EDITIONEN·FR·ქართ
ASF Logo

Accréditation Sans Frontières

International Accreditation of Healthcare Facilities

Start accreditation  
ASF Standards · Laboratory · Standard 7

Standard 7 — Examination: Process & Safety

8 criteria · 6 non-negotiable · 2 core · Version 1.0

Criteria in this standard

7.1 — Method Validation Before Clinical Use
7.2 — Internal Quality Control Run Before Patient Results Released
7.3 — External Quality Assessment Participation
7.4 — Critical Value Communication, Verified Received
7.5 — Personal Protective Equipment, Available and Used
7.6 — Exposure Incident: Documented Response Protocol
7.7 — Information System Access Controlled and Logged
7.8 — Result Amendment Traceable to Original
7.1

Method Validation Before Clinical Use

Non-Negotiable

Every examination method used for reportable patient results has documented validation or verification data — accuracy, precision, reportable range, reference intervals — completed before the method was first used clinically, not retrospectively assembled.

In plain terms: Every test method was properly checked and proven to work correctly before it was ever used on a real patient — not after, as paperwork catching up.

Facility category Standalone lab Hospital lab Clinic lab
Applicability Full Full Full

Why this matters

Validation completed after a method is already in clinical use defeats its entire purpose — the point is to confirm a method works correctly before any patient result depends on it, not to retroactively document that it probably worked. A validation date that falls after the method’s first recorded clinical use is a clear, specific signal that paperwork was assembled to satisfy a review rather than to genuinely protect patients.

What good looks like

  • Validation completion predates first clinical use, verifiably.
  • Manufacturer data is locally verified, not relied on alone.
  • Reference intervals match the laboratory’s actual patient population.

Common failure modes

  • Validation paperwork is dated after clinical use already began.
  • Manufacturer insert data is used without any local verification.
  • Reference intervals are copied wholesale without population relevance review.

Worked example

In practice
A laboratory introducing a new point-of-care glucose method.
BeforeA new point-of-care device was deployed to a clinic site and used clinically for several weeks before the central laboratory got around to completing formal validation documentation, which was assembled retrospectively using data gathered after clinical use had already begun.
ActionThe laboratory instituted a hard rule: no new method or device goes into clinical use anywhere, including satellite sites, until validation is complete and formally signed off by the quality manager.
AfterThe Monitor reviewed the most recently introduced method and found validation completion dated eleven days before its first recorded clinical use. Criterion verified.

If you are starting from zero — do this first

  1. Pull validation records for your most recently introduced method and check the dates.
  2. Confirm manufacturer data has been locally verified, not simply adopted.
  3. Check reference intervals against your own patient population’s characteristics.
The most common mistake: Deploying a new method to a satellite or point-of-care site quickly, with formal validation documentation only catching up afterward because the central laboratory’s usual process didn’t extend to that location in practice.

Self-assessment questions

1. Pick a currently used method — does validation data exist with a completion date before the method’s first clinical use date? — Validation dated after clinical use began indicates retrospective paperwork, not real verification.
Evidence: Validation record with completion date
2. Where a manufacturer’s validation data is relied upon instead of in-house validation, has the laboratory verified it performs equivalently in its own hands? — Manufacturer data alone, without local verification, is insufficient for most method categories.
Evidence: Local verification record
3. Are reference intervals appropriate to the actual patient population served, not simply copied from a manufacturer insert? — A population mismatch between the reference source and the laboratory’s actual patients is a real, specific risk.
Evidence: Reference interval verification record

Common reasons for a PARTIAL answer

  • Satellite or point-of-care deployments bypass the central validation process.
  • Reference intervals are adopted wholesale without population review.

Implementation plan

When What
Week 1 Audit recent method introductions for validation-before-use dating.
Week 2 Build a hard rule extending to every site, including satellite locations.
Week 3 Review reference intervals for population appropriateness.
Ongoing Require validation sign-off before any new method deployment, anywhere.

How the Monitor verifies this

Method What Detail
DOCUMENT Validation date cross-check Compares validation completion date against first recorded clinical use date.

