Home / Insights / Quality control

QC lockout and operator lockout for point-of-care analysers

Quality controlPublished 2026-11-1710 min readInternational

A practical guide to QC lockout and operator lockout in point-of-care testing: what triggers each one, how they are enforced on the device or in software, how to release them safely, and what the records must show an assessor in the UK and the US.

Written and reviewed by the Catenix team. How these guides are written and checked.

In brief

  • A QC lockout stops patient testing on an analyser, or holds its patient results, when quality control is overdue, has failed, has broken a Westgard rule or was run on an expired control lot.
  • An operator lockout stops a named person from running patient tests when their competency or certification for that device and test is missing or expired.
  • Analysers that support POCT1-A2 device management can usually be locked and unlocked remotely; analysers that cannot be locked need a software-side result hold instead, so enforcement always depends on the analyser's interface.
  • Releasing a lockout should require a documented corrective action and a passing QC run, recorded against the person who released it, because that record is what an ISO 15189, UKAS, CQC, CLIA or CAP assessor will ask to see.
  • Badly tuned lockouts that fire on noise teach staff to work around the system, so rule selection, QC schedules and expiry warnings matter as much as the lock itself.

Take it with you: POCT connectivity buyer's guide · POCT readiness checklist (PDF, no form).

What a QC lockout is, and where it lives

A quality control (QC) lockout is a rule that stops patient testing when the evidence that an analyser is working correctly is missing or bad. An operator lockout is the same idea applied to people: a patient test cannot be run by someone who has no in-date record of competence on that device. Together they turn a point-of-care testing (POCT) policy from a document into a control.

There are two places a lockout can live, and the difference matters when you buy or configure a system.

Device-level lockout

The analyser itself refuses to run a patient sample. Many connected devices hold their own QC schedule and operator list, and will show a message such as "QC required" or "operator not authorised" at the point of testing. Some enforce this with settings held only on the device. Others take their rules and their operator list from a data manager or middleware over a network connection, which means the rules can be changed centrally and the lock can be applied or lifted remotely.

Software-level lockout

The analyser runs the test, but the software that receives the result holds it. The result is flagged, kept out of the patient record or the laboratory information system (LIS), and placed in a review queue until someone with authority releases or rejects it. This is the fallback for devices that cannot be locked remotely, and it is also a useful second layer behind a device lock, because it catches anything that slips through.

A device-level lock is stronger: the patient result never exists. A software-level hold is weaker but still meaningful: the result exists, but it cannot reach the record or the clinician until it has been reviewed. Most real services run a mix of the two, because most fleets mix device types.

What should trigger a lockout

The triggers below are the ones most services configure. Not every trigger suits every device. Write down which ones apply to which analyser, and why, because an assessor will ask.

TriggerWhat it meansTypical response
QC dueThe scheduled QC interval (per shift, daily, weekly, per new lot or after maintenance) has passed without a QC run.Warn before the deadline, then lock or hold patient testing until QC is run and passes.
QC failedA control result falls outside its acceptable range.Lock or hold immediately, investigate, repeat QC, release only after a pass and a documented review.
Westgard rule violationA rejection rule is broken, for example 1-3s (one result beyond three standard deviations), 2-2s (two consecutive results beyond two standard deviations on the same side) or R-4s (a range of four standard deviations within a run).Lock or hold on rejection rules; treat warning rules such as 1-2s as a prompt to look, not a lock.
Expired control or reagent lotQC was run on control material or reagent past its expiry or open stability date, so the QC result proves nothing.Reject the QC run, lock until QC is repeated on in-date material.
Expired operator certificationThe person logged in has no current competency record for this device and test.Block patient testing for that person, allow QC-only or supervised mode if your policy permits.
Unknown or unregistered operatorThe operator ID entered is not on the authorised list at all.Block patient testing and log the attempt.

