Drug Product Stability Trending – Sample Size for Confidence Limits?
January 30, 2020 at 4:31 pm #245877
GoldchemistParticipant@KevinMurray Include @KevinMurray in your post and this person will
be notified via email.
I am currently using ICH guideline Q1A_R2 as a guideline for trending stability data. It is recommended to determine the time at which the 95% one-sided confidence limit for the mean curve intersects the acceptance criterion. If the confidence limit intersects the acceptance criterion prior to the proposed expiration date….I open an out of trend investigation. I consistently open investigations for stability studies with only 6 months of data (3 time points) at the licensed storage condition. I understand that, by means of the confidence limit formula, larger sample data sets will provide a more accurate assessment of the population parameter because there is less standard error due to the inverse square root relationship between confidence limits and sample sizes. More data equals more confidence. By means of a risk assessment I’m sure I can justify waiting until 15 months to check for statistical trends.
My question is….Is there a minimum sample size recommended for calculating the confidence limits for stability trending purposes? Does anyone know of a specific reference that recommends waiting until I have say 15 months of stability data (0,3,6,9,12,15 months)?
Any feedback is very much appreciated.
You must be logged in to reply to this topic.