ICH Q10: Pharmaceutical Quality System
A practical guide to the pharmaceutical quality system model and the responsibilities, evidence and decisions that connect product knowledge to continual improvement.

On this page · 12 sections
- 1What ICH Q10 is—and what it does not replace
- 2Lifecycle application: development through product discontinuation
- 3Management responsibility: ownership, resources and escalation
- 4PQS element: monitoring process performance and product quality
- 5PQS element: CAPA connected to investigation evidence
- 6PQS elements: change management and product-performance review
- 7PQS improvement and the knowledge and risk enablers
- 8Jurisdiction and product scope: Q10 alongside regional GMP
- 9Evidence model, system interfaces and exception recovery
- 10Worked cases: signals, transfer gaps and duplicate evidence
- 11Implementation checklist: test the connections, not just the procedures
- 12V5 Ultimate support: assembled evidence for human review
How does ICH Q10 apply to your shop floor?
Pick your industry and scale — Ask V5 rewrites the definition in your context, gives a worked example, and shows what V5 does on day one.
01What ICH Q10 is—and what it does not replace
ICH Q10 is a harmonised guideline describing a pharmaceutical quality system model for drug substances and drug products throughout their lifecycle, intended for use alongside regional GMP requirements. That relationship is set out in ICH Q10 Section 1.3. The cross-regional pharmaceutical model covers systems supporting development and manufacture of active pharmaceutical ingredients and drug products, including biotechnology and biological products, under Section 1.2.
For pharmaceutical APIs and drug products in this cross-regional model, ICH Q10 Section 1.1 identifies content additional to current regional GMP requirements as optional. That qualification concerns the additional guidance—not the applicability of regional GMP requirements.
Operationally, it helps to separate the reference model from the site arrangements and records used to implement it. The following are practical distinctions, not additional legal definitions.
| Term | Practical meaning | Boundary |
|---|---|---|
| ICH Q10 | Guideline model for a pharmaceutical quality system | Used alongside regional GMP; not a replacement regulatory code |
| Site PQS | Implemented responsibilities and quality processes | The working organisation, not merely a quality manual |
| Regional GMP requirements | Jurisdiction-specific regulatory baseline | Applicability depends on the market, product and activity |
| Process validation | Process-focused evidence and evaluation | An assurance activity, not the entire PQS |
| Electronic batch record | Organised execution evidence | A supporting record, not the quality system or an approval decision |
A useful site-level test is whether people can explain how a quality concern reaches an accountable decision-maker, which evidence informs the decision, and where the rationale is retained. A procedure library alone does not answer those questions. Neither does a software installation: records can support review, but human reviewers remain responsible for approval and release decisions. Treat Q10 as the organising model, the site PQS as its operational implementation, and supporting systems as tools within that implementation.
02Lifecycle application: development through product discontinuation
For pharmaceutical APIs and drug products in the cross-regional ICH model, ICH Q10 Annex 2 identifies pharmaceutical development, technology transfer, commercial manufacturing and product discontinuation as the product lifecycle. For that same pharmaceutical scope, Section 1.2 recommends applying Q10 elements appropriately and proportionately to each stage, recognising their different goals.
Operationally, teams commonly translate this into changing decisions and explicit evidence handovers rather than applying an identical checklist everywhere. The following is an illustrative operating model, not a prescribed organisation chart.
| Stage | Main decision | Evidence carried forward | Receiving owner |
|---|---|---|---|
| Pharmaceutical development | Capture product and process assumptions | Design rationale, study findings, unresolved uncertainties | Transfer lead and receiving technical team |
| Technology transfer | Resolve differences and knowledge gaps; accept the handover or return it for further work | Site comparisons, gap assessments, acceptance rationale | Receiving manufacturing, laboratory and quality leads |
| Commercial manufacturing | Connect execution with performance assessment | Batch history, trend interpretation, updated process understanding | Process owner and product-quality review team |
| Product discontinuation | Assign continuing record and assessment ownership | Record and sample inventories, open assessments, reporting responsibilities | Named records custodian and continuing quality owner |
At technology transfer, a useful acceptance record states what the receiving team understands, which differences remain unresolved, who owns them and why the proposed next activity is justified. Sending documents alone does not demonstrate that the receiving team can use the knowledge. Where acceptance remains conditional, record the conditions and the decision owner rather than leaving an ambiguous handover.
