Role of variability in design, implementation, and interpretation of microbial inactivation studies

In developing a Ct (concentration × time) table for the inactivation of Cryptosporidium with ozone it has been necessary to incorporate an uncertainty factor that can be applied to calculated Ct values to take account of the significant variability in inactivation results. This paper presents an approach for partitioning the overall variability into two types. That due to analytical and experimental error is shown to clearly outweigh that due to differences in water matrices and oocyst resistance to disinfectant. The results support the use of confidence intervals calculated about the mean inactivation coefficients calculated with existing data, leading to the operation of ozonation with sufficient safety factors to protect public health.