POCT operator competency: a programme that stands up to assessment

What ISO 15189:2022 and MHRA guidance ask for

ISO 15189:2022 is the international standard for medical laboratories and, in the United Kingdom, the standard the United Kingdom Accreditation Service (UKAS) assesses against. Point-of-care testing (POCT) sits inside it. Annex A, which replaced the separate standard ISO 22870:2016, applies the laboratory's requirements to testing performed near the patient by staff who are usually not laboratory trained.

Clause 6.2 (personnel) sets out what a competence programme must do. The laboratory defines the competence required for each role, provides training, assesses competence against defined criteria before someone works unsupervised, authorises them for specific tasks, reassesses at intervals and keeps records of all of it. Annex A adds that this applies to every person who performs POCT, and that the laboratory remains responsible for the programme even when the operators report to a ward manager, a practice manager or a pharmacy superintendent. Our summary of what ISO 15189:2022 asks of a POCT service covers the other clauses.

In the UK, the Medicines and Healthcare products Regulatory Agency (MHRA) guidance Management and use of IVD point of care test devices says the same in plainer terms: only trained and competency-assessed staff should use POCT devices, records should be kept, and the POCT committee should oversee the training programme. UKAS assessors read the two documents together. Neither prescribes a method or an interval. That is deliberate. The service decides, writes the decision down, and then has to show that it happens.

Two other clauses matter. Clause 7.3.7.3 (external quality assessment) requires EQA samples to be handled by the personnel who routinely perform the examinations, which makes an EQA return evidence about an individual operator and not only about the device. Clause 8.7 (nonconformities and corrective actions) means a failed assessment, or a lapsed operator found still testing, is handled as a nonconformity with a documented action rather than quietly tidied up.

Training is not competency

The most common confusion at assessment is a training record presented as a competency record. They are different things.

Training is the input. The person attended a session, read the standard operating procedure (SOP), watched the manufacturer's video or completed an e-learning module. A training record proves exposure to information.

Competency is the demonstrated output. The person performed the test correctly, on the real device, in the real setting, and someone qualified to judge watched them do it and signed to say so. A competency record proves ability at a point in time, which is why it needs an expiry date.

The distinction has practical consequences. An e-learning certificate on its own is not evidence of competence, because it does not show that the person can collect a sample, run a control, respond to a flag or record a result. Equally, a person can be competent without a formal course, for example an experienced nurse assessed on a device used daily for years. The programme should require both: training as the entry ticket, assessment as the gate, and periodic reassessment as the check that the gate still holds.

Assessors also look for the reverse case: a person assessed once, years ago, who has been running tests ever since with no evidence that anyone looked again. That record shows historic competence, not current competence.

For an international reference point, the Clinical Laboratory Improvement Amendments (CLIA) regulations in the United States, at 42 CFR Part 493, list six elements of competency assessment for non-waived testing: direct observation of test performance, monitoring of recording and reporting, review of QC and proficiency testing records, direct observation of maintenance and function checks, testing of previously analysed or blind samples, and assessment of problem-solving skills. CLIA also requires assessment twice in the first year and annually thereafter. ISO 15189 does not prescribe any of this, but the six elements are a sound checklist in any country.

A workable programme for decentralised sites

A hospital laboratory with one POCT coordinator, a group of GP practices, a chain of private clinics and a community diagnostic centre (CDC) all face the same problem: the operators are many, spread out, turn over often and are not employed by the laboratory. The programme has to be simple enough to run everywhere and strict enough to satisfy an assessor. The elements below are the workable minimum.

Initial training

Cover the device, the SOP, sample collection and handling, patient identification, quality control (QC), what to do when a result is flagged or the device errors, how the result is recorded and where it goes, and how to report an incident. Use the manufacturer's material where it is good, but the SOP is yours and the training must reference it. Record the date, the trainer, the version of the content and the trainee's confirmation.

Direct observation

An assessor watches the operator perform the whole procedure on the device they will use, from patient identification to result recording, against a checklist derived from the SOP. The assessor must themselves be competent on the device and authorised to assess. On a large site, train link nurses or senior operators as assessors, keep the list of assessors, and have the POCT coordinator sample their assessments each year.