The Westgard rules deserve a note. They were designed so that a combination of rules catches real error while keeping false rejections low. Locking on 1-2s alone, which a control will breach by chance roughly once in twenty runs, produces a stream of false lockouts. Our guide to Westgard rules for POCT teams explains which rules to use as warnings and which as rejections, and the Levey-Jennings and Westgard reference shows how they look on a chart.

How lockout works in practice: POCT1-A2 and the rest

Whether a lockout can be enforced at the device depends on how the analyser talks to the outside world. This is the single most important question to settle before you promise a ward or a clinic that a failed QC will stop testing.

Analysers with POCT1-A2 device management

POCT1-A2 is the connectivity standard published by the Clinical and Laboratory Standards Institute (CLSI) for point-of-care devices. As well as carrying results, it defines device management messages. These include an operator list that the data manager can send to the device, and directives that ask the device to change state, which many devices use to lock and unlock patient testing. With a device like this, the middleware decides that QC has failed or that an operator has lapsed, and the analyser is told to stop accepting patient samples or that person's ID. Support for these messages varies by manufacturer, model and firmware, so confirm it device by device. Our POCT1-A2 guide covers the message types in more detail.

Analysers that cannot be locked remotely

Many analysers connect over HL7 v2 or ASTM E1394, or through a serial or USB link, and only send results out. They do not accept a remote lock instruction. Some have their own on-board QC lockout and operator list configured through the device menu, which works but has to be kept in step with the central record by hand. For the rest, the control has to sit in software: the result arrives, the software sees that QC was failed or overdue, or that the operator was not authorised, and holds the result instead of releasing it to the record.

A software hold needs two things to be credible. The held result must not reach the patient record, the LIS, the electronic patient record (EPR) or, in US terms, the electronic health record (EHR), until it is released. And the device screen still shows a number to the operator, so your procedure has to say clearly that a result shown on the device but held in software must not be acted on or written down. Train for that explicitly.

Paper and standalone devices

Devices with no connection at all cannot be locked by software. The control is the procedure, the QC log and the supervisor. If a device of this kind is used for decisions that matter, connecting it, or replacing it with one that can be connected, is usually the real fix.

Operator lockout tied to competency and training records

Operator lockout only works if the list of authorised operators is the same thing as the competency record. When they are two separate lists, one on the device and one in a spreadsheet, they drift: a nurse whose competency expired last month is still on the device, and a new starter assessed yesterday cannot log in.

The workable model is a single competency record, held per person, per device and per test, with an assessor, a date and an expiry on every line. The authorised operator list on each device, or the operator check in the middleware, is generated from that record. When the expiry date passes, the person drops off the list. When they are reassessed, they come back on. Our article on building a POCT operator competency programme covers how to assess and record competence in a way that supports this.

Four details make operator lockout fair and safe:

  • Warnings before expiry. Tell the operator and their manager at 60 and 30 days. Nobody should discover their lapse at the bedside.
  • A training or supervised mode. A person being trained should be able to run QC or supervised tests, recorded as such, without being blocked from learning.
  • No shared logins. A generic ward ID defeats the lock and is a finding in its own right.
  • An urgent-care route. Decide in advance what happens if the only person on shift is locked out in an emergency: a named alternative operator, a laboratory sample, or a documented override with review. Do not leave it to improvisation.

How to release a lockout safely

The release is where a lockout earns or loses its value. A lock that anyone can clear with one tap gives a false sense of control. A release that needs three signatures and a phone call to the laboratory will be bypassed. The checklist below is a sound middle ground for a QC lockout.

  1. Stop and record. The lockout itself is logged automatically with the device, the trigger, the time and the operator.
  2. Investigate the likely cause. Check the control and reagent lots and their expiry and open dates, storage temperature, sample handling, recent maintenance and any device error codes.
  3. Take corrective action. Replace the control material, recalibrate, clean, or take the device out of service, as the procedure says.
  4. Repeat QC and get a pass. The release should depend on a passing QC run, not on someone deciding the failure did not matter.
  5. Review patient results since the last good QC. Decide whether any patient results run since the last passing QC need to be reviewed or repeated, and record the decision. This step is often missed.
  6. Release, by an authorised person. Define who can release: the POCT coordinator, a named senior operator or a laboratory scientist. The release is recorded against that person's identity, with the reason.
  7. Raise a nonconformity if it repeats. Repeated failures on one device or under one operator are a pattern to investigate formally, not a series of isolated events.

