ASF Standards · Laboratory · Standard 2

Standard 2 — Management Requirements

8 criteria · 4 non-negotiable · 4 core · Version 1.0

Criteria in this standard

2.1 — Internal Audit, On a Fixed Schedule
2.2 — Corrective Action Closes the Root Cause
2.3 — Preventive Action, Not Only Reactive
2.4 — Management Review, With Minutes and Actions
2.5 — Complaints Logged, Investigated, Closed
2.6 — Impartiality: No Commercial Pressure on Results
2.7 — Conflict of Interest Declared and Reviewed
2.8 — External Service and Supply Evaluation

2.1

Internal Audit, On a Fixed Schedule

Non-Negotiable

The full quality management system is internally audited at least once per year against a documented audit schedule, covering every section of the laboratory over a defined cycle, not only the sections most convenient to review.

In plain terms: Every part of the lab gets checked by someone looking specifically for problems, on a real schedule — not just the easy, visible sections.

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

Why this matters

Internal audit is the laboratory checking itself honestly before an external body does. Audits that only ever cover the same convenient sections — the main chemistry bench, say — while never reaching the satellite point-of-care site or the after-hours coverage arrangement, create a false sense of security. The gaps nobody audits are exactly where problems accumulate unnoticed.

What good looks like

A schedule covering every section across a defined cycle, not just favorites.
Audits actually happen on or near their scheduled date.
Findings are detailed enough for someone outside the audit to understand them.

Common failure modes

The same easy sections get audited repeatedly; harder-to-reach ones never do.
Audits slip months behind schedule with no one tracking the slippage.
Findings are a checklist of ticks with no real detail behind them.

Worked example

In practice

A laboratory with a main site and a small satellite point-of-care testing location.

BeforeThree consecutive annual audits had covered the main chemistry and hematology benches in detail. The satellite point-of-care site, five minutes away, had never once been included — it simply wasn’t on anyone’s list.

ActionThe quality manager built a formal three-year audit cycle explicitly naming every physical location and section, including the satellite site, and put the schedule on a shared calendar visible to the whole team.

AfterThe Monitor reviewed the three-year schedule and confirmed the satellite site’s first audit had been completed on time, with detailed findings on file. Criterion verified.

If you are starting from zero — do this first

List every physical location and section the laboratory operates.
Check which of these your current or most recent audit actually covered.
Build a multi-year cycle that explicitly includes everything, not just the obvious parts.
Put real dates on a calendar, not a vague annual intention.

The most common mistake: A satellite or after-hours service location quietly falls outside the audit scope because nobody explicitly included it when the schedule was first built.

Self-assessment questions

1. Does a documented audit schedule exist covering every laboratory section within a defined cycle? — Not just the sections an auditor finds easiest to reach.
Evidence: Audit schedule

2. Did the most recent audit actually happen on or near its scheduled date? — A schedule that is routinely missed is not functioning.
Evidence: Audit completion records

3. Are audit findings recorded in enough detail that someone outside the audit could understand what was actually checked? — A checklist with ticks and no detail is not a usable audit record.
Evidence: Audit report

Common reasons for a PARTIAL answer

A remote or satellite location was never explicitly added to scope.
Audits happen but findings are too brief to act on.

Implementation plan

When	What
Week 1	List every section and location requiring audit coverage.
Week 2	Build a multi-year cycle with explicit scheduled dates.
Week 3	Share the schedule and assign an auditor to the first session.
Ongoing	Track schedule adherence and escalate any slippage.

How the Monitor verifies this

Method	What	Detail
DOCUMENT	Schedule and report review	Checks the audit schedule’s completeness against actual facility sections, and reviews recent audit report detail.

Evidence base

International Organization for Standardization. ISO 15189:2022, Medical laboratories — Requirements for quality and competence, Clause 8.8. Geneva: ISO; 2022.

International Organization for Standardization. ISO 19011:2018, Guidelines for auditing management systems. Geneva: ISO; 2018.