Evidence base

International Organization for Standardization. ISO 15189:2022, Medical laboratories — Requirements for quality and competence, Clause 7.3.2. Geneva: ISO; 2022.
Clinical and Laboratory Standards Institute. EP15-A3: User Verification of Precision and Estimation of Bias. Wayne (PA): CLSI; 2014.
7.2

Internal Quality Control Run Before Patient Results Released

Non-Negotiable

Internal quality control is run and reviewed as acceptable before any patient result from that run is released, with a written, followed procedure for what happens when quality control fails.

In plain terms: Quality control gets checked and approved first, before any real patient results from that same run go out — every single time, no exceptions.

Facility category Standalone lab Hospital lab Clinic lab
Applicability Full Full Full

Why this matters

Releasing patient results before confirming quality control is acceptable inverts the entire purpose of quality control — the check exists specifically to catch a problem before it reaches a patient, and releasing first removes that protection entirely. This failure mode often emerges not from carelessness but from workflow pressure, where results are queued for release automatically while quality control review happens “in parallel” rather than strictly first.

What good looks like

  • No instance exists of results released before quality control review.
  • A real, tested response procedure exists for quality control failure.
  • Quality control frequency is justified, not just minimized.

Common failure modes

  • Automated systems release results before a human confirms quality control passed.
  • A written failure-response procedure exists but has never actually been followed in practice.
  • Quality control frequency is set at the bare minimum with no documented justification.

Worked example

In practice
A chemistry analyzer with automated result release configured by default.
BeforeThe laboratory information system’s default configuration released patient results automatically as they were generated, with quality control results reviewed separately afterward by a technologist, rather than release being gated on prior quality control confirmation.
ActionThe quality manager reconfigured the system so result release is explicitly held pending confirmed quality control acceptance for that specific run, closing the gap between generation and gated release.
AfterThe Monitor reviewed system configuration and recent run logs and found no instance of results released ahead of quality control confirmation. Criterion verified.

If you are starting from zero — do this first

  1. Check your actual system configuration — is release truly gated on quality control, or just reviewed in parallel?
  2. Test your quality control failure response procedure for real, don’t just read it.
  3. Document the justification for your current quality control frequency.
The most common mistake: Assuming that reviewing quality control “shortly after” result release is equivalent to reviewing it before release, when the sequence itself is what protects the patient.

Self-assessment questions

1. Is there any instance of patient results being released before quality control for that run was reviewed? — A single such instance is a direct patient-safety failure, not a minor process gap.
Evidence: System configuration and release logs
2. When quality control fails, is there a documented instance of the correct response being followed — investigation, correction, re-run — before results were released? — Not just a written procedure that’s never actually been tested in practice.
Evidence: Quality control failure response record
3. Is quality control frequency appropriate to test volume and stability, not merely the minimum the laboratory can get away with? — A written justification for the chosen frequency, not an arbitrary number.
Evidence: Frequency justification document

Common reasons for a PARTIAL answer

  • Automated release isn’t genuinely gated on prior quality control confirmation.
  • The failure-response procedure exists but has never actually been exercised.

Implementation plan

When What
Week 1 Audit system configuration for true release-gating on quality control.
Week 2 Reconfigure if a gap is found.
Week 3 Document quality control frequency justification.
Ongoing Periodically audit for any release-before-review instance.

How the Monitor verifies this

Method What Detail
DOCUMENT Configuration and log audit Checks system configuration and historical logs for any release-before-review instance.

Evidence base

International Organization for Standardization. ISO 15189:2022, Medical laboratories — Requirements for quality and competence, Clause 7.3.5. Geneva: ISO; 2022.
Westgard JO. Basic QC Practices, 4th ed. Madison (WI): Westgard QC; 2016.
7.3

External Quality Assessment Participation

Non-Negotiable

The laboratory participates in an external quality assessment or proficiency testing scheme for every discipline it operates where a scheme is available, with results reviewed by the accountable person and unsatisfactory performance investigated.

In plain terms: The lab proves to an outside party, regularly, that it’s actually getting results right — not just trusting its own internal checks.