For an operator lockout, the release is a completed competency assessment entered by an authorised assessor. Nobody should be able to restore an operator by editing an expiry date without an assessment behind it.

Avoiding false lockouts that teach staff to work around the system

Every unnecessary lockout costs trust. Staff who are locked out for no good reason, at a busy moment, learn to keep a spare login, to run QC twice until it passes, or to write the device result on paper. At that point the control is worse than none, because the records say one thing and the practice says another.

The common causes of false lockouts are fixable:

  • Locking on warning rules. Use 1-2s as a prompt, and lock on rejection rules such as 1-3s, 2-2s and R-4s.
  • Wrong target values or ranges. Ranges copied from a package insert without local verification, or not updated when a new control lot arrives, trigger failures that are not real. Set and review ranges for each lot.
  • QC schedules that do not match the shift pattern. A daily QC deadline at 08:00 on a ward whose first test is at 07:30 creates a lockout every morning. Schedule QC around how the site actually works.
  • No warning before a deadline. Prompt QC and competency renewals before they fall due, on the device and to the coordinator.
  • Repeat-until-pass. Make repeated QC runs visible, so that one failure followed by two passes is reviewed rather than hidden.

Review lockout frequency per device and per site every month. A device that locks out every week either has a real problem or a badly set rule, and both are worth finding.

What the records must show an assessor

The standards do not require a lockout as such. They require that quality control is performed and acted on, that only competent people test, and that you can prove both. A lockout is the most reliable way to make that true, and its log is the most direct evidence that it is.

United Kingdom

ISO 15189:2022 is the standard the United Kingdom Accreditation Service (UKAS) assesses medical laboratories against, and Annex A applies it to POCT. Clause 7.3.7.2 requires internal quality control, with results reviewed against defined criteria and patient results withheld when QC indicates a problem. Clause 6.2 covers personnel competence and authorisation. Clauses 7.5 and 8.7 cover nonconforming work and corrective action. The Medicines and Healthcare products Regulatory Agency (MHRA) guidance on the management and use of IVD point of care test devices recommends devices with operator identification and QC controls. For services regulated by the Care Quality Commission (CQC), the relevant tests are safe care and treatment (Regulation 12) and good governance (Regulation 17) under the Health and Social Care Act 2008 (Regulated Activities) Regulations 2014.

United States

The Clinical Laboratory Improvement Amendments (CLIA) regulations at 42 CFR Part 493 require control procedures for non-waived testing, including that patient results are not reported when QC is unacceptable, and require competency assessment for testing personnel. The College of American Pathologists (CAP) Point-of-Care Testing Checklist covers QC, operator training and competency for accredited sites. Our CAP point-of-care readiness guide covers inspection preparation in more depth.

What to have ready

  • For any patient result: who ran it, their competency status at that moment, and the QC status of the device at that moment.
  • For any QC failure: the lockout, the investigation, the corrective action, the repeat QC, the review of affected patient results, and who released the device.
  • For any operator: their competency lines per device and test, with assessor and expiry, and evidence they could not test while lapsed.
  • An audit trail that shows these records were not changed after the event, or shows every change with who made it and when.

What to look for in the software, and a short implementation plan

Questions to ask a vendor

  1. For each of our analyser models, can you lock patient testing at the device, or only hold results in software? Show us which.
  2. Which triggers can lock or hold: QC due, QC failed, which Westgard rules, expired lots, expired operator competency?
  3. Is the authorised operator list generated from the competency record, and how quickly does a change reach the device?
  4. Who can release a lockout, what must they record, and is a passing QC run required before release?
  5. What happens when the network is down: does the lock still hold, and are results held until reviewed?
  6. Can we report lockout frequency, time locked and releases by device, site and operator?
  7. Is the audit trail tamper-evident, and can it show every lock, release and override with the person responsible?

