Development of a Statistical Approach for Microbial Monitoring in Non-sterile Pharmaceutical Environments

Little attention has generally been given to non-sterile pharmaceutical manufacturing, because these products represent less microbial contamination risks compared to parenterals. Nevertheless, an extensive risk management program is required to prevent risk to patients and incidence of complaints or product recalls. Few scientific studies discuss on non-sterile production environment, which led to poor references for unreliable statistical approaches. The aim of this study was therefore to develop a statistical approach for microbial monitoring and limit calculation to be applied on non-sterile pharmaceutical industries, assessing two manufacturing facilities environments located in intertropical zones on different hemispheres. The study evaluated a 2-year data of environmental monitoring active and passive air and surface microbiological recovery for both facilities using JMP™ as a statistical tool. Process and non-process rooms were evaluated to propose the statistical approach for monitoring and limit calculation. Fit tests showed that microbiological environmental monitoring results for both facilities followed Gamma-Poisson distribution. Capability analyses performed against regulatory limits show that all data sub-groups were in control (Cpk > 1.3). Alert and action limits were proposed according to the actual distribution curve. The study succeeded in developing a step-by-step statistical approach for assessing the EM program, which will support most non-sterile industries. The establishment of adequate limits is essential for false alarm prevention and for cost reduction with overreactions to in-control results. This study can also be extended to other process parameter control with similar distribution, expanding the applicability.