QC performance

Running QC correctly, on schedule, and acting on a failure is part of competence. Review the operator's QC history: were controls run when required, were failures acknowledged and repeated rather than ignored, were the results within range. A run of failed QC under one operator's identity while everyone else's passes is a competence signal before it is a device signal. Westgard rules give you an objective way to define a failure.

EQA participation

Where the site is enrolled in an external quality assessment (EQA) scheme, rotate the EQA samples among operators so that participation itself becomes evidence, and record who ran each sample. A poor return traced to one operator is a reason to reassess that person. A poor return across all operators points to a shared cause: investigate the device, method, material handling and training together.

Reassessment interval

The standard does not set one. Annual reassessment is the most common choice in UK POCT policies and is easy to defend. Some services use a longer interval for simple, low-risk devices and a shorter one for complex ones such as blood gas analysers. Whatever you choose, write the rationale into the policy, because the assessor will ask why.

What triggers retraining or early reassessment

1. A gap in practice: no tests run on the device for a defined period, or return from long-term leave.
2. A pattern of QC failures, or a poor EQA return, attributable to the operator.
3. An incident or near miss involving the operator, including a wrong-patient result.
4. A change of device model, test method, SOP or software that alters the procedure.
5. A failed or incomplete reassessment.
6. Transfer from a site that uses different devices for the same test.

How to record it: per person, per device, per test

Records are where programmes fail at assessment. A well-run ward can have every nurse trained and still produce no evidence of it in a form an assessor can use. Clause 6.2.5 (personnel records) requires records of qualifications, training, competence assessment and authorisation to be kept and to be retrievable. The assessor will pick a name from the device's operator list and ask to see that person's record, then pick a result and ask who ran it.

Record competence at three levels:

• Per person: role, site, employer (substantive, bank or agency), assessor and training history.
• Per device: a nurse competent on a blood glucose meter is not competent on a blood gas analyser. Each device type, and where the procedure differs, each model, is a separate line.
• Per test: a multi-analyte device may have tests with different sample handling or acceptance rules. Where the SOP differs by test, so does the competence.

Every line carries a date assessed, an assessor, an outcome and an expiry date. The matrix below is a working format. Small sites often keep it in a spreadsheet, and the last section explains why that stops working past a certain size.

Two habits make the matrix credible. Keep the observation checklist with the record, not in a separate folder, so the assessor sees what was assessed and not just that something was. And keep the assessor's own competency line in the same matrix, so the chain of authority is visible.

Operator lockout as the enforcement mechanism

A competency record on its own is a description. Operator lockout turns it into a control. Most connected POCT devices require the operator to identify themselves, by barcode, PIN or badge, before a patient test can be run. The device, or the middleware behind it, checks that identity against a list of authorised operators. If the person is not on the list, or their competency has expired, the patient test does not run. A QC-only or supervised mode can be permitted so that a person in training can still learn on the device.

Lockout matters because it closes the gap that policies cannot. Without it, a lapsed operator keeps testing because the ward is busy, and the audit trail then shows tests run by someone who was not authorised to run them. The MHRA guidance recommends devices with operator identification and access control for exactly this reason. With lockout, the record and the reality are the same thing.

Three practical points. First, the operator list on the device has to be kept in step with the competency record; if these are two separate systems, someone has to update both, and they drift. Second, a generic or shared login defeats the purpose and is a finding in itself. Third, announce the lockout rule well before it bites, with expiry warnings to the operator and their manager at 60 and 30 days, or the first week of enforcement becomes a service incident.

Common findings at assessment

The findings below recur in accreditation assessments and internal audits of POCT services. Most of them are record problems rather than practice problems, which is what makes them avoidable.

1. Training records presented as competency evidence, with no observation, no assessor and no expiry.
2. No defined competence criteria: an observation was signed, but there is no checklist, or the checklist was not derived from the current SOP.
3. Assessors who were never themselves assessed or authorised to assess.
4. Records that cannot be produced per person and per device on request, or that live only on a site the coordinator cannot reach.
5. Expired competency with patient results still appearing under that operator's identity.
6. Shared or generic device logins, which make every result untraceable to a person.
7. No reassessment interval in the policy, or an interval with no rationale.
8. QC and EQA failures reviewed only per device, never per operator, so individual retraining is never triggered.
9. Bank, agency, locum and student staff outside the programme even though they run tests.
10. A device or method change with no retraining, because the new model "does the same test".

