Overview. Deviation management provides a structured response when operations do not proceed as approved or expected. CAPA should address demonstrated causes and reduce recurrence. Cannabis operations face recurring deviation themes around botanical variability, microbiology, sampling, potency, drying and supplier performance.
Deviation detection and containment
The event should be recorded promptly and immediate actions taken to protect product, data and operations. Potentially affected materials should be placed under appropriate status control.
Classification and impact
Severity should reflect patient risk, product impact, compliance and recurrence. Classification should not be reduced to whether a batch can still be released.
Investigation planning
The investigation should define the problem clearly, preserve evidence and assess relevant people, materials, equipment, methods, environment and data.
Cannabis-specific investigation examples
High microbial results may involve cultivation, drying, sampling or laboratory recovery. Potency variability may involve heterogeneity, mixing, fill drift or analytical extraction. Low extraction yield may involve raw-material moisture, particle size or equipment performance.
Root-cause analysis
Tools such as five whys, fishbone analysis and fault trees can support thinking, but the conclusion must be evidence based. 'Operator error' is rarely an adequate root cause without understanding system factors.
CAPA design
Corrections fix the immediate issue; corrective actions address the identified cause; preventive actions reduce similar risks elsewhere. Actions should be proportionate and owned.
Effectiveness checks and trend review
CAPA should include measurable success criteria. Repeated events, overdue actions and ineffective CAPA should be visible to management.
Practical reference table
| Deviation | Possible root cause | Potential CAPA |
|---|---|---|
| High microbial flower result | Uneven drying or sampling weakness | Drying-map improvement and sampling redesign |
| Potency variability in oil | Insufficient bulk mixing | Mixing study and in-process controls |
| Residual solvent OOS | Vacuum endpoint not controlled | Validated endpoint and alarm |
| Label discrepancy | Unrecorded reject destruction | Reconciliation and line-control improvement |
| Repeated HPLC reintegration | Weak processing rules | Controlled integration procedure and review |
Control and decision path
Frequently asked questions
What is the difference between correction and CAPA?
A correction addresses the immediate problem; CAPA addresses causes and recurrence risk.
Is operator error a root cause?
Usually not by itself. The investigation should identify why the system allowed the error.
When should a batch be held?
When the deviation could affect identity, strength, quality, safety or compliance.
What makes an effectiveness check useful?
It uses predefined evidence showing that recurrence or risk was actually reduced.
Should minor deviations be trended?
Yes. Repeated minor events may reveal a major systemic weakness.
Primary references and guidance
- EU GMP Part I, Chapter 1
- ICH Q9(R1)
- ICH Q10
- PIC/S PE 009
- FDA Quality Systems Approach guidance
- EU GMP Part I, Chapters 4, 5 and 6
- MHRA investigation expectations
Confirm the current effective revision and national applicability before operational or regulatory use.