to an interface engine, LIS or EHR.
Which is faster to go live?
Lab middleware, for analysers it already supports, because the drivers, QC and rules exist. An interface engine project starts with design and build.
Does lab middleware replace an interface engine?
Not in a hospital that already uses an engine across many systems; there the two work together. In a clinic or small lab, middleware usually removes the need for an engine altogether.
How long does it take to build an analyser interface on an interface engine?
It depends on the analyser, the engineer's experience and the validation needed. The link layer, parsing, QC handling and testing all take time, so ask for an estimate per analyser that includes validation.
Can an interface engine file QC results?
It can route QC messages, but it has no QC model of its own. You would need somewhere to store QC by lot and level, chart it, apply Westgard rules and link the outcome to patient release.
Sources and further reading
Read next
More for coordinators.
Analyser connection guides
Step-by-step guides to connecting point-of-care and lab analysers to your LIS or EHR over HL7, POCT1-A2 and ASTM, with bidirectional, wire-level...
Read →ASTM to HL7 conversion explained
How lab middleware converts analyser output in ASTM E1394 into HL7 v2 ORU messages or FHIR: record mapping, test codes, units, flags, QC and...
Read →Catenix device connectivity
Bidirectional point-of-care device connectivity over HL7, POCT1-A2 and ASTM, with LIS/EHR integration and a self-healing edge gateway.
Read →Take it with you
The buyer's guide, in your inbox.
Nine pages on what POCT connectivity is, the standards in plain English, the cost of doing it by hand and eight questions to ask any supplier.
Get the POCT connectivity buyer's guide
Nine pages, vendor-neutral: what connectivity is, the standards in plain English, the cost of doing it by hand and eight questions to ask any supplier. The PDF also opens directly from the downloads page; leave an email only if you want it in your inbox.
See it done in software.
A 30-minute walkthrough on a live tenant, with your analysers and your allowable-error limits.