Home / The POCT Briefing / No. 1
The result that never reached the record.
Retyped results, the middleware-or-LIMS question, a copyable RFP, and what community diagnostic centres have to get right. This is the web version of the September issue; sign up to get the next one by email.
Ask a POCT coordinator where their service loses time and the answer is rarely the analyser. It is the step after it: a number read off a screen, typed into a record, copied onto a report, then into an invoice. Every copy is a chance to get it wrong, and the error is usually found by the wrong person, later.
This first issue is about that step: how often it goes wrong, what it costs, which software actually removes it, and how to write a tender that gets you the right kind.
Lead: transcription errors in POCT
Transcription errors: causes, cost and how to remove them. Transposed digits, wrong patient, wrong unit, a decimal in the wrong place, a flag that never made it across. The guide lists the error types, explains why they are under-reported, compares manual, semi-automated and connected workflows in one table, and ends with an audit you can run this month.
Also this month
02POCT middleware vs LIMS: what each does and which you need
The three architectures you will actually meet, and a decision table by setting: acute trust, community diagnostic centre, private clinic, GP practice, occupational health.
03A POCT software RFP you can copy
Fifty-five requirements in nine groups, each with a priority column and a space for the vendor's answer, plus the mistakes that make POCT tenders go wrong.
04Community diagnostic centres: what the connectivity has to do
Results into the LIMS and the referring clinician's record, operator lockout, QC and EQA evidence across sites, and a readiness checklist for a CDC that still runs POCT on paper.
05One page to send to IT: how results reach Epic, EMIS Web, SystmOne and the LIMS
Eighteen plain-language integration guides, UK and US, each describing the interface the clinical system accepts and what your integration team has to confirm.
From the bench
Decide the acceptance criteria before you look at the data. When you verify a new analyser, write down the total allowable error you will accept (CLIA limits, RCPA specifications or biological variation) before the first replicate is run. Deciding after the data are in is how verifications quietly turn into opinions. The full sequence, EP15 through EP28, is in How to verify a new POCT analyser.
And one thing about Catenix
Everything above is vendor-neutral. If you want to see the retyping step removed on your own analysers, with the result filed against the right patient in EMIS Web, SystmOne, your LIMS or your EPR, a 30-minute walkthrough on a live tenant is the fastest way to judge it. Book a walkthrough, or take the readiness checklist (PDF) first.
Written and reviewed by the Catenix team. How these are written and checked. Catenix does not interpret clinical results and provides no clinical decision support.
Next issue: October
Who is allowed to press the button.
Operator competency, lockout, and the records an assessor asks for. Get it by email.
Get the POCT Briefing
One email a month for the people who run point-of-care testing. One theme, four to six things worth reading, one tip you can use the same week. No clinical advice, no invented statistics, one button about Catenix at the end.