2.2

Corrective Action Closes the Root Cause

Non-Negotiable

Every nonconformance with potential impact on patient results triggers a documented corrective action that identifies the root cause, not only the immediate symptom, and the action is verified as effective after implementation.

In plain terms: When something goes wrong, the lab fixes the real reason it happened — not just the thing that was visible that day — and checks afterward that the fix actually worked.

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

Why this matters

“Staff error” is the single most common entry on a corrective action form, and it is almost always incomplete as a root cause. Staff rarely make errors for no reason — a confusing label, an unclear SOP step, inadequate training, chronic understaffing at a particular shift. Stopping at “staff error, retrained” without asking what allowed the error to happen leaves the actual cause fully intact, waiting to produce the same failure again.

What good looks like

Root cause identified beyond the immediate symptom.
Effectiveness checked after a defined interval, not assumed.
Recurrence of the same nonconformance type triggers fresh investigation.

Common failure modes

“Staff error, retrained” as the entire investigation.
No follow-up check that the action actually worked.
The same failure recurs months later with no connection drawn to the earlier one.

Worked example

In practice

A laboratory investigating a repeated mislabeling incident.

BeforeA specimen was mislabeled for the second time in three months. Both incidents were closed with “staff retrained on labeling procedure” and no further investigation.

ActionAfter the second incident, the quality manager investigated further and found both involved the same busy collection window when two phlebotomists worked side by side with visually similar pre-printed labels for different patients. The corrective action changed the label stock design and physically separated the two workstations during that window.

AfterThe Monitor reviewed the corrective action record and found a clearly identified root cause distinct from “staff error,” plus a documented effectiveness check three months later showing no recurrence. Criterion verified.

If you are starting from zero — do this first

Pull your last five corrective actions and check how many say only “staff error.”
For each, ask what allowed the error to happen — that’s the real root cause.
Build in a scheduled follow-up check for every corrective action going forward.

The most common mistake: Treating retraining as both the root cause finding and the complete fix, when retraining addresses only one possible contributing factor.

Self-assessment questions

1. Pick a recent corrective action at random — does it name a root cause, or only restate the symptom? — “Staff error” is a symptom; what allowed the error to happen is the root cause.
Evidence: Corrective action record

2. Was the corrective action’s effectiveness checked after a defined interval, not just implemented and assumed to have worked? — A documented follow-up check, not an assumption.
Evidence: Effectiveness check record

3. Where the same type of nonconformance has recurred, was that recurrence itself investigated as a sign the original action failed? — A repeat failure without a fresh investigation suggests the corrective action process is not functioning.
Evidence: Nonconformance log showing trend review

Common reasons for a PARTIAL answer

Root cause analysis stops at the first obvious answer.
Effectiveness checks are skipped once the immediate fix is in place.

Implementation plan

When	What
Week 1	Review recent corrective actions for root-cause depth.
Week 2	Train staff involved in investigations on basic root-cause techniques.
Week 3	Add a mandatory effectiveness-check field to the corrective action form.
Ongoing	Trend nonconformance types to catch recurrence early.

How the Monitor verifies this

Method	What	Detail
DOCUMENT	Corrective action record review	Selects recent corrective actions and assesses root-cause depth and effectiveness follow-up.

Evidence base

International Organization for Standardization. ISO 15189:2022, Medical laboratories — Requirements for quality and competence, Clause 8.7. Geneva: ISO; 2022.

2.3

Preventive Action, Not Only Reactive

Core

The laboratory identifies potential nonconformances before they occur — from trend data, near-misses, risk register entries, or external quality assessment patterns — and acts on them, not only responding after an actual failure.

In plain terms: The lab doesn’t just fix things after they break — it actually looks for early warning signs and acts on those too.