Facility category Standalone lab Hospital lab Clinic lab
Applicability Full Full Full

Why this matters

Internal quality control confirms consistency within a laboratory’s own system, but it cannot reveal a systematic bias shared across that entire system — external quality assessment is specifically designed to catch exactly that blind spot by comparing results against an independent, external benchmark. A discipline quietly excluded from participation, often simply because it’s smaller or newer than the laboratory’s main services, loses this safeguard entirely without anyone necessarily deciding that deliberately.

What good looks like

  • Every operated discipline with an available scheme is currently enrolled.
  • Results are genuinely reviewed by the accountable person, not just filed.
  • Unsatisfactory performance triggers a real, documented investigation.

Common failure modes

  • A newer or smaller discipline was never enrolled in a scheme at all.
  • Certificates are filed without genuine review of the underlying performance.
  • An unsatisfactory result has no documented follow-up.

Worked example

In practice
A laboratory that recently added a small coagulation testing service.
BeforeThe laboratory’s established chemistry and hematology sections were enrolled in external quality assessment, but the recently added coagulation service, run by just one technologist, had never been enrolled — nobody had specifically added it when the service launched.
ActionThe quality manager built a checklist requiring external quality assessment enrollment as a mandatory step before any new testing discipline launches, and immediately enrolled the coagulation service.
AfterThe Monitor confirmed current, active enrollment across every operated discipline, including the recently added coagulation service, with documented review of the most recent round’s results. Criterion verified.

If you are starting from zero — do this first

  1. List every testing discipline currently operated, including smaller or newer ones.
  2. Check enrollment status for each against available schemes.
  3. Enroll any gaps found immediately.
The most common mistake: A newer or smaller testing service launches without external quality assessment enrollment being explicitly built into the launch checklist, simply falling through the cracks of an otherwise solid program.

Self-assessment questions

1. Does current enrollment and recent participation exist for every operated discipline with an available scheme? — A discipline quietly excluded from participation is a specific, checkable gap.
Evidence: Enrollment records
2. Is there documented review of each round’s results by the accountable person, not just filing the certificate? — Participation without genuine review of the results defeats the purpose.
Evidence: Review sign-off record
3. Where performance was unsatisfactory, was a documented investigation and corrective action completed? — An unsatisfactory result with no follow-up record is a significant gap.
Evidence: Investigation record

Common reasons for a PARTIAL answer

  • A newer discipline was never formally enrolled.
  • Certificates are filed but genuine review isn’t documented.

Implementation plan

When What
Week 1 List all operated disciplines and check enrollment status.
Week 2 Enroll any gaps found.
Week 3 Build a mandatory enrollment step into new-service launch checklists.
Ongoing Review every round’s results with the accountable person.

How the Monitor verifies this

Method What Detail
DOCUMENT Enrollment and review check Verifies current enrollment for every operated discipline and checks for documented result review.

Evidence base

International Organization for Standardization. ISO/IEC 17043:2023, Conformity assessment — General requirements for the competence of proficiency testing providers. Geneva: ISO; 2023.
International Organization for Standardization. ISO 15189:2022, Medical laboratories — Requirements for quality and competence, Clause 7.3.6. Geneva: ISO; 2022.
7.4

Critical Value Communication, Verified Received

Non-Negotiable

Critical or panic values are communicated directly to a responsible clinician without delay, with verification that the message was actually received and understood — not left on voicemail or sent by a method with no confirmation of receipt — and the communication is logged.

In plain terms: A genuinely dangerous result reaches a real clinician, directly, fast — and the lab actually confirms it landed, not just that a call was attempted.

Facility category Standalone lab Hospital lab Clinic lab
Applicability Full Full Full

Why this matters

A critical value left on voicemail or sent through a channel with no delivery confirmation can sit unseen for hours, during which exactly the urgent clinical response the alert was meant to trigger never happens. The read-back verification step — having the receiving clinician repeat the value back — exists specifically because verbal communication under time pressure is genuinely error-prone, and this is the step most often skipped when staff are busy.

