Generalized methodology for evaluation of parenteral inspection procedures.

Abstract

A procedure to permit objective comparison and thus validation of alternate particulate inspection techniques has been developed and is described. This procedure is based upon the finding that particulate inspection methodologies, human and robotic, are probabilistic rather than deterministic in nature. This is due in major part to the size and contrast of the particulate contamination encountered. Two criteria are utilized for particulate inspection evaluation: (a) the efficiency of rejection of the containers to satisfy CMP requirements: and (b) the economic effectiveness of the inspection in terms of labor and falsely rejected good material. The effect of a probabilistic inspection on the nominal batch quality and the repticability of the quality assurance sampling inspection are also discussed. A tentative conclusion is reached that a deterministic sampling inspection is unduly responsive to minimal particulates and should be more sensitive to the low concentrations of major particulates encountered. A self-contained step-by-step guide to the use of the described procedure for validation of new techniques or instruments is included in an Appendix.