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

Why this matters

A laboratory that only ever acts after an actual failure is, by definition, always one step behind. The data that could have predicted the failure — a slow drift in quality control values, a near-miss that didn’t quite become an incident, a pattern across external quality assessment rounds — is usually sitting right there, unused, because nobody is specifically assigned to look for it before something breaks.

What good looks like

At least one real example of action taken from a trend, not an actual failure.
External quality assessment performance trended across rounds, not just pass/fail.
A named person responsible for trend review on a schedule.

Common failure modes

Every documented action traces back to an actual incident.
Quality data is collected but nobody is assigned to review trends.
A slow drift goes unnoticed until it crosses an acceptable limit.

Worked example

In practice

A chemistry section tracking internal quality control trends.

BeforeQuality control values for a glucose analyzer had been drifting slowly toward the upper control limit over six weeks, still technically passing each individual run, but nobody had plotted the trend across runs to notice the pattern.

ActionA newly assigned quality reviewer, checking Levey-Jennings charts weekly rather than only per-run pass/fail, caught the drift and scheduled preventive maintenance before the analyzer actually failed control.

AfterThe Monitor reviewed the quality reviewer’s trend log and found the documented preventive action taken before any actual control failure occurred. Criterion verified.

If you are starting from zero — do this first

Name a specific person responsible for trend review, not just incident response.
Set a weekly or monthly schedule for reviewing quality control trends, not just pass/fail.
Look back at external quality assessment results for any multi-round pattern.

The most common mistake: Checking quality control only as pass/fail per run, which can hide a genuine trend developing across multiple passing runs.

Self-assessment questions

1. Can the laboratory show at least one preventive action taken from trend data or a near-miss, not triggered by an actual failure? — If every action on file follows an actual incident, prevention is not genuinely functioning.
Evidence: Preventive action record

2. Is external quality assessment performance trended over time, not just checked pass/fail each round? — A single-round pass can mask a worsening trend across several rounds.
Evidence: EQA trend analysis

3. Is there a named person responsible for reviewing trends on a schedule? — Trend data nobody is assigned to review does not drive prevention.
Evidence: Role assignment, review schedule

Common reasons for a PARTIAL answer

Data is collected but review responsibility isn’t clearly assigned.
Trend charts exist but aren’t actually looked at regularly.

Implementation plan

When	What
Week 1	Assign a named person to trend review responsibility.
Week 2	Set up Levey-Jennings or equivalent trend charting if not already in use.
Week 3	Review existing EQA history for any missed multi-round patterns.
Ongoing	Hold a scheduled trend review, separate from routine pass/fail checks.

How the Monitor verifies this

Method	What	Detail
DOCUMENT	Trend record review	Looks for at least one preventive action genuinely triggered by trend data rather than an actual failure.

Evidence base

International Organization for Standardization. ISO 15189:2022, Medical laboratories — Requirements for quality and competence, Clause 8.5. Geneva: ISO; 2022.

Westgard JO. Basic QC Practices, 4th ed. Madison (WI): Westgard QC; 2016.

2.4

Management Review, With Minutes and Actions

Core

Senior management formally reviews the quality management system at least annually, with documented minutes, inputs covering audits/complaints/nonconformances/external quality assessment, and tracked output actions with owners and dates.

In plain terms: Leadership actually sits down once a year, looks at the real quality data, writes down what they decided, and someone checks those decisions actually happen.

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

Why this matters

Management review is where quality data is supposed to actually reach the people with authority to allocate resources and change direction. A review that happens informally in a hallway conversation, with nothing written down and no required inputs reviewed systematically, provides none of the accountability or traceability the formal process is meant to create.

What good looks like

Documented minutes with named attendees exist for the most recent cycle.
Required inputs — audits, complaints, nonconformances, EQA — are all covered.
Output actions from the previous review are traceable to completion.

Common failure modes

Review happens but only covers budget, not quality inputs.
No written minutes, just a verbal assurance “we discussed it.”
Agreed actions from last year’s review were never followed up.

Worked example

In practice

A laboratory’s first formal management review cycle.

