Out-of-Specification (OOS) Investigations Training

About this course

Why OOS investigations matter to your site

An out-of-specification result is one of the first records an inspector asks to see, because it shows how a laboratory and its quality unit behave when a number goes wrong. In FDA's fiscal year 2025 drug inspection data, 21 CFR 211.192, the duty to investigate thoroughly, was the second most cited line, with 164 citations. Recent warning letters show the cost of a weak file: in February 2026 FDA wrote to a maker whose batch was released and shipped to the US after a retest within limits, with the investigation started late and no root cause found. One late investigation became a request for an independent review of past invalidated results. For the analyst, the supervisor, the QA reviewer and the site, the quality of each OOS file is a direct measure of the quality system.

What makes this course different

The course follows the structure of FDA's current OOS guidance, Revision 1 of May 2022, step by step, and cites the EU GMP Chapter 6 points in one line each where they add something, such as the procedure for out-of-trend results and the reporting of confirmed stability failures. It does not stop at the rules. Each module applies the method to one company, JKL Pharma, and follows a single failing assay from the bench to the batch decision, the field alert question, the root cause and an effective CAPA, with the documents filled in. Every module also sets a common mistake against what good looks like: trial injections, retests repeated until a number fits, averages that hide variation, and inconclusive investigations treated as clean. Real 2025 and 2026 FDA warning letters show how inspectors judge each of these, and four practice exercises with model answers let you apply the method to JKL or to your own laboratory.

Who should take it

The course is written for everyone who touches an OOS result: QC analysts who take the first actions at the bench, laboratory supervisors who run the first assessment, QA reviewers and investigators who lead the full-scale investigation, and the production and technical staff who join it. It also suits regulatory affairs staff who may need to tell FDA about a failing batch already on the market within three working days, and contract laboratories, whose responsibility FDA's guidance treats as equivalent to a manufacturer's. Sites that supply both the US and the EU will find one method that serves both markets, with the different reporting rules named clearly. The course also explains the 1993 US court decision that shaped FDA's position on outlier tests, so that the reasoning behind the rule is clear.

What you can do afterwards

After the course, you can keep the evidence in the first hour, test each hypothesis on the material kept at the bench, and set aside a result only on a cause shown by evidence. You can plan retests in advance with a second analyst, report every individual result, and use averages and outlier tests only where the guidance allows. You can extend a confirmed failure to the other batches that may share its cause, decide when a field alert is due, and close a CAPA only when its effectiveness check is met. The scenario-based final exam leads to the Sterling Academy Certificate of Competence for those who score 60 percent or more, with one retake included. It is a training certificate, not an approval by FDA or any EU authority.