Each of these maps to a clause. The first four are clause 6.2 findings, the fifth and sixth are audit-trail and authorisation findings, the eighth is a clause 7.3.7 finding, and all of them are raised as nonconformities under clause 8.7 with a corrective action expected.

Where software helps

A spreadsheet works for one site and twenty operators. It stops working when there are twelve sites, two hundred operators, five device types and a coordinator who cannot be in every building. The failure mode is always the same: the spreadsheet is out of date, the device operator lists are out of step with it, and nobody knows which results were run by whom.

POCT software should do four things for a competency programme. Record competence per person, per device and per test, with the observation checklist attached. Warn the operator, their manager and the coordinator before expiry. Push the authorised operator list to the devices, or enforce it in the middleware, so that lockout follows the record automatically. And link every result, QC run and EQA sample to the operator identity that ran it, so that competence can be reviewed from the evidence rather than from memory.

Catenix records operator competency and certification against the person who ran each test, with expiry reminders and lockout, and handles the resulting nonconformances and corrective actions in its ISO 15189-aligned quality management module. Whatever software you use, the test is simple: can you produce, in under a minute, the competency record for the operator who ran a given result, and can you show that a lapsed operator could not have run it.

Questions people ask

What is POCT operator competency?

POCT operator competency is the demonstrated ability of a named person to perform a specific point-of-care test on a specific device correctly, from patient identification to recording the result. It is evidenced by direct observation against a checklist, review of the person's quality control and EQA performance, and a signed record with an expiry date. It is different from training, which only shows the person was taught.

How often should POCT competency be reassessed?

ISO 15189:2022 requires periodic reassessment but does not fix an interval; the laboratory sets one and justifies it. Annual reassessment is the most common choice in UK POCT policies. Reassess earlier after a gap in practice, a run of QC failures or a poor EQA return traced to the operator, an incident, or a change of device, method or procedure. In the United States, CLIA requires personnel performing non-waived testing to be assessed twice in the first year and annually thereafter; waived testing has no federal competency interval, though accreditation programmes and good practice add one.

What is the difference between training and competency assessment?

Training is the input: a session, a module or reading the procedure. It proves the person was exposed to the information. Competency assessment is the output: a qualified assessor watches the person perform the test on the real device and confirms it was done correctly, alongside a review of their QC and EQA results. Assessors accept a training certificate as a prerequisite, not as evidence of competence.

Does ISO 15189:2022 require competency assessment for POCT operators?

Yes. Clause 6.2 requires the laboratory to define competence requirements, assess competence before people work unsupervised, authorise them for specific tasks, reassess at intervals and keep records. Annex A, which replaced ISO 22870, applies these requirements to everyone who performs point-of-care testing under the laboratory's responsibility, including nurses, healthcare assistants, pharmacists and GPs who are not laboratory staff.

What is operator lockout in point-of-care testing?

Operator lockout is a control on a connected POCT device or its middleware that prevents a patient test from running unless the operator identifies themselves and has an in-date competency record for that device and test. It is the mechanism that makes the competency record enforceable rather than descriptive. Most systems allow a supervised or QC-only mode so trainees can still practise.

Can a nurse be competent on one POCT device but not another?

Yes, and the record should show it. Competence is assessed per device and, where procedures differ, per test. A nurse competent on a blood glucose meter has not been assessed on a blood gas analyser or an HbA1c analyser, and a new model of the same device may need reassessment if the procedure changed. Each device and test is a separate line in the competency matrix with its own expiry.

Sources and further reading

• ISO 15189:2022 Medical laboratories: requirements for quality and competence (clause 6.2 and Annex A)
• MHRA, Management and use of IVD point of care test devices
• CLIA regulations, 42 CFR Part 493, competency assessment requirements for non-waived testing
• CLSI POCT04, Essential Tools for Implementation and Management of a Point-of-Care Testing Program