A recurring deviation is not automatic proof that the previous CAPA failed. It is, however, a signal that the earlier investigation, root-cause conclusion, CAPA scope and effectiveness check should be challenged against the new evidence.
The practical question is not simply, “Has this happened before?” It is: “Does the new event reveal the same failure mechanism, an incomplete control strategy or a wider system weakness?”
What recurrence means—and what it does not mean
Two events can look similar but arise from different mechanisms. Conversely, events recorded under different categories can share the same systemic cause. A reliable recurrence assessment therefore cannot depend only on the deviation title, equipment number or event code.
A repeat should trigger a structured comparison of:
- the failure mode and the point in the process where it occurred;
- the affected product, batch, equipment, shift, supplier and operating conditions;
- the evidence used to establish the previous root cause;
- the actions implemented and the risks they were intended to control;
- the approved effectiveness criteria, observation window and data source.
Do not label the old CAPA “ineffective” before comparing the mechanisms. Do not label the new event “isolated” before testing the possible connection.
The regulatory baseline
EU GMP Chapter 1: investigate the system, then monitor effectiveness
EU GMP Chapter 1 expects an appropriate level of root-cause analysis during deviation investigations. Where human error is suspected, the conclusion should be justified after checking that process, procedural or system-based problems have not been overlooked. Appropriate CAPAs should be taken, and their effectiveness should be monitored and assessed using Quality Risk Management principles.
The same chapter also expects Product Quality Reviews to consider significant deviations, their investigations and the effectiveness of the resulting CAPAs. This makes recurrence analysis part of the wider state-of-control assessment—not merely a field in an individual deviation record.
ICH Q10: CAPA is a feedback and improvement system
ICH Q10 places CAPA alongside process and product monitoring, change management and management review as a core element of the Pharmaceutical Quality System. It expects a structured, risk-proportionate investigation process and states that, in commercial manufacturing, the effectiveness of actions should be evaluated.
That connection matters: a recurrence can be a signal from monitoring that should feed back into investigation, process understanding and continual improvement.
United States: extend the investigation where the evidence points
Under 21 CFR 211.192, unexplained discrepancies or failures to meet specifications must be thoroughly investigated. The investigation must extend to other batches of the same drug product and to other products that may be associated with the specific failure or discrepancy, with conclusions and follow-up documented.
This does not create a universal rule that every repeat event has the same cause. It does reinforce the need to test the extent of the problem rather than confining the assessment to the record immediately in front of the investigator.
A four-step recurrence review: Signal, Compare, Challenge, Verify
1. Signal: define the possible recurrence
Start with a neutral signal statement. Describe what appears connected and why, without prejudging the conclusion.
Example: “A second line-clearance discrepancy occurred on the same packaging family within the approved CAPA observation window, involving a different operator.”
Search beyond exact matches. Include related terminology, adjacent process steps, similar failure modes, other lines or products using the same procedure, and complaints, audit findings or maintenance records that may reveal the same weakness.
2. Compare: test whether the mechanisms are related
Build a side-by-side comparison of the old and new events. Ask whether the same control failed, whether a different control failed for the same reason, or whether the similarity is superficial.
- What was physically observed in each event?
- Which barrier should have prevented or detected it?
- What evidence supports the earlier root-cause conclusion?
- Did the new event occur inside or outside the scope of the old CAPA?
- Were the relevant operating conditions materially different?
3. Challenge: reassess root cause and CAPA scope
If the events are related, reopen the reasoning—not only the action list. A repeat may show that the previous cause was a symptom, that the action controlled only one local manifestation or that implementation changed the procedure without changing the work system.
Be especially cautious when the earlier conclusion was “human error” and the new event involves another person. The MHRA Inspectorate has explicitly highlighted that repeated related problems across different people make human error less likely to be the true root cause, and that retraining alone is unlikely to be effective where system factors remain.
4. Verify: evaluate effectiveness with predefined evidence
CAPA completion and CAPA effectiveness answer different questions. Completion shows that an action was implemented. Effectiveness shows that the intended risk reduction was achieved, without unacceptable unintended consequences.
A defensible effectiveness check should specify:
- the risk or failure mechanism expected to change;
- a measurable acceptance criterion;
- the population, data source and observation period;
- who reviews the evidence and when;
- what result triggers extension, escalation or a new investigation.