BeforeAnnual “management review” had consisted of the Director briefly mentioning quality metrics during a general staff meeting, with no minutes, no structured agenda, and no tracked actions.

ActionThe Director instituted a formal annual review meeting with a fixed agenda covering all required inputs, took written minutes with named attendees, and created an action tracker with owners and target dates for every output decision.

AfterThe Monitor reviewed the minutes from the most recent cycle and the action tracker, confirming three of four prior actions were completed on time and the fourth had a documented, reasonable extension. Criterion verified.

If you are starting from zero — do this first

Set a fixed annual date for the formal management review.
Build an agenda explicitly covering audits, complaints, nonconformances, and EQA.
Take real minutes, name attendees, and assign owners to every output action.

The most common mistake: Holding a review that covers operational and budget topics thoroughly while quality-specific inputs get only a passing mention.

Self-assessment questions

1. Do management review minutes exist for the most recent cycle, with named attendees? — Not a verbal assurance that “we discussed it.”
Evidence: Signed meeting minutes

2. Do the minutes cover the required inputs — audit results, complaints, nonconformances, external quality assessment performance? — A review that only covers budget is not a quality management review.
Evidence: Meeting agenda and minutes

3. Are output actions from the previous review traceable to completion, with a named owner for each? — Actions agreed and never followed up are a common, specific failure.
Evidence: Action tracker

Common reasons for a PARTIAL answer

Minutes exist but omit required quality inputs.
Actions are agreed verbally but never tracked to closure.

Implementation plan

When	What
Week 1	Set the annual review date and build a structured agenda template.
Week 2	Gather required inputs — audit, complaint, nonconformance, EQA data.
Week 3	Hold the review, take formal minutes, assign action owners and dates.
Ongoing	Track output actions to completion before the next cycle.

How the Monitor verifies this

Method	What	Detail
DOCUMENT	Minutes and action tracker review	Confirms minutes cover all required inputs and checks prior action completion status.

Evidence base

International Organization for Standardization. ISO 15189:2022, Medical laboratories — Requirements for quality and competence, Clause 8.9. Geneva: ISO; 2022.

2.5

Complaints Logged, Investigated, Closed

Core

Every complaint, from any source — patient, clinician, referring facility — is logged in a single register, investigated, and closed with a documented outcome communicated back to the complainant where contact information exists.

In plain terms: Every complaint gets written down in one place, actually looked into, and the person who complained finds out what happened.

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

Why this matters

A complaint handled verbally and resolved on the spot, with nothing logged, is invisible to the quality system — it cannot be trended, cannot inform prevention, and leaves no record that the concern was ever taken seriously. Complaints are often the earliest, most direct signal a laboratory receives that something in its process isn’t working as intended.

What good looks like

A single complaints register captures every complaint, regardless of source.
Each entry has a documented investigation outcome.
The complainant is informed of the outcome where contact exists.

Common failure modes

Complaints handled verbally by whoever answers the phone, never logged.
A complaint is received but the outcome is never communicated back.
Multiple informal logs exist across different staff, none complete.

Worked example

In practice

A laboratory receiving a complaint from a referring clinic about delayed results.

BeforeA referring clinic called to complain about repeatedly delayed turnaround on referred tests. The front-desk staff member apologized, promised to “look into it,” and never logged the call anywhere.

ActionThe laboratory introduced a mandatory complaint log at every point of contact, investigated the delay — tracing it to a courier scheduling gap — fixed the underlying issue, and called the clinic back with the specific finding and the fix.

AfterThe Monitor selected the logged complaint and found a documented investigation, a specific root cause, and a dated record of the callback informing the clinic of the outcome. Criterion verified.

If you are starting from zero — do this first

Create one single, accessible complaints log, not department-specific lists.
Tell all staff: every complaint gets logged, no exceptions, even if resolved immediately.
Build in a step to close the loop with the complainant.