What good looks like

  • A written list defines exactly what counts as a critical value, per analyte.
  • Communication logs show genuine read-back verification, not just a call attempt.
  • A defined escalation path exists for a failed first contact attempt.

Common failure modes

  • “Critical” is left to individual judgment with no written definition.
  • A voicemail is left and treated as communication complete.
  • No escalation path exists if the first call attempt isn’t answered.

Worked example

In practice
A critical potassium result identified late in the evening shift.
BeforeThe technologist called the ordering physician’s listed number, reached voicemail, left a message describing the result, and logged the call as “critical value communicated” with no further action, since the call itself had been attempted.
ActionThe laboratory built a formal escalation procedure requiring a second and third contact attempt through alternative channels, including the covering physician or charge nurse, if the first attempt isn’t answered within a defined window, with read-back verification required at whatever point contact is made.
AfterThe Monitor reviewed the communication log for a recent critical value and found a documented read-back verification with the responding clinician, completed within the required window. Criterion verified.

If you are starting from zero — do this first

  1. Confirm a written critical value list exists for your major analytes.
  2. Check whether your communication log shows read-back verification or just call attempts.
  3. Build a defined escalation path for failed first contact.
The most common mistake: Treating a voicemail message as completed critical value communication, when the entire point of the requirement is confirmed, verified receipt by a responsible clinician.

Self-assessment questions

1. Is there a written list of what constitutes a critical value for each relevant analyte? — Without a defined list, “critical” becomes a matter of individual judgment rather than a consistent system.
Evidence: Written critical value list
2. Does the communication log show read-back verification or equivalent confirmation of understanding, not just that a call was placed? — A call attempted is not the same as a value successfully communicated and understood.
Evidence: Communication log with read-back record
3. Is there a documented escalation path when the first contact attempt fails to reach a responsible clinician? — A single failed call attempt with no escalation leaves a critical value effectively unreported.
Evidence: Escalation procedure

Common reasons for a PARTIAL answer

  • Voicemail is treated as sufficient communication.
  • No escalation path exists for a failed first attempt.

Implementation plan

When What
Week 1 Confirm or build a written critical value list.
Week 2 Build a required read-back verification step into the communication log.
Week 3 Build and brief staff on an escalation path for failed contact.
Ongoing Audit critical value logs for genuine read-back completion.

How the Monitor verifies this

Method What Detail
DOCUMENT Communication log review Reviews recent critical value communications for read-back verification and timely resolution.

Evidence base

World Health Organization. Communication During Patient Hand-Overs. Patient Safety Solutions, Volume 1, Solution 3. Geneva: WHO; 2007.
College of American Pathologists. Critical Values/Critical Results Reporting. Northfield (IL): CAP; 2021.
7.5

Personal Protective Equipment, Available and Used

Non-Negotiable

Appropriate personal protective equipment is available at the point of use for every task requiring it, in adequate supply, and observed to be actually worn, not merely available but unused under time pressure.

In plain terms: The right protective gear isn’t just sitting somewhere in the lab — it’s right there where it’s needed, and staff actually wear it, genuinely, even when busy.

Facility category Standalone lab Hospital lab Clinic lab
Applicability Full Full Full

Why this matters

PPE compliance is one of the most well-documented gaps between written policy and actual practice — availability alone says nothing about use, and self-reported compliance is notoriously unreliable since staff know the expected answer. Direct observation during an actual, unannounced moment of bench work is the only way to genuinely assess whether PPE use is a real habit or a policy that exists mainly on paper.

What good looks like

  • PPE is actually observed worn correctly during real bench work.
  • PPE is physically located at the point of use, not requiring a trip to retrieve.
  • Any stockout is logged as a safety event, not quietly absorbed.

Common failure modes

  • PPE is available but not consistently worn under time pressure.
  • PPE is stored somewhere inconvenient, discouraging use.
  • Stockouts happen occasionally and go unrecorded.