During commercial operation, teams commonly connect inherited assumptions to a continued process verification plan, making it possible to revisit them against manufacturing experience. This keeps development reasoning available to the people interpreting routine performance.
For pharmaceutical APIs and drug products in the cross-regional model, ICH Q10 Section 1.2 includes documentation retention, sample retention, and continued product assessment and reporting within discontinuation activities. An operational handover therefore commonly assigns continuing ownership rather than treating the end of manufacturing as closure of every responsibility.
03Management responsibility: ownership, resources and escalation
For pharmaceutical APIs and drug products in the cross-regional model, ICH Q10 Section 2.1(a) places ultimate responsibility for an effective pharmaceutical quality system with senior management. This guidance also expects roles, responsibilities and authorities to be defined, communicated and implemented throughout the company.
Turn accountability into a responsibility map
A practical site governance map separates leadership accountability from individual quality decisions. It can connect the quality assurance process to named owners without making senior management the approver of every investigation:
- Translate the site's quality policy into objectives with observable evidence of progress.
- Name each process owner and the functions needed to support that process.
- Distinguish who performs a task from who evaluates its evidence and has decision authority.
- Connect unresolved risks to resource decisions, recording the decision owner, rationale and follow-up.
For pharmaceutical APIs and drug products in the cross-regional model, ICH Q10 Section 2.5(a) recommends that management establish and implement appropriate internal communication processes.
For example, assume investigations repeatedly stall because the site lacks technical capacity to interpret process data. A practical escalation presents the recurring evidence gap, associated uncertainty and options for specialist support or training. Management addresses capacity and priorities; designated quality decision-makers retain their assigned review and approval authority. A staffing decision is not evidence that an investigation's conclusion is sound.
Make external interfaces explicit
For contract-manufacturer interfaces, use the quality agreement to ask: who communicates quality signals, provides investigation evidence, evaluates proposed changes and accepts transferred knowledge? These are agreement-design questions, not a universal allocation of legal responsibility. Practical arrangements also identify the escalation contact when evidence or a decision remains outstanding.
04PQS element: monitoring process performance and product quality
For pharmaceutical APIs and drug products, the cross-regional ICH Q10 guidance identifies a process performance and product quality monitoring system as a PQS element in ICH Q10 Section 3.2.
In site practice, an interpretable monitoring specification connects each measurement to its manufacturing context and a review decision. The following is an illustrative design, not a prescribed Q10 template.
| Field | What to define |
|---|---|
| Product and process scope | Product, site, process stage and applicable process version. |
| Characteristic and unit | Named measurement, unit and connection to a critical quality attribute or critical process parameter, where relevant. |
| Sampling context | Location, timing, method and operating conditions. |
| Source record | Traceable batch, sample, test or equipment record. |
| Grouping logic | Which observations can be compared, and why. |
| Signal interpretation | What prompts review and what uncertainty remains. |
| Review owner and escalation route | Responsible reviewer and route for unresolved or potentially significant findings. |
Results from different process versions or operating conditions may need separate interpretation: pooling them can conceal a shift or suggest deterioration where only the population changed. Production context tagging helps preserve those distinctions. Any combined analysis benefits from a documented comparability rationale.
Specification assessment asks whether an individual result meets its acceptance criteria; trend interpretation asks what comparable observations suggest over time. Accepted results can contribute to a concerning pattern, but a pattern alone does not establish a cause. Before interpreting it, check for missing observations, inconsistent units and changed sampling practices that could distort the apparent direction or variability.
The practical output is a documented quality signal: the observation, evidence links, relevant context, limitations and reviewer decision, with a rationale for follow-up or continued observation. The worked monitoring case below illustrates this reasoning; the later jurisdiction section addresses the FDA process-validation relationship.
05PQS element: CAPA connected to investigation evidence
For pharmaceutical APIs and drug products in the cross-regional model, ICH Q10 Section 3.2 identifies the corrective action and preventive action (CAPA) system as a PQS element. This is guidance, not a standalone legal obligation.
