Quality Systems
The CAPA subsystem is the most inspected element of FDA's Quality System Inspection Technique (QSIT). A broken CAPA system is the root cause of most FDA 483 observations — and a well-functioning one is the leading indicator of a quality culture that prevents them.
Assess Your CAPA SystemCAPA in Practice
FDA's Quality System Inspection Technique (QSIT) identifies CAPA as one of four major subsystems that investigators evaluate in every device inspection. Under 21 CFR 820.100, manufacturers must establish procedures for corrective and preventive action that include: analyzing quality data sources to identify existing and potential causes of nonconforming product or processes, investigating the cause of nonconformities, identifying and implementing corrective and preventive actions, verifying or validating those actions, disseminating information related to quality problems to those responsible for quality, and reviewing submitted information and relevant CAPA activities.
In practice, FDA investigators inspect a sample of CAPA records to assess whether the system is working. They look for specific evidence: Was the problem triaged appropriately? Is there a documented root cause analysis that goes beyond surface-level description? Are corrective actions specific and tied to the root cause? Were preventive actions extended to similar processes or products? Was effectiveness verified — and is there objective evidence of that verification? Is the CAPA closed in a reasonable timeframe?
The most common CAPA-related 483 observations fall into three patterns: inadequate root cause analysis (identified symptom rather than cause), missing effectiveness verification (the action was implemented but there's no documented proof it worked), and scope too narrow (the CAPA addressed the specific instance but didn't assess whether the same problem exists elsewhere in the organization).
Root cause analysis is where CAPA systems most consistently fail FDA inspection. "Operator error" is never an acceptable root cause for a quality system issue — if an operator made an error, the question is why the system allowed that error to occur without detection. Was training inadequate? Was the procedure unclear? Was the work environment designed in a way that made the error probable? Was there no independent verification step? Each of those is a root cause; "operator error" is a symptom.
We use structured RCA methodologies matched to the complexity of the issue: 5-Why analysis for straightforward problems, Ishikawa (fishbone) diagrams for multi-factor process issues, fault tree analysis for complex systemic failures, and FMEA-based approaches for proactive risk identification. The RCA methodology used should be documented, the analysis should show the work, and the conclusion — the identified root cause — should be clearly stated and defended.
21 CFR 820.100(a)(7) requires that manufacturers verify or validate corrective and preventive actions to ensure those actions do not adversely affect the finished device and that they are effective. Yet effectiveness verification is the most commonly observed CAPA deficiency — either it isn't done, it's done too early (before there's sufficient data to demonstrate the fix worked), or the verification method doesn't actually test whether the root cause was eliminated.
We design effectiveness checks that are temporally appropriate (scheduled after enough time has passed to generate meaningful data), methodologically sound (testing the right metric to confirm the root cause is gone), and documented in a way that constitutes objective evidence of effectiveness. For systemic issues, we design monitoring programs that continue after CAPA closure to provide early warning if the problem recurs.
Get Started
Most companies don't know their CAPA system has deficiencies until an FDA investigator points them out. A 30-minute assessment call can tell you where your system stands — and what needs to change before your next inspection.