A deviation is an event. The batch number on the form is where someone noticed it. The single most common impact assessment failure is treating the two as the same thing: the event affected a room, a piece of equipment, a material lot or a shift, and the assessment only looked at one batch.
The scoping questions
Before assessing impact, define the scope. These questions should be mandatory fields, and 'none' should require a rationale.
- Which other batches were exposed to the same condition? (Same room, same equipment, same material lot, same time window, same operator, same method.)
- Which other products? Multi-product equipment and shared utilities spread impact across the portfolio.
- Which equipment or systems? An out-of-calibration instrument affects everything it measured since the last good calibration.
- Which records? If the event has a data integrity dimension, which other records were produced under the same conditions?
- Which validated states? Does the event challenge a process, cleaning or computerised system validation?
- Which regulatory commitments? Registered parameters, stability commitments, agreed CAPA from a previous inspection.
Product and patient impact
Impact assessment answers one question with evidence: could this event have affected the quality of product that could reach a patient? Structure it as a chain of reasoning from the event, through the mechanism by which it could affect a quality attribute, to the evidence that it did or did not.
Batch disposition
The disposition decision (release, reject, rework, hold pending further data) belongs to the Qualified Person or the quality unit, and the record must show the basis for it. An assessment that concludes 'no impact' without saying what evidence supports the conclusion is a finding under Chapter 1 and Annex 16 regardless of whether the batch was in fact fine.
Module 5 of the course provides an impact assessment template built around these questions and walks through a multi-batch, multi-product example end to end.