Every manufacturing site has deviations. What inspectors look at is whether yours are classified consistently, investigated in proportion to their risk, and traced to a cause that actually explains what happened.
What the EU GMP Guide expects
Chapter 1 of the EU GMP Guide requires that an appropriate level of root cause analysis is applied when deviations are investigated. Where the true root cause cannot be determined, the most likely root causes should be identified and addressed. Where human error is suspected or identified as the cause, this should be justified, and the investigation must show that process, procedure or system-related causes have not been overlooked. Appropriate corrective and preventive actions should then be identified, and their effectiveness monitored.
Classification: proportionate, not arbitrary
A deviation procedure should define categories, typically minor, major and critical, with clear criteria that do not depend on who happens to be reviewing the event. The classification should consider:
- the potential impact on patient safety, product quality and data integrity;
- whether other batches, products or sites may be affected;
- whether the marketing authorisation or a validated state is touched;
- whether the event has occurred before.
Classification should be allowed to change as the investigation progresses. A deviation downgraded at the outset and never revisited is a common finding.
Investigation: scope before cause
A sound investigation first establishes the facts and the scope: which batches, equipment, materials and time period may be involved. Only then does it move to cause. Useful practice includes:
- starting the record at the time of discovery, with immediate containment actions documented;
- interviewing the people involved promptly and recording what was observed, not what was assumed;
- using a structured tool such as a fishbone diagram or 5 Whys, chosen to suit the complexity of the event;
- testing the proposed root cause against the evidence, asking whether it explains every observation.
The trap of “operator error”
Inspectors regularly challenge conclusions that stop at an individual. Retraining is rarely a sufficient corrective action by itself if the procedure was ambiguous, the layout invited the mistake or the workload made it likely. Treat human error as a prompt to look further, not as the end of the investigation.
Closing the loop
A deviation is not closed when the report is signed. Impact on batch disposition should be documented for the Qualified Person, linked change controls and CAPA should be tracked, and deviation data should be trended so that recurring causes are visible. Trends feed the product quality review and management review, which is where systemic problems become visible.
A short self-check
- Can you show, for a sample of closed deviations, that scope was defined before cause was proposed?
- Do your classification criteria produce the same category whichever reviewer applies them?
- Are recurring causes visible in a trend report that management actually reads?
How PQRA helps
PQRA supports pharmaceutical manufacturers and marketing authorisation holders in Greece and across Europe with deviation procedures, classification criteria, investigation templates and independent review of closed investigations. We can also assess a sample of your recent deviations against the expectations above and advise where the evidence for root cause is thin.
Contact PQRA to discuss how we can support your quality system.


No responses yet