Handling Repeated Deviations
A deviation occurs, an investigation is completed, the root cause is identified, and corrective actions are implemented. The CAPA effectiveness check is satisfactory, and the record is closed.
Several months later, a similar deviation occurs.
What should happen next?
Repeated deviations are among the most important signals a pharmaceutical quality system can receive. They may indicate that the previous root cause was incomplete, that CAPA did not address the actual failure mechanism, or that an existing control is no longer adequate.
However, recurrence does not automatically establish that the previous investigation or CAPA was wrong.
The challenge is to understand what the new event reveals about the earlier conclusions, current process controls, and remaining risk.
What Makes a Deviation Repeated?
A repeated deviation occurs when an event is sufficiently related to a previous event to warrant assessment of a possible common failure mechanism, recurring weakness, or shared control deficiency.
Two deviations do not need to look identical to be related.
For example, one deviation may involve an incorrect material selection during dispensing, while another involves an incorrect material selection during packaging. The process steps differ, but both could involve similar labeling, identification, or verification weakness.
Conversely, two deviations may have the same description but entirely different causes.
A recurring equipment alarm could result from a mechanical failure in one case and an electrical fault in another.
The relationship between deviations should therefore be evaluated through several dimensions:
Similarity of the observed failure
Common or related root causes
Shared equipment, systems, procedures, or controls
Common human-performance or workflow conditions
Recurrence within a relevant timeframe
Relationship to previous CAPA actions
The objective is to identify whether the events reflect a common or continuing risk.
Start with the New Event
When a potential repeat deviation is identified, the immediate priority remains the current event.
The organization should evaluate containment, affected product or process status, potential quality impact, and necessary escalation.
The historical relationship is important, but it does not replace the current impact assessment.
A new occurrence may affect a different batch, process stage, equipment item, or operating condition. The investigation must establish the specific circumstances of the current event before drawing conclusions from earlier cases.
Initial risk classification and investigation depth may also need adjustment once recurrence is recognized.
As discussed in Pharmaceutical Investigations & CAPA, deviation management connects event recognition, investigation, root cause, CAPA, and recurrence control within the pharmaceutical quality system.
A repeat event tests whether those elements are functioning together.
Review the Previous Investigation
The earlier investigation is a valuable source of information. It should be examined critically rather than accepted as a settled explanation for the new event.
A useful review considers four areas:
Was the previous root cause adequately established?
Examine the evidence that supported the original conclusion.
Was the root cause demonstrated through relevant records, observations, technical assessment, or other evidence? Were plausible alternative causes assessed? Did the conclusion explain both the failure and the relevant prevention or detection controls?
If the earlier investigation attributed an event to operator error, for example, determine whether task design, procedure clarity, training, equipment interface, and verification controls were meaningfully evaluated.
Was the original scope sufficient?
Review the affected population, historical search, equipment or system boundaries, and documented exclusions.
A new event may reveal that the earlier investigation considered only one batch or equipment item even though the underlying condition existed elsewhere.
As explained in Defining Investigation Scope in GMP Deviations, investigation boundaries should follow plausible exposure and the documented failure mechanism.
Did the CAPA address the identified cause?
Determine what the previous CAPA was intended to change.
If the earlier cause was inadequate verification, did the CAPA strengthen verification? If a system-design weakness was identified, did the action improve the system or merely remind personnel to be more careful?
A completed action may still leave the original failure condition unchanged.
What did the effectiveness check demonstrate?
Review the effectiveness criteria, monitoring period, data source, observation population, results, and conclusions.
A previous check may have been completed correctly but may have had limited opportunity to detect recurrence.
These findings help determine whether the new event challenges the previous conclusion or represents a distinct failure.
Compare the Actual Failure Mechanisms
A useful repeat-deviation assessment compares the events side by side.
| Comparison Area | Previous Deviation | Current Deviation |
|---|---|---|
| Observed failure | What occurred? | What occurred now? |
| Process conditions | Under what conditions? | Are those conditions comparable? |
| Root cause | What cause was supported? | Is the same mechanism plausible? |
| Existing controls | What should have prevented or detected the event? | Did the same control operate or fail? |
| CAPA | What changed after the first event? | Was that change applicable to this event? |
| Impact and exposure | What population was affected? | Is there new or broader exposure? |
This comparison prevents two common mistakes: treating every similar event as proof of CAPA failure and treating each new deviation as unrelated simply because its circumstances differ.
When a Deviation Recurs After CAPA Closure
Consider a CAPA that was implemented, monitored for six months, found effective, and formally closed. Three months later, a similar deviation occurs.
Does the new event invalidate the earlier effectiveness decision?
Not automatically.
The earlier effectiveness conclusion was based on the evidence available during the defined monitoring period. The new event introduces additional evidence that must be evaluated.
The review should establish:
whether the new event involves the failure mechanism the CAPA was designed to prevent
whether the CAPA was implemented in the affected area or population
whether the strengthened control remained operational
whether the new event arose under conditions covered by the effectiveness assessment
whether the earlier monitoring period and number of opportunities were sufficient
whether the event reveals a new or previously unrecognized failure mechanism
Three outcomes are possible.