In practical quality-management usage, correction addresses the immediate problem; corrective action addresses an identified cause; preventive action reduces the likelihood of a potential problem. These are operational distinctions, not quotations from a formal standard. Correcting a record, for example, does not explain why the error occurred or prevent another occurrence.
Build a traceable investigation chain
- Identify the source event, assess its possible impact and record containment separately from the longer-term action.
- Compare competing causal explanations through root cause analysis. Connect each explanation to supporting or contradictory evidence rather than selecting the easiest explanation to close.
- Select an action that addresses the supported cause, assign an accountable owner and define an effectiveness criterion before evaluating the result.
Action completion asks whether the agreed work was performed. An effectiveness check asks whether subsequent evidence supports the intended result under relevant operating conditions. A revised instruction can be complete while the underlying failure continues. Define what evidence would demonstrate improvement, where it will come from and what finding would trigger reassessment.
If an action is ineffective, reassess the causal explanation, action design or implementation. When reopening the issue, preserve the original rationale, evidence and decisions; add the new findings rather than overwriting the earlier conclusion.
Electronic-record FAQ: do signatures prove effectiveness?
No. Recorded signatures and completed tasks document actions or decisions, not proof that an action worked. Attributable recording supports traceability; reviewers still judge whether the evidence satisfies the effectiveness criterion.
06PQS elements: change management and product-performance review
For pharmaceutical APIs and drug products in its cross-regional guidance model, ICH Q10 Section 3.2 identifies change management and management review of process performance and product quality as PQS elements. Operationally, these connect proposed improvements with evidence about their effects.
Assess the change before implementing it
An illustrative change-control assessment records the intended benefit, affected product and process versions, quality risks, supporting evaluation, document and training impacts, implementation conditions and post-change checks. Defining the checks before implementation makes the eventual effectiveness decision less vulnerable to hindsight.
For example, a proposed equipment-setting change could identify the applicable recipe version, the evidence supporting the setting and the observations that would trigger reassessment. Market-specific regulatory assessment remains a separate applicability question for the relevant product and jurisdiction—not an assumption that every change follows a universal submission rule.
Turn review findings into owned decisions
A practical periodic product-performance review combines monitoring signals, investigations, actions and change outcomes. Its output is a decision with an owner and follow-up evidence: continue observation, investigate further, revise an action or propose another change. Keeping pre-change and post-change results tied to their process versions helps reviewers interpret whether the intervention achieved its intended benefit.
This aggregate assessment is distinct from disposition of an individual batch. Questions about the effectiveness of the PQS itself belong in the following section.
Illustrative site workflow
Monitoring signal → investigation and assessment → proposed action or change → human authorisation → implementation → effectiveness check → product-performance review. Ineffective action → reassess cause, risks and supporting evidence → revised proposal → human authorisation. This is a proposed operating sequence, not a prescribed regulatory sequence. Review findings can also initiate a new assessment rather than merely close an existing action.
- SignalLink the observation to its product, process and source record.
- AssessEvaluate impact, uncertainty and any immediate containment needs.
- InvestigateTest plausible causes against traceable evidence.
- DecideSelect actions, owners and effectiveness criteria.
- AuthoriseObtain human approval before implementing the agreed change.
- ImplementRecord execution, training and affected document versions.
- ReviewCheck effectiveness; reopen unresolved issues and update shared knowledge.
07PQS improvement and the knowledge and risk enablers
For pharmaceutical APIs and drug products in the cross-regional ICH Q10 guidance model, Section 3 focuses on continual improvement of process performance and product quality, while Section 4 addresses managing and continually improving the pharmaceutical quality system itself. Operationally, distinguish improving what the process delivers from improving how the organisation identifies, evaluates and resolves quality issues.
For pharmaceutical APIs and drug products in that cross-regional model, ICH Q10 Annex 2 identifies knowledge management and quality risk management as enablers applicable throughout the lifecycle stages. The following are practical applications, not prescribed methods:
- Knowledge management: preserve why process settings were selected, which explanations were rejected and why, what transfer assumptions remain unresolved, and the basis for change decisions. Use retrievable links to supporting records so a later reviewer can reconstruct the reasoning. Link that rationale into change control, rather than retaining only the approved outcome.