CAPA completion versus effectiveness
| Question | Completion evidence | Effectiveness evidence |
|---|---|---|
| What is being demonstrated? | The approved action was implemented. | The action achieved its intended control objective. |
| Typical evidence | Approved SOP, training record, installed device, closed change. | Trend, error-rate reduction, challenge test, audit sample, process performance. |
| Timing | At implementation or shortly afterwards. | After a justified observation window or sufficient opportunities for recurrence. |
| Failure signal | Action overdue or not implemented as approved. | Acceptance criterion missed, repeat mechanism observed or new risk introduced. |
| Decision | Close the implementation task. | Confirm effectiveness, extend monitoring, modify the CAPA or reopen the investigation. |
Fictional case: the line-clearance deviation that returned
A packaging site records a line-clearance deviation after obsolete printed material is found during start-up. The investigation concludes that the operator did not follow the checklist. The CAPA revises the work instruction and retrains the team. Training is completed, and the CAPA is closed.
Four months later, a similar event occurs on another shift. Treating the event as automatic proof of CAPA failure would be premature. Treating it as an unrelated operator error would also be weak.
The comparison shows that both events occurred during short changeovers, when the physical line-clearance sequence and the electronic checklist were not aligned. The second operator followed the displayed sequence, but a removable guard created a blind inspection point. The earlier “human error” conclusion had not tested the work system.
The revised CAPA therefore addresses the mechanism: redesign the clearance sequence, add a poka-yoke visual control at the blind point, update the electronic checklist, verify the change under representative changeovers and monitor a defined number of opportunities across all affected shifts.
The lesson is not that training is never useful. It is that training should address a demonstrated knowledge or skill gap and should not substitute for correcting a process, design or system weakness.
Practical review checklist
- Recurrence logic: Is the connection based on mechanism and evidence, not only on labels?
- Extent: Have other batches, products, lines, shifts, suppliers and related PQS records been considered where relevant?
- Root cause: Does the evidence support the conclusion, and have system causes been actively tested?
- CAPA linkage: Does each action address a confirmed or well-supported causal factor?
- Scope: Was the action applied wherever the same mechanism can exist?
- Effectiveness criteria: Were measurable criteria and an appropriate observation window defined before review?
- Implementation: Is there evidence that the change works in routine conditions, not only on paper?
- Escalation: Is the response clear if the criterion is missed or another related signal appears?
- Management review: Are recurrent signals and CAPA performance visible at the right governance level?
FAQ
Does every recurring deviation prove that the previous CAPA was ineffective?
No. Similar events may have different causes, and different-looking events may share a cause. The organisation should compare the failure mechanisms, CAPA scope and effectiveness evidence before concluding.
Should a CAPA be reopened when a repeat occurs?
Not automatically. First assess the relationship and the approved decision rules. A related repeat may justify reopening the investigation, revising the CAPA, extending its scope or recording a failed effectiveness check. The rationale should be documented.
Is retraining an acceptable CAPA?
Yes, when evidence shows a genuine knowledge or skill gap and the training is targeted and verified. It is weak when used as the default response to a problem caused by procedure design, equipment, workload, interfaces or other system conditions.
How long should an effectiveness check last?
There is no universal duration. The observation window should be long enough—or contain enough process opportunities—to detect the failure mechanism with meaningful confidence, and should be proportionate to risk and event frequency.
What if the true root cause cannot be confirmed?
EU GMP Chapter 1 allows consideration of the most likely root cause or causes when the true cause cannot be determined. The uncertainty should be explicit, plausible causes should be addressed based on risk, and monitoring should test the assumptions.
Official sources
- European Commission, EudraLex Volume 4, EU GMP Chapter 1: Pharmaceutical Quality System
- ICH Q10, Pharmaceutical Quality System
- 21 CFR 211.192, Production record review
- MHRA Inspectorate, Investigations in the GMDP environment
- PIC/S PI 054-1, PQS effectiveness and risk-based change management
Make deviation and CAPA decisions easier to defend
The GMP Deviations & CAPA: Effective Implementation and Audit Defensibility guide provides a structured process, root-cause tools, effectiveness-check logic and ready-to-use templates for investigations that need to stand up under inspection.