Worked example

In practice
A busy microbiology bench during a high-volume morning.
BeforeGloves and eye protection were stored in a supply cabinet at the far end of the room. During a high-volume morning, several staff members worked without eye protection because retrieving it from across the room felt disruptive to workflow.
ActionThe laboratory relocated PPE supplies directly to each bench station, eliminating the retrieval friction, and the supervisor began conducting brief, unannounced compliance spot-checks during actual peak periods.
AfterThe Monitor observed bench work during a genuinely busy period and found consistent, correct PPE use across all observed staff. Criterion verified.

If you are starting from zero — do this first

  1. Observe actual bench work during a busy period, not a calm one.
  2. Check PPE’s physical location relative to each workstation.
  3. Relocate supplies if retrieval friction is discouraging use.
The most common mistake: Storing PPE in a central supply location rather than directly at each point of use, creating just enough friction that compliance quietly drops during busy periods.

Self-assessment questions

1. Observe actual bench work — is PPE being worn correctly for the task being performed? — Observed directly, not asked about, since self-report on safety compliance is unreliable.
Evidence: Direct observation, ideally unannounced
2. Is PPE actually available at the point of use, not stored somewhere requiring staff to leave the bench to retrieve it? — Inconvenient access is a well-documented cause of PPE non-compliance.
Evidence: Bench inspection
3. Has PPE ever run out during a shift, and if so, was this logged as a safety event? — Stockouts that go unrecorded suggest supply monitoring is inadequate.
Evidence: Safety event log

Common reasons for a PARTIAL answer

  • PPE is available but located too far from actual point of use.
  • Compliance drops specifically during high-volume periods.

Implementation plan

When What
Week 1 Observe actual PPE use during a genuinely busy period.
Week 2 Relocate PPE supplies to each point of use if friction is found.
Week 3 Build a stockout logging mechanism.
Ongoing Conduct periodic unannounced compliance spot-checks.

How the Monitor verifies this

Method What Detail
OBSERVE Bench work observation Directly observes actual bench work, ideally during a busy period, for genuine PPE compliance.

Evidence base

World Health Organization. Laboratory biosafety manual, 4th ed. Geneva: WHO; 2020.
7.6

Exposure Incident: Documented Response Protocol

Non-Negotiable

A written protocol governs response to a biological exposure incident — needlestick, splash, spill — including immediate first aid, reporting, and post-exposure medical follow-up, known to staff and accessible at the point of need.

In plain terms: If someone has an actual exposure incident, there’s a real, known plan everyone can follow immediately — not a document to go find and read for the first time in that moment.

Facility category Standalone lab Hospital lab Clinic lab
Applicability Full Full Full

Why this matters

The first minutes after a biological exposure incident genuinely matter for post-exposure outcomes, and this is precisely the moment staff have the least capacity to calmly look up and read an unfamiliar procedure from scratch. A protocol that exists only as a document, without having been actively taught or drilled, functions very differently in theory than it does during the stress of a real event.

What good looks like

  • Staff can describe the first steps without needing to consult a document.
  • A recent actual incident’s response genuinely matched the written protocol.
  • Post-exposure medical follow-up is a real, responsive arrangement.

Common failure modes

  • Staff know a protocol exists but can’t describe its actual first steps.
  • The referenced occupational health service doesn’t respond promptly when called.
  • The protocol has never been drilled, only distributed as a document.

Worked example

In practice
A laboratory reviewing readiness after a near-miss sharps incident.
BeforeA written exposure protocol existed in the laboratory’s policy manual, but when asked, several staff members could not describe its first steps and had never actually practiced it, having only read it once during initial onboarding years earlier.
ActionThe laboratory conducted a tabletop drill walking through an actual exposure scenario step by step, confirmed the referenced occupational health service’s actual response time by testing it directly, and posted a simplified quick-reference card at each bench.
AfterThe Monitor asked two staff members to describe the first steps of the exposure protocol without reference material, and both answered correctly and confidently. Criterion verified.

If you are starting from zero — do this first

  1. Ask two or three staff members right now to describe the exposure protocol’s first steps.
  2. Test your referenced occupational health service’s actual response time.
  3. Consider a quick-reference card at each bench, not just a policy manual entry.
The most common mistake: Treating onboarding-stage reading of the exposure protocol as sufficient preparation, with no refresher or drill ever repeated afterward.

