FDA 483 OBSERVATION · #140 OF 1,923 MOST-CITED
Calibration - adequacy
FDA cited this observation 215 times across 188 firms in recent inspections. Below: how it breaks down by program area and fiscal year, and real examples of how FDA worded it on Form FDA 483.
What this grouping is not
An observation-type page groups FDA’s published citation records. The grouping, the compliance steps and the root causes are SPEQ synthesis, not an FDA classification of any citation, and a count measures what FDA published rather than risk.
What this observation means, and how to comply
SPEQ's practitioner reading of this finding. The regulatory basis is a cited fact; the interpretation, compliance steps, and root causes are SPEQ synthesis — not an FDA classification of the citation.
A calibration program exists but is inadequate in execution — for example, calibration is not traceable to a recognized standard, accuracy and precision limits are undefined, intervals are unjustified, or out-of-tolerance events are not assessed for impact on previously accepted product, as required by §820.72(a). (The equivalent drug-CGMP finding maps to 21 CFR 211.68.)
Why FDA writes it — An inadequate calibration program gives false confidence: instruments carry current stickers yet may not actually measure true. FDA protects against measurement error silently corrupting acceptance decisions — the danger is not the missing program but the ineffective one that everyone believes is working.
- Establish calibration traceable to national or international standards, or document the basis when no standard exists.
- Define accuracy and precision limits for each instrument and set intervals justified by risk and use, not habit.
- When equipment is found out of tolerance, document an impact assessment on all product accepted since the last good calibration.
- Record the identity of each calibrated device and the calibration results per §820.72(a).
- Calibration is outsourced and accepted on the vendor certificate without confirming traceability or limits.
- Out-of-tolerance findings are corrected quietly with no backward look at product already released.
How to close this finding — and stop it recurring
What this observation is associated with
Firms cited for this observation also appear in FDA compliance actions — warning letters, injunctions, or seizures — at 1.1× the rate of the average FDA-cited firm (8.5% vs 7.8%), across 188 FEI-matched firms.
This is an association, not causation, and reflects co-occurrence in FDA's records — not a claim that the observation caused or preceded the action.
How FDA worded it — real examples
Your process monitoring equipment is not calibrated to ensure that it reads accurately.
Your process monitoring equipment is not calibrated to ensure that it reads accurately.
Your process monitoring equipment is not calibrated to ensure that it reads accurately.
Your process monitoring equipment is not calibrated to ensure that it reads accurately.
Your process monitoring equipment is not calibrated to ensure that it reads accurately.
Your process monitoring equipment is not calibrated to ensure that it reads accurately.
Related SPEQ references
Where to read the requirements behind this observation area on SPEQ. These are SPEQ's topic and audience pages — not an FDA classification of the citation.
Examples are a representative sample of FDA's published inspection observations. Open the 483 observation explorer →