How do prospective, concurrent and retrospective validation differ?

Why

Prospective validation is the default. Annex 15 says processes using the traditional approach should undergo prospective validation, wherever possible, before the product is certified. The FDA's 2011 guidance similarly expects PPQ to be completed successfully before commercial distribution.

Concurrent validation means routine production starts before the validation program is complete. Annex 15 permits it only in exceptional circumstances with a strong benefit-risk ratio for patients, such as a medicine in short supply. The decision must be justified, recorded in the validation master plan and approved, and each batch must have enough data to show it is uniform and meets criteria before the QP certifies it.

The FDA guidance calls this concurrent release of PPQ batches and gives examples such as infrequently made products or very short-lived products, again with careful oversight. Retrospective validation used historical production data to argue that a process already in use was validated. Annex 15 states plainly that it is no longer an acceptable approach, because past data were not collected under a plan designed to detect variability.

For example, a legacy product with ten years of passing batches still needs prospective work if it moves site. Historical data remain useful for setting the validation strategy and for the hybrid approach, but not as the validation itself.