The original failure mechanism has returned. This may indicate that the CAPA was ineffective, lost effectiveness, or did not establish sustainable control.
The visible event is similar, but the cause is different. The previous CAPA may still have been appropriate and effective for its intended failure mechanism.
The relationship remains uncertain. Additional investigation may be needed before deciding whether the previous effectiveness conclusion remains supportable.
A previously satisfactory effectiveness check is part of the evidence. It does not make the CAPA permanently immune to reassessment.
Should the Previous Investigation Be Reopened?
A repeat event should trigger an assessment of whether previous conclusions require reconsideration. It does not necessarily require reopening every related closed investigation.
The decision should follow the site’s approved procedures and the significance of the new evidence.
Reopening or formal reassessment may be appropriate when the new event materially challenges the earlier root cause, impact assessment, scope, CAPA justification, or effectiveness conclusionl.
In other circumstances, the organization may investigate the new event separately while formally linking it to the earlier record and documenting why reopening is unnecessary.
The important point is maintaining traceability between the new information and any prior conclusions it affects.
Reassess CAPA Before Repeating It
When a deviation returns, a common response is to assign the same CAPA again.
An operator may be retrained, an SOP reviewed, or another reminder issued.
Repeating an action can be appropriate if the investigation establishes that the original action was not implemented correctly or was not maintained. However, repeatedly selecting the same action without evaluating why the failure returned provides little assurance of improved control.
The new investigation should determine whether the response needs to address:
an incomplete or incorrect original root cause
a CAPA implemented in too narrow a population
failure to sustain the implemented control
a contributing cause omitted from the earlier CAPA
a new failure mechanism requiring a different action
a broader system weakness across equipment, processes, or functions
Recurrence generally warrants closer scrutiny of a proposed no-CAPA decision. A decision that no additional CAPA is required can still be defensible, but it needs specific evidence and reasoning.
The conditions supporting such decisions are explored in When CAPA Is Not Required.
Practical Example: Repeated Documentation Errors
Consider a missing second-person verification entry in a manufacturing batch record.
The first investigation concluded that the operator overlooked the requirement. The CAPA consisted of retraining the relevant operators, followed by a review of completed records. No further missing entries were identified during the effectiveness period.
Four months after closure, the same documentation discrepancy occurs on another shift.
The new investigation finds that the verification field is separated from the process step where verification is performed. Operators frequently return to complete the entry after the activity.
This new evidence raises an important question: was the original investigation sufficiently focused on workflow and record design?
The response may require an improved verification point, revised batch-record layout, or a process control that confirms verification at the time of execution.
The earlier retraining may have been completed correctly. However, the new evidence suggests that the underlying condition enabling the omission may not have been adequately addressed.
The repeat event has contributed new understanding of the process.
Look Beyond the Number of Occurrences
Deviation trending helps identify recurring signals, but occurrence counts alone can be misleading.
Three similar events over a short period may be significant. However, the interpretation also depends on the number of batches manufactured, operations performed, samples tested, equipment cycles, or other opportunities for the failure to occur.
A meaningful recurrence assessment may consider frequency relative to exposure, severity, time between events, common causes, and the performance of existing controls.
It should also consider whether the event is appearing across multiple shifts, products, equipment items, or departments.
Increasing frequency, broader distribution, or recurrence following CAPA may justify escalation to a cross-functional or management-level review.
Regulatory Perspective
Under 21 CFR 211.192, applicable unexplained discrepancies and failures must be thoroughly investigated, including other potentially associated batches, with documented conclusions and follow-up.
ICH Q10 - Pharmaceutical Quality System describes a CAPA system informed by investigations and quality trends, with investigation effort proportionate to risk and evaluation of CAPA effectiveness.
Neither establishes a universal rule that every repeat event automatically invalidates previous CAPA closure. The organization must evaluate the new evidence and determine the appropriate investigation, CAPA, and follow-up response.
QA Review Perspective
When reviewing a repeated deviation, QA should be able to trace the relationship between the current and earlier events.
The most useful questions are:
Were potentially related deviations identified and compared?
Is the relationship based on evidence or only similar descriptions?
Does the previous root cause remain credible?
Was the earlier CAPA applicable to the current failure?
Do the effectiveness results still support the earlier conclusion?
Has the current impact and broader exposure been assessed?
Is the decision to revise CAPA, reopen an investigation, or maintain existing conclusions justified?
Repeated deviations provide an opportunity to test whether previous investigations produced lasting improvement.
A strong quality system recognizes these signals, challenges earlier assumptions when necessary, and updates controls as new evidence develops.
The measure of investigation effectiveness is the organization’s ability to learn from what has happened and strengthen control of what may happen again.
Explore more on Investigations & CAPA Excellence
Browse VerethiQ resources on deviation handling, root cause analysis, investigation quality, CAPA design, effectiveness checks, recurrence prevention, and investigation governance.
