How to Conduct a Deviation Investigation
From "something departed from procedure" to a defensible, closed record.
What a how-to is not
A how-to is SPEQ’s practitioner method, not a procedure. It does not replace your own SOP, it is not a validated approach, and the judgement calls in it belong to your quality unit.
A deviation is any departure from an approved procedure, specification, or standard. The investigation decides whether the batch is safe, why it happened, and whether it will happen again. Handled well it is a routine quality event; handled poorly it becomes the thread an inspector pulls.
- 1
Document and contain immediately
Record the deviation contemporaneously — what happened, when, who, what product is affected — and contain any affected material before it progresses. A late or reconstructed record weakens everything downstream.
- 2
Classify the deviation
Assign a criticality (minor / major / critical) based on potential impact to product quality and the patient. Classification sets the depth of investigation and the approval level required.
- 3
Investigate to root cause
Use a structured tool and drive to a systemic cause. Resist "operator error" as an endpoint — it is almost always a symptom of a system that made the error easy.
- 4
Assess product impact
Determine, with a documented rationale, whether product quality could be affected — this batch and any other batch that shares the cause. The impact assessment is what justifies the disposition.
- 5
Decide the disposition
Based on the impact assessment, disposition the affected material — release, reject, rework, or further evaluation — with QA approval and a science-based rationale.
- 6
Trigger a CAPA if warranted
If the cause is systemic or recurrent, raise a CAPA. A single significant deviation, or a trend of minor ones, is a signal that the system — not just the batch — needs attention.
- 7
Close with QA approval and full traceability
Close the deviation with the investigation, impact assessment, disposition, and any CAPA all traceable and approved by QA.
- !Writing the record from memory at the end of the shift instead of contemporaneously.
- !Stopping the root-cause analysis at "operator error".
- !Skipping the product-impact assessment, or not extending it to sister batches.
- !Downplaying a real deviation so it does not draw attention — which reads as a culture that hides problems.
How to Conduct a Deviation Investigation: frequently asked questions
Common questions on conduct a deviation investigation.
What is the difference between a deviation and a CAPA?
A deviation is the event — a departure from a procedure or specification — and its investigation and disposition. A CAPA is the follow-on action taken when the cause is systemic or recurrent. Not every deviation raises a CAPA, but a deviation investigation is where you decide whether one is needed.
How do you classify a deviation?
By its potential impact on product quality and the patient — typically minor, major, or critical. The classification drives how deep the investigation goes and who must approve it. Classify early, and re-classify if the investigation reveals greater impact.
Why is "operator error" a weak root cause?
Because it is almost always a symptom, not a cause. If two materials were easy to confuse or a step was ambiguous, the system made the error easy — and only a systemic fix prevents recurrence. Stopping at "operator error" leaves that system in place.
When must a deviation trigger a product-impact assessment?
Always, when the deviation could plausibly affect a batch that has been or will be released. The impact assessment — documented, science-based, and extended to any sister batches sharing the cause — is what justifies the disposition decision.