- Quality risk management: make uncertainty, potential quality impact and the reasoning behind priorities visible. A quality risk register can connect an assessment with its assumptions and supporting evidence; a risk rating alone does not explain the decision.
For example, assume a site repeatedly escalates quality concerns late. Treat this as a possible system weakness, not automatically an isolated training issue. Examine whether decision authority is unclear or relevant evidence is difficult to retrieve. A governance change might clarify escalation ownership or make decision records accessible. Evaluate whether subsequent cases reach the appropriate decision-maker with usable evidence and whether decisions become better supported. Closing the governance action records completion; examining how later decisions are made tests whether the PQS actually improved.
08Jurisdiction and product scope: Q10 alongside regional GMP
A practical applicability assessment starts with the product, manufacturing activity and destination market—not the site’s choice of quality-system model. Use the matrix below to separate guidance relationships from the legal assessment of applicable current good manufacturing practice.
| Jurisdiction or framework | Product scope | Source category | Relationship to Q10 | Local applicability question |
|---|---|---|---|---|
| Cross-regional ICH model | Pharmaceutical APIs and drug products | Guidance | Q10 provides the pharmaceutical quality-system model | Which regional GMP text applies to this product and activity? |
| US FDA validation guidance | Human and animal drugs and biological products, including APIs | Guidance | Lifecycle alignment with Q10 | Which recommendations and separately applicable regulations concern this process? |
| European Union | Medicinal-product GMP scope check | Jurisdiction-specific legal assessment | Assess regional applicability separately | What does current EU GMP text cover for this product and activity? |
| United Kingdom | Medicinal-product GMP scope check | Jurisdiction-specific legal assessment | Assess regional applicability separately | What does current UK GMP text cover for this product and activity? |
For US human and animal drugs and biological products, including APIs, FDA Process Validation guidance, Section I describes general principles and approaches to process validation. For those US pharmaceutical products, it aligns validation activities with a product lifecycle concept and existing FDA guidance. This US pharmaceutical guidance presents recommendations, not independently enforceable responsibilities or a replacement for applicable regulations.
For pharmaceutical APIs and drug products in the cross-regional model, ICH Q10 Section 1.1 positions Q10 as including applicable GMP regulations and complementing Q8 Pharmaceutical Development. That relationship does not establish the local applicability of a particular GMP provision.
For food, supplements, devices and non-pharmaceutical chemicals, reuse the operating examples only as analogies. A separate product-and-jurisdiction assessment is needed before translating them into site controls; pharmaceutical guidance is not a universal compliance claim.
09Evidence model, system interfaces and exception recovery
The following data-model diagram illustrates a proposed site design for tracing pharmaceutical quality decisions. It is an information-management practice, not a vendor database schema or a prescribed ICH Q10 structure.
Product and process context → signal → assessment → action → review decision
Connect these records through stable identifiers: product, batch, process version, signal, assessment, action and decision IDs. Retain source-record references and the version used for each assessment, alongside named reviewers, decision status and rationale. This makes it possible to reconstruct what a reviewer considered, rather than merely retrieve the latest value. Such links are a practical application of data integrity by design.
For an illustrative manufacturing–laboratory–quality interface, designate manufacturing as authoritative for execution context, the laboratory for analytical results, and the quality system for assessments and decisions. In MES-to-LIMS integration, preserve the originating result identifier, units, method version and associated process version; reconcile transferred records against their sources. A copied result and its source are related records, not independent observations, so do not count both in a trend.
Design manufacturing-to-quality interfaces to surface missing records, isolate duplicate messages and preserve failed-transfer history. After replay, reconcile identities, versions and values, then route affected assessments or decisions for human reassessment. Record whether the decision changed and why. Keep data-recovery status separate from product or process acceptability: successful transfer resolves a technical exception, not the quality question. Assign an owner to each unresolved exception so recovery and reassessment remain connected.
Product and Process
Context shared by all linked records.
- product_id
- process_version
- lifecycle_stage
- knowledge_record_id
Quality Signal
Observation linked to its originating evidence.
- signal_id
- batch_id
- source_record_id
- observed_at
Assessment
Reasoned evaluation linked to the signal.
- assessment_id
- signal_id
- risk_rationale
- reviewer_id
Action and Change
Response linked to assessment and effectiveness checks.
