A Novel Approach to the Statistical Evaluation of Media Fill Tests by the Difference from No Contamination Data

Abstract

It is obvious that, in the media fill test and process simulation test, positive numbers in total fills should not have any significant difference from zero or asepsis. There are many reports concerning the definition of “sterility” or “asepsis.” However, any scientific and practical methods to demonstrate “no significant difference from zero” have not been reported up to now. The existing criteria, such as “less than 0.1%,” “less than 0.05%,” and “less than two positives” are not appropriate to assure the integrity of processes, and sometimes lead to erroneous results. The purpose of this report is to demonstrate novel, reasonable and practical methods and criteria based on scientific and statistical consideration.

According to the ISO 13408-1 Aseptic Processing of Health Care Products, Part 1 (1998), General Requirement for Aseptic Processing, the action level for the number of positive units in media fill tests is specified as 0.1%, and the alert level is 0.05%. In this paper it is shown that the existing ISO standard and other official methods are inappropriate in that zero contaminated units (sterile product) is outside the confidence range of probable distribution of contaminated units, even though the contaminated units are less than 0.1% in larger numbers of fills, and even less than 0.05%. This indicates that the limit of 0.1% or 0.05% is inappropriate in cases of larger numbers of fills. For sterile products, the number of contaminated units other than “zero” at the statistical confidence range must be judged to be contaminated units in process and as non-sterile. In order to harmonize this criteria—“no significant difference from zero”—with the existing criteria, the new criteria may be combined with only the existing criteria of 0.05% in smaller number of fills.