Self-assessment questions

1. Can staff describe the first steps of the exposure protocol without consulting a document? — In an actual exposure, staff will not have time to look up a procedure from scratch.
Evidence: Staff interview
2. Is there a documented recent exposure incident, and did the actual response match the written protocol? — Where no incident has occurred, confirm the protocol has at least been drilled or walked through.
Evidence: Incident record or drill record
3. Is post-exposure medical follow-up actually accessible, not just named in the policy with no real arrangement in place? — A referenced occupational health service that doesn’t actually respond when called is not a functioning arrangement.
Evidence: Occupational health service responsiveness check

Common reasons for a PARTIAL answer

  • Staff read the protocol once at onboarding and never again.
  • The occupational health referral exists on paper but hasn’t been tested for real responsiveness.

Implementation plan

When What
Week 1 Ask current staff to describe the protocol’s first steps.
Week 2 Conduct a tabletop drill and test occupational health responsiveness.
Week 3 Post a quick-reference card at each bench.
Ongoing Refresh training periodically, not only at onboarding.

How the Monitor verifies this

Method What Detail
ASK Staff knowledge check Asks staff to describe the protocol’s first steps without reference material.

Evidence base

World Health Organization. Laboratory biosafety manual, 4th ed. Geneva: WHO; 2020.
Centers for Disease Control and Prevention. Updated U.S. Public Health Service Guidelines for the Management of Occupational Exposures. MMWR Recomm Rep. 2013;62(RR-10):1-19.
7.7

Information System Access Controlled and Logged

Core

Access to the laboratory information system is controlled through individual, non-shared user credentials, with access levels matched to role, and a log of who accessed or modified patient data.

In plain terms: Every person has their own login, not a shared one — and what each person can see or change in the system actually matches what their role genuinely needs.

Facility category Standalone lab Hospital lab Clinic lab
Applicability Full Full Full

Why this matters

A shared generic login makes it fundamentally impossible to answer “who actually did this” when a question arises about a specific data access or change — the accountability that individual credentials provide isn’t a bureaucratic nicety, it’s the only mechanism that makes any later investigation of an error or concern possible at all. Over-broad access permissions compound this risk by giving more staff than necessary the ability to alter released results.

What good looks like

  • Every staff member has individual, non-shared credentials.
  • Access levels genuinely differ by role, restricting amendment permission appropriately.
  • The access log is actually reviewed periodically, not just generated.

Common failure modes

  • A shared generic login is used for convenience, especially on shared terminals.
  • Every staff member has full edit access regardless of actual role need.
  • The access log exists technically but nobody ever reviews it.

Worked example

In practice
A shared terminal used by multiple rotating phlebotomists.
BeforeA single generic “phlebotomy” login was used on the shared terminal for convenience, since rotating staff found individual logins slower during busy periods, making it impossible to determine which specific staff member performed any given accessioning action.
ActionThe laboratory implemented individual badge-swipe login specifically designed to be fast enough not to disrupt workflow, eliminating the convenience justification for shared credentials.
AfterThe Monitor reviewed system access logs and confirmed every action traced to an individual, identifiable staff member with no shared login in use. Criterion verified.

If you are starting from zero — do this first

  1. Check whether any shared or generic logins are currently in use.
  2. Review access levels against actual role requirements.
  3. Schedule a periodic access log review if none currently happens.
The most common mistake: A shared login persisting specifically on a busy, multi-user terminal because individual login felt too slow, without exploring a faster individual-login method like badge swipe.

Self-assessment questions

1. Are login credentials individual, or does staff share a single generic login? — Shared logins make it impossible to know who actually performed a given action.
Evidence: System credential audit
2. Do access levels differ by role, with results-amendment permission restricted to appropriate staff? — Every staff member having full edit access to released results is a specific, checkable risk.
Evidence: Access level configuration
3. Is the access log actually reviewed periodically, not just generated and left unexamined? — A log nobody reviews provides no real oversight.
Evidence: Log review record

Common reasons for a PARTIAL answer

  • Shared logins persist on busy, multi-user terminals.
  • Access logs exist but are never actually reviewed.