- action_id
- assessment_id
- change_id
- action_owner_id
- effectiveness_criterion
Review decision
Authorised decision linked to actions and supporting evidence.
- decision_id
- action_id
- evidence_record_id
- approver_id
- decision_status
10Worked cases: signals, transfer gaps and duplicate evidence
These hypothetical cases illustrate operational reasoning, not regulatory thresholds or actual outcomes.
Monitoring: accepted results with a downward pattern
- Assumptions and units: A commercial oral-drug process has comparable assay measurements expressed in mg per tablet; batch results are accepted.
- Observed evidence: Results show a downward pattern, without an established cause.
- Reasoning: Confirm comparability of methods, sampling, calculation basis and product strength. Assess potentially affected batches and product-quality implications. Compare material variation, processing changes and analytical effects as alternative explanations, using laboratory records and batch execution history.
- Decision: If the pattern persists after comparability checks, pursue a focused investigation and risk-based monitoring. The pattern alone establishes neither a root cause nor a failed batch.
- Retained record: Preserve the dataset, comparability checks, impact assessment, competing explanations, investigation evidence and rationale for the next decision.
Transfer: different dryer, unresolved process understanding
- Assumptions and units: A receiving site uses a different dryer; temperature is recorded in degrees Celsius and moisture in mg per g.
- Observed evidence: Operating settings are copied, but their relationship to receiving-site drying performance remains unestablished.
- Reasoning: Matching settings does not establish matching heat transfer, airflow or moisture distribution. Identify missing knowledge about load, sensor location, sampling and endpoint behaviour. Define an evaluation connecting operating conditions to moisture distribution and relevant product attributes.
- Decision: The receiving team could accept the transfer if results support the intended process performance against predefined criteria; unresolved gaps could instead warrant further evaluation.
- Retained record: Link questions, evaluation criteria and results through requirements traceability, recording the receiving team's acceptance decision and rationale.
Integration: replayed evidence inflates an event count
- Assumptions and units: An interface replay creates duplicate records for one originating event; the metric counts events per batch.
- Observed evidence: Records share a source identifier and replay history.
- Reasoning: Trace that identifier to the originating event, distinguish duplication from genuinely separate events, isolate the duplicate from counting and reconcile the dataset.
- Decision: Recalculate the metric and reconsider any affected review conclusion; retain it only if the reconciled evidence still supports it.
- Retained record: Preserve original records, replay details, reconciliation rationale and review disposition for audit trail review, rather than erasing the recovery history.
11Implementation checklist: test the connections, not just the procedures
Use this practical self-assessment to test the connections described above, not merely whether procedures exist. Follow a recent issue from its initial signal through review and follow-up, asking whether another reviewer could reconstruct the decision without relying on personal recollection.
| Failure mode | What it obscures | Practical control |
|---|---|---|
| Orphaned trend | Missing product or process context | Stable evidence links |
| Administratively closed action | Unknown effectiveness | Predefined assessment criterion |
| Unreviewed process change | Unclear resulting state | Linked implementation and follow-up records |
| Missing interface records | Incomplete decision inputs | Reconciliation and exception ownership |
Implementation and review checklist
- Establish product and jurisdiction scope before choosing the site's assessment criteria.
- Map lifecycle handovers, including the knowledge and records each receiving team needs.
- Assign decision and escalation owners, including ownership of unresolved cross-team issues.
- Define monitoring context so signals remain connected to products, processes and operating conditions.
- Connect investigations with actions and changes; locate the planned effectiveness assessment.
- Test interface recovery, including reconciliation and exception ownership across MES–QMS interfaces.
- Trace review decisions to the evidence considered and the resulting follow-up.
- Reassess whether system changes solved the identified weakness rather than merely completing assigned tasks.
For a small site, start with existing responsibilities, records and review forums. Address missing connections before building a duplicate system. Record who owns each gap, what evidence would demonstrate resolution and where that conclusion will be reviewed. This is an improvement exercise, not a certification checklist.