Catenix captures QC as it happens, applies Westgard multi-rule checks, and can hold a device or its patient results until a failure is reviewed, with operator competency recorded against the person who ran each test and every lock and release kept in a tamper-evident audit trail. Whether the lock is applied at the device or as a software result hold depends on the analyser's interface, and we confirm that per model before go-live. There is a fuller overview on our QC lockout and operator competency page.

A short implementation plan

  1. Inventory the fleet. List every analyser, its interface and whether it can be locked remotely, locked on board, or only held in software.
  2. Clean the competency record. One line per person, device and test, with assessor and expiry. Remove shared logins.
  3. Set the rules. QC schedules matched to shift patterns, rejection rules per analyte, lot ranges verified, warning periods defined.
  4. Write the release procedure. Who releases, what they record, how affected patient results are reviewed, and the emergency route.
  5. Run in warning-only mode first. For two to four weeks, log what would have locked without locking. Fix false triggers.
  6. Switch on enforcement site by site. Announce it, train it, and review lockout data monthly at the POCT committee.

Questions people ask

Is there software that locks out a point-of-care analyser when QC fails?

Yes. POCT middleware and data management software can lock an analyser, or hold its patient results, when QC is overdue, fails, breaks a Westgard rejection rule or was run on an expired lot. Where the analyser supports remote device management, usually through POCT1-A2, the lock is applied at the device. Where it does not, the software holds results until they are reviewed. Check enforcement model by model, because it depends on the analyser's interface.

What is the difference between a QC lockout and an operator lockout?

A QC lockout acts on the device: patient testing stops, or results are held, because the evidence that the analyser is working is missing or failed. An operator lockout acts on the person: a named operator cannot run patient tests because their competency record for that device and test is missing or expired. Both are usually configured in the same system and both should be logged with who released them.

Can every POCT analyser be locked remotely?

No. Analysers that support POCT1-A2 device management can often receive an operator list and lock or unlock instructions from a data manager, though support varies by model and firmware. Many analysers that send results over HL7 v2, ASTM or a serial link cannot be locked remotely. For those, the control is a software-side result hold, an on-board lockout configured on the device, or a written procedure.

Who should be allowed to release a QC lockout?

Define it in your POCT policy. Common choices are the POCT coordinator, a laboratory scientist or a named senior operator who has been trained in QC troubleshooting. Whoever it is, the release should require a documented corrective action and a passing repeat QC run, plus a decision about patient results run since the last good QC, all recorded against that person's identity in the audit trail.

Which Westgard rules should trigger a POCT lockout?

Lock on rejection rules and use warning rules as prompts. Many POCT services lock on 1-3s, 2-2s and R-4s, and treat 1-2s as a warning to look at the chart, because a single control breaches two standard deviations by chance about once in twenty runs. The right selection depends on the test's performance and the number of controls, so agree it with the laboratory and document the rationale.

What evidence of lockouts do UKAS, CQC, CLIA or CAP assessors look for?

They look for proof that QC failures stopped or held patient testing, that only competent operators ran tests, and that each failure was investigated and released by an authorised person. In practice that means a log of every lockout and release, the corrective action and repeat QC, a review of affected patient results, per-operator competency records with expiry dates, and an audit trail showing nothing was changed after the event.

Sources and further reading

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.

We use your email only to send the guide and, if you ask, to follow up once. No newsletter unless you tick the box. See our privacy notice.

Thank you. The guide is on its way. You can also open it now.
That did not send. Open the guide directly or email contact@catenix.com.

See it done in software.

A 30-minute walkthrough on a live tenant, with your analysers and your allowable-error limits.