The most common mistake: Treating a complaint that was resolved quickly and politely as not needing to be logged, when logging matters regardless of how smoothly it was handled in the moment.

Self-assessment questions

1. Does a single complaints register exist, or are complaints handled informally by whoever receives them? — Complaints handled verbally and never logged are effectively invisible to the quality system.
Evidence: Complaints register

2. Is there a documented outcome for each logged complaint, not just a record that it was received? — Receipt without resolution is not complaint handling.
Evidence: Investigation outcome record

3. Where contact information exists, was the complainant actually informed of the outcome? — Closing a complaint internally without telling the person who raised it undermines trust.
Evidence: Callback or communication record

Common reasons for a PARTIAL answer

Verbal complaints resolved on the spot are never logged.
The register exists but outcomes are rarely communicated back.

Implementation plan

When	What
Week 1	Build or consolidate a single complaints register.
Week 2	Brief all staff that every complaint, however resolved, gets logged.
Week 3	Add a required closing-the-loop step to the process.
Ongoing	Trend complaint types quarterly for recurring themes.

How the Monitor verifies this

Method	What	Detail
DOCUMENT	Register review	Selects a logged complaint and verifies investigation outcome and complainant communication.

Evidence base

International Organization for Standardization. ISO 15189:2022, Medical laboratories — Requirements for quality and competence, Clause 7.7 (Complaints). Geneva: ISO; 2022.

2.6

Impartiality: No Commercial Pressure on Results

Non-Negotiable

No staff member’s remuneration, performance evaluation, or continued employment is tied, directly or indirectly, to the volume or outcome of specific test results. Laboratory personnel are protected from undue commercial or financial pressure that could influence technical judgment.

In plain terms: Nobody’s paycheck depends on what a test result says, and staff genuinely feel safe flagging a problem even if it slows things down.

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

Why this matters

Impartiality is foundational to laboratory credibility — a result is only trustworthy if nothing in how staff are compensated or evaluated creates an incentive, conscious or not, to produce a particular outcome. Volume-linked bonuses are the most direct version of this risk, but subtler pressures — an unspoken expectation that flagging problems reflects badly on performance reviews — can be just as corrosive and are harder to detect from policy documents alone.

What good looks like

No compensation link to test volume, patterns, or referral relationships.
A written, genuine non-retaliation policy for raising quality concerns.
Staff report feeling safe raising concerns, even when confidentially asked.

Common failure modes

A volume-based bonus structure for technical staff.
An unwritten culture where flagging delays is discouraged informally.
Staff privately admit hesitation to raise concerns despite a written policy.

Worked example

In practice

A laboratory reviewing its compensation structure during a quality assessment.

BeforeTechnical staff received a small monthly bonus tied partly to total specimens processed, originally intended to reward productivity, but which created a subtle incentive to rush rather than pause for a questionable result.

ActionManagement restructured the bonus to be based entirely on quality metrics — EQA performance, audit findings, documentation completeness — removing any volume component, and communicated the change and its reasoning directly to staff.

AfterThe Monitor reviewed the compensation structure and confirmed no volume or result-outcome link existed, and confidential staff conversations confirmed genuine comfort raising concerns. Criterion verified.

If you are starting from zero — do this first

Review every compensation and bonus structure for any volume or outcome link.
Write a clear non-retaliation policy if none exists.
Ask staff confidentially, separately from management, whether they’d feel safe raising a concern.

The most common mistake: A productivity bonus introduced with good intentions that inadvertently creates volume pressure nobody explicitly designed for.

Self-assessment questions

1. Is any part of technical staff compensation linked to test volume, result patterns, or referral relationships? — A volume-linked bonus for technical staff creates exactly the pressure this criterion prohibits.
Evidence: Compensation policy

2. Is there a documented policy explicitly protecting staff who report a quality concern from retaliation? — Protection that exists informally but isn’t written down fails under pressure.
Evidence: Non-retaliation policy

3. Would a technician who identified a result requiring re-testing feel safe raising it, even if it delays turnaround? — Tested through confidential staff conversation, not management’s own assurance.
Evidence: Confidential staff interview

Common reasons for a PARTIAL answer

A legacy bonus structure was never reviewed for this specific risk.
A written policy exists but staff don’t genuinely feel it applies in practice.