12V5 Ultimate support: assembled evidence for human review
Frequently asked questions
Q.Does adopting ICH Q10 replace a pharmaceutical site's GMP obligations?+
No. Treat Q10 adoption as quality-system improvement, not as evidence that every applicable GMP obligation has been met. A practical implementation keeps a separate view of applicable legal duties and uses the PQS to organise responsibilities, controls and improvement work. During a gap assessment, distinguish a missing operational control from a missing document describing that control. Keep the site's cGMP framework as the compliance baseline rather than treating a Q10-labelled manual as a substitute.
Q.How can a small pharmaceutical site apply Q10 without building a parallel quality system?+
Start by mapping existing procedures, records and decision owners to the intended PQS functions. For pharmaceutical companies applying the international guideline, ICH Q10 section 1.8(f) recommends monitoring, CAPA, change management and management review as system elements. A practical small-site approach connects existing change control and CAPA workflows rather than creating duplicate registers. Identify gaps in ownership, escalation and feedback, then add only the controls needed to close those gaps.
Q.What should a contract manufacturer and product owner exchange to make their PQS interfaces work?+
A practical exchange includes current process instructions, relevant development knowledge, quality trends, deviations, proposed changes and unresolved risks. The useful test is whether each party can explain what information it receives, who evaluates it and what decision follows. Use the quality agreement to document responsibilities, escalation routes and access arrangements. In routine CMO/CDMO management, reconcile open issues across both organisations rather than assuming that sending a report means the recipient has assessed its implications.
Q.When can a quality trend justify investigation even if every batch meets specification?+
A trend can justify investigation when it suggests a meaningful change in process behaviour, even without a failing batch. For pharmaceutical companies applying international ICH Q10 section 3.2.2, monitoring trends are among the inputs identified for investigation and CAPA. In practice, examine persistent drift, widening variability or repeated adjustments against process knowledge and data reliability. An out-of-trend assessment can distinguish an emerging process problem from sampling effects or measurement variation before selecting an appropriate response.
Q.How is management review of the PQS different from batch disposition?+
In practical terms, batch disposition addresses whether a particular batch is suitable for its intended next step; PQS review addresses whether the system is functioning effectively across products, processes and recurring issues. A useful management review considers patterns, unresolved risks, resource constraints and improvement priorities. Its outputs typically identify decisions, accountable owners and follow-up evidence. Keep those system-level actions distinct from the documented assessment and authorisation associated with batch release; neither decision substitutes for the other.
Q.What should a receiving site do when technology-transfer knowledge is incomplete?+
Treat missing knowledge as an explicit transfer risk rather than filling gaps with undocumented assumptions. A practical receiving-site assessment lists what is unknown, which process decisions depend on it and what evidence could resolve it. During technology transfer, request the rationale behind controls, not just operating instructions. Use targeted studies, source-site discussions or additional monitoring where justified. Record unresolved questions in a quality risk register, with owners and decision criteria for progressing or pausing affected activities.
Q.What quality-system work continues after a pharmaceutical product is discontinued?+
For pharmaceutical companies applying the international guideline, ICH Q10 section 3.2.2, Table II recommends continuing CAPA after discontinuation. A practical closeout therefore keeps unresolved investigations visible rather than closing them merely because manufacturing has stopped. Consider whether findings affect remaining stock or related products, preserve access to relevant records and assign continuing ownership. An effectiveness check can assess whether completed actions achieved their purpose even when no further batches of the discontinued product are planned.
Q.Can electronic batch records replace the judgement needed to close a CAPA?+
No. An electronic batch record can support the evidence review, but CAPA closure is a reasoned quality decision, not simply a completed workflow. In practice, the reviewer assesses whether the proposed cause fits the evidence, the actions address that cause and the effectiveness assessment is convincing. If there has been insufficient opportunity to observe recurrence, document that limitation and define follow-up. Human reviewers retain responsibility for approving closure; the presence of complete electronic records does not demonstrate effectiveness.
Primary sources
Explore this topic
ICH Q10 sits inside 2 overlapping topic clusters in our glossary. Every neighbour is one click away.
Drug-product cGMP rules, ICH Q-series, and the regulators that enforce them.
ICH Q8/Q11/Q12 toolkit — Quality by Design, design space, control strategy, CPV and lifecycle management.
Want to see how ICH Q10 could fit into your own records and workflows? Explore the related V5 pages or talk to our team about what applies to your operation.
