Case Study: OOS Investigation Using FMEA
Scenario Overview
A pharmaceutical manufacturer receives an Out-of-Specification (OOS) result during finished product assay testing.
The product specification requires:
95.0% - 105.0% label claim
The initial laboratory result reports:
92.8% assay
The batch has not yet been released.
An OOS investigation is initiated.
At the start of the investigation, the organization does not know whether the failure originated from:
laboratory error
sample preparation
analytical method issues
manufacturing process variation
raw material variability
equipment performance
broader process control weakness
Rather than immediately pursuing a single explanation, the investigation team decides to use Failure Mode and Effects Analysis (FMEA) to evaluate potential failure pathways systematically.
Initial Investigation Challenge
The team faces a common problem.
Several explanations appear plausible.
Each explanation carries different implications for:
product quality
patient risk
batch disposition
CAPA requirements
regulatory exposure
Without structured evaluation, investigators risk:
confirmation bias
premature conclusions
incomplete root cause analysis
Structured failure analysis can improve visibility of possible causes before evidence becomes sufficient to identify the most likely explanation.
Step 1 — Identify Potential Failure Modes
The investigation team develops an initial list of possible failure modes.
| Failure Mode | Potential Effect |
|---|---|
| Sample preparation error | Incorrect assay result |
| Instrument calibration issue | Analytical inaccuracy |
| Analytical method execution error | Invalid result |
| Blend uniformity problem | Reduced potency |
| Raw material potency variation | Reduced assay result |
| Incorrect manufacturing parameter | Product quality impact |
At this stage, none of these explanations have been confirmed.
The purpose is to ensure plausible failure pathways remain visible.
Step 2 — Evaluate Severity
The team evaluates potential impact if each failure mode proves true.
Several failure modes receive elevated severity ratings because they could affect:
product quality
dose delivery
patient exposure
release decisions
For example:
| Failure Mode | Severity Consideration |
|---|---|
| Sample preparation error | Limited impact if isolated laboratory event |
| Blend uniformity issue | Potential batch-wide impact |
| Manufacturing process issue | Potential systemic impact |
| Raw material variability | Potential multiple-batch impact |
Severity helps prioritize investigation effort but does not identify root cause.
Step 3 — Evaluate Occurrence
Investigators review historical information.
Questions include:
Has this failure occurred previously?
Are similar deviations trending upward?
Have related CAPAs been effective?
Is supplier performance stable?
Are manufacturing parameters historically capable?
The review identifies:
no prior laboratory assay failures
three recent blend uniformity deviations within six months
one open CAPA related to mixing consistency
This information increases concern regarding manufacturing-related failure pathways.
Recurrence often changes significance of operational exposure.
Step 4 — Evaluate Detectability
The team next evaluates detectability.
Key questions include:
Would existing controls identify the failure before release?
How quickly would the issue become visible?
Are monitoring systems reliable?
The evaluation reveals:
| Failure Mode | Detectability Assessment |
|---|---|
| Sample preparation error | High detectability through retesting |
| Instrument calibration issue | High detectability through calibration review |
| Blend uniformity issue | Moderate detectability |
| Manufacturing variability | Moderate to low detectability |
| Raw material variability | Moderate detectability |
This analysis becomes important because lower detectability increases uncertainty regarding the true extent of exposure.
Failures that remain difficult to identify often require greater investigation attention.
Step 5 — Prioritize Investigation Focus
The team combines available evidence.
Although laboratory error remains possible, current information suggests greater concern regarding:
blend uniformity performance
manufacturing consistency
effectiveness of prior corrective actions
The investigation scope is expanded to include:
batch manufacturing records
mixing parameter review
historical trend evaluation
review of previous blend-related deviations
FMEA does not identify the answer.
It helps direct investigation effort toward the most meaningful failure pathways.
Investigation Outcome
The expanded review identifies:
inconsistent mixing performance
incomplete implementation of a prior CAPA
recurring blend variability across multiple batches
The OOS result is determined to reflect an actual manufacturing issue rather than laboratory error.
The original assumption of isolated analytical failure is rejected.
This demonstrates one of the primary benefits of structured FMEA analysis:
the investigation remained open to evolving evidence rather than becoming anchored to the first plausible explanation.
CAPA Development
Based on the findings, the organization implements:
revised mixing parameter controls
enhanced process monitoring
verification of CAPA implementation effectiveness
additional operator qualification activities
trend monitoring for blend consistency
The CAPA strategy focuses on reducing recurrence risk rather than simply addressing the individual OOS result.
Corrective actions should reduce operational exposure rather than merely close investigation records.
Batch Disposition Decision
Because the investigation confirms a manufacturing-related potency issue, the affected batch is rejected.
Additional actions include:
evaluation of potentially impacted batches
reassessment of process capability
review of historical trend information
Release decisions should reflect actual operational understanding rather than procedural completion alone.
Lessons Learned
Several important lessons emerge from this case:
Initial assumptions may be incorrect.
Recurrence trends often provide critical context.
Detectability influences investigation confidence.
CAPA effectiveness should remain visible over time.
Structured failure analysis improves investigation quality.
Most importantly:
The value of FMEA was not that it produced a numerical score.
Its value was that if prevented the investigation from narrowing prematurely around an unproven explanation.
What Good Looks Like
A mature OOS investigation:
evaluates multiple failures pathways
integrates recurrence information
considers detectability limitations
updates conclusions as evidence develops
links findings to CAPA and disposition decisions
The objective is not simply identifying what failed.
The objective is understanding why the failure occurred, how much exposure exists, and what controls are necessary to prevent recurrence.
Operational Perspective
Many weak OOS investigations begin with a likely explanation and spend the remainder of the investigation attempting to confirm it.
Strong investigations operate differently.
They maintain visibility of multiple plausible failure pathways until evidence becomes sufficient to narrow conclusions confidently.
FMEA provides structure for this process by helping investigators evaluate where operational exposure is most likely to exist while reducing the risk of confirmation bias and premature closure.