Implementation plan

When	What
Week 1	Audit all compensation structures for volume or outcome links.
Week 2	Write or strengthen the non-retaliation policy.
Week 3	Communicate the policy directly and explain any compensation changes.
Ongoing	Periodically check staff comfort confidentially, not just via policy review.

How the Monitor verifies this

Method	What	Detail
DOCUMENT	Compensation structure review	Examines compensation policy for any volume or outcome link.
ASK	Confidential staff interview	Speaks with staff privately, away from management, about comfort raising concerns.

Evidence base

International Organization for Standardization. ISO 15189:2022, Medical laboratories — Requirements for quality and competence, Clause 4.1. Geneva: ISO; 2022.

International Organization for Standardization. ISO/IEC 17025:2017, General requirements for the competence of testing and calibration laboratories, Clause 4.1. Geneva: ISO; 2017.

2.7

Conflict of Interest Declared and Reviewed

Core

Staff and management with influence over testing, purchasing, or reporting decisions declare financial or personal interests that could reasonably affect impartiality, and these declarations are reviewed on a fixed schedule, not only collected once at hiring.

In plain terms: People who make decisions affecting testing or purchasing say upfront if they have a personal financial stake — and this gets checked again periodically, not just once when they were hired.

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

Why this matters

Circumstances change after hiring — a staff member’s spouse takes a sales role at a reagent supplier, a manager acquires a financial stake in a referral laboratory. A conflict-of-interest declaration collected once, years ago, and never refreshed captures none of this. The purpose of periodic review is specifically to catch exactly these changes before they quietly influence a purchasing or referral decision.

What good looks like

Current declarations exist for everyone in a position of influence.
A named person or body actually reviews declarations and decides on action.
Identified conflicts have a documented mitigation on record.

Common failure modes

Declarations collected once at hiring and never refreshed.
Declarations filed but nobody formally reviews them.
A genuine conflict is identified but no mitigation is documented.

Worked example

In practice

A laboratory’s annual conflict-of-interest review cycle.

BeforeDeclarations had been collected once, at hiring, for all purchasing-influential staff. No refresh process existed. A purchasing manager’s recent family connection to a reagent distributor had never been captured.

ActionThe laboratory introduced an annual declaration refresh, reviewed by the quality committee. The purchasing manager’s new family connection was disclosed in the next cycle, and the committee documented a mitigation — a second person co-signing all orders from that specific distributor.

AfterThe Monitor reviewed the annual refresh process and the specific mitigation record, confirming the conflict had been identified and genuinely addressed. Criterion verified.

If you are starting from zero — do this first

Identify everyone with real influence over testing, purchasing, or reporting decisions.
Collect current declarations if none exist or they’re outdated.
Set an annual refresh schedule and name who reviews them.

The most common mistake: Treating the hiring-stage declaration as a one-time, permanent record rather than a starting point that needs refreshing as circumstances change.

Self-assessment questions

1. Do current declarations exist for everyone in a position of influence, not just a one-time declaration from years ago? — Circumstances change; a declaration never refreshed becomes stale.
Evidence: Current declaration records

2. Is there a named person or body responsible for reviewing declarations and deciding what action, if any, is needed? — A declaration filed and never reviewed by anyone accomplishes nothing.
Evidence: Review committee record

3. Where a genuine conflict was identified, is there a record of what mitigation was actually applied? — Identification without documented mitigation is incomplete.
Evidence: Mitigation record

Common reasons for a PARTIAL answer

Declarations exist but have never been refreshed since hiring.
A conflict was noted but no concrete mitigation was documented.