Implementation plan

When What
Week 1 Audit current login practice for any shared credentials.
Week 2 Implement individual credentials, including fast methods like badge swipe where needed.
Week 3 Review and correct access levels by role.
Ongoing Review access logs on a fixed schedule.

How the Monitor verifies this

Method What Detail
DOCUMENT Credential and access audit Checks for individual, non-shared credentials and reviews access level assignments.

Evidence base

International Organization for Standardization. ISO 15189:2022, Medical laboratories — Requirements for quality and competence, Clause 7.3.9. Geneva: ISO; 2022.
7.8

Result Amendment Traceable to Original

Core

Where a released result is amended, the original value, the amended value, the reason, the person making the change, and the date are all retained and visible together — the original is never simply overwritten or deleted.

In plain terms: If a result gets corrected after release, you can still see what it originally said, what it was changed to, why, and by whom — nothing just quietly disappears.

Facility category Standalone lab Hospital lab Clinic lab
Applicability Full Full Full

Why this matters

An overwritten original value removes the ability to understand what actually happened — whether the original was a transcription error, an instrument malfunction, or something more concerning, and whether a clinician may have already acted on the incorrect original value. Full traceability, including notification to the ordering clinician, is what allows both internal quality review and appropriate downstream clinical correction.

What good looks like

  • The original value remains visible alongside any amendment, never deleted.
  • Every amendment shows a documented reason and who made it.
  • The ordering clinician is notified when a previously released result is amended.

Common failure modes

  • The original value is overwritten with no trace remaining.
  • An amendment exists but with no documented reason attached.
  • The ordering clinician never learns the result they already saw was amended.

Worked example

In practice
A transcription error discovered after result release.
BeforeA manually entered result was found to contain a transcription error the day after release. The value was corrected directly in the system, overwriting the original, with no notification sent to the ordering physician who had already seen and potentially acted on the incorrect value.
ActionThe laboratory configured the system to preserve amended values as an addition rather than an overwrite, visible alongside the original with reason and amender name, and built a mandatory clinician notification step for any amendment to a previously released result.
AfterThe Monitor selected a recent amendment and found the original value still visible, a documented reason, and a recorded clinician notification. Criterion verified.

If you are starting from zero — do this first

  1. Check your system configuration — does amendment overwrite or preserve the original?
  2. Build a mandatory reason field for any amendment.
  3. Build a required clinician notification step for amendments to released results.
The most common mistake: Correcting an error directly by overwriting the original value, treating the correction itself as the priority while overlooking that the ordering clinician already saw and may have acted on the wrong value.

Self-assessment questions

1. Pick an amended result at random — is the original value still visible alongside the amendment, not replaced? — An overwritten original with no trace is a significant traceability failure.
Evidence: Amendment record
2. Does the amendment record show who made the change and why? — An amendment with no attributed reason undermines confidence in the correction itself.
Evidence: Amendment reason and attribution
3. Was the ordering clinician notified when a previously released result was amended? — An amendment the requesting clinician never learns about can lead to clinical decisions based on the wrong value.
Evidence: Clinician notification record

Common reasons for a PARTIAL answer

  • System configuration overwrites rather than preserves the original.
  • Clinician notification isn’t consistently completed for amendments.

Implementation plan

When What
Week 1 Check system configuration for amendment preservation versus overwrite.
Week 2 Reconfigure if the original is currently overwritten.
Week 3 Build a mandatory clinician notification step.
Ongoing Audit amendment records periodically for completeness.

How the Monitor verifies this

Method What Detail
DOCUMENT Amendment record review Selects an amended result and verifies original preservation, documented reason, and clinician notification.

Evidence base

International Organization for Standardization. ISO 15189:2022, Medical laboratories — Requirements for quality and competence, Clause 7.4.4. Geneva: ISO; 2022.
← Laboratory Standards — all standards
Standard 8 — Post-Examination: Reporting & Critical Values →
© 2026 Accréditation Sans Frontières · PHIG · Sheni Network