Statistical Analysis in biotech process validation

Hi All,

I would like to know your opinion regarding the statistical analysis for setting process validation acceptance criteria, used in bioprocessing.
I currently use mean + / - 3SD, or calculation of the tolerance interval (TI) with k+/- SD as an acceptance criteria (when N<15).
I would like to know:

  • What if my data are not normally distributed? I can´t apply these tools, so… What can I do? Is enough only meet the specification range?
  • It is valid to use ANOVA as a statistical tool to compare process parameters of n batches to demonstrate reproducibility?
    Can you recommend me another tools, widely used for validation?
    I would really appreciate your advice.

Thanks in advance

You need look PDA Technical report No: 59