Implementation plan

When	What
Week 1	Identify all staff in positions of influence over testing or purchasing.
Week 2	Collect or refresh declarations for everyone identified.
Week 3	Assign review responsibility and set an annual refresh schedule.
Ongoing	Document mitigation whenever a genuine conflict is identified.

How the Monitor verifies this

Method	What	Detail
DOCUMENT	Declaration and mitigation review	Checks declaration currency and any documented mitigation for identified conflicts.

Evidence base

International Organization for Standardization. ISO 15189:2022, Medical laboratories — Requirements for quality and competence, Clause 4.1.4. Geneva: ISO; 2022.

2.8

External Service and Supply Evaluation

Core

Reference laboratories, equipment service providers, and critical reagent suppliers are formally evaluated before use and periodically thereafter, against documented criteria relevant to the service each provides.

In plain terms: The lab doesn’t just start working with a supplier or reference lab informally — it checks them properly first, and keeps checking over time.

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

Why this matters

A laboratory’s result quality depends partly on parties it doesn’t directly control — the reference laboratory running referred tests, the technician servicing an analyzer, the supplier shipping reagents. Relationships with these parties often begin informally, through a personal recommendation or convenience, and formal evaluation is treated as paperwork to complete later. Later rarely comes, and the laboratory is left relying on providers it has never actually assessed.

What good looks like

A documented evaluation exists before the relationship begins in earnest.
Evaluation is repeated periodically, not only at the start.
Unsatisfactory performance has a documented laboratory response.

Common failure modes

A long-standing relationship was never formally evaluated at all.
Initial evaluation happened but was never repeated.
Known performance issues continue with no documented response.

Worked example

In practice

A laboratory formalizing its reference laboratory relationship.

BeforeThe laboratory had referred specialized tests to the same reference laboratory for six years, based on an original personal connection between directors. No formal evaluation had ever been documented.

ActionThe quality manager built a formal evaluation covering turnaround time, accreditation status, and result accuracy on split samples, completed it retrospectively, and scheduled annual re-evaluation going forward.

AfterThe Monitor reviewed the evaluation record and the scheduled annual re-evaluation and confirmed the relationship, though long-standing, now had a genuine documented basis. Criterion verified.

If you are starting from zero — do this first

List every reference laboratory, equipment servicer, and critical reagent supplier currently used.
Check which of these have a documented evaluation on file.
Build evaluations for any gaps, even for long-standing relationships.

The most common mistake: Assuming a long-standing, trusted relationship doesn’t need formal evaluation because “we know them well” — familiarity is not the same as documented assessment.

Self-assessment questions

1. Is there a documented evaluation on file for the laboratory’s current reference laboratory and key equipment service providers? — Not a relationship that began informally and was never formally assessed.
Evidence: Supplier evaluation record

2. Are these evaluations repeated on a periodic schedule, not only at the start of the relationship? — A provider’s performance can decline over years without a fresh check.
Evidence: Re-evaluation schedule and records

3. Where a provider’s performance has been unsatisfactory, is there a record of how the laboratory responded? — Continuing with an underperforming provider with no documented response is a gap.
Evidence: Performance response record

Common reasons for a PARTIAL answer

Long-standing suppliers were grandfathered in without ever being formally evaluated.
Evaluation happened once years ago and was never repeated.

Implementation plan

When	What
Week 1	List all critical external service providers and suppliers.
Week 2	Build or complete formal evaluations for each, including long-standing ones.
Week 3	Set a periodic re-evaluation schedule.
Ongoing	Document any response to identified performance issues.

How the Monitor verifies this

Method	What	Detail
DOCUMENT	Evaluation record review	Checks for documented initial and periodic evaluation of key external providers.

Evidence base

International Organization for Standardization. ISO 15189:2022, Medical laboratories — Requirements for quality and competence, Clause 6.6. Geneva: ISO; 2022.

← Laboratory Standards — all standards
Standard 3 — Personnel & Competence →