You c
an now register for this event. Registration fees are as follows:
\n-
\;Members of PSI \;= Free of charge
\n- \;Non-Members of
PSI \;= Free of charge
\nTo register for the session\, please&nb
sp\;
click here.
In the pharmaceutical industry\, all analytica
l methods must be shown to deliver unbiased and precise results. In an ass
ay qualification or validation study\, the trueness\, accuracy and interme
diate precision are usually assessed by comparing the measured concentrati
ons to their nominal levels. Trueness is assessed by using confidence inte
rvals of mean measured concentration\, accuracy by prediction intervals fo
r a future measured concentration\, and the intermediate precision by the
total variance.
\nICH and USP guidelines alike request that all relev
ant sources of variability must be studied\, e.g. the effect of different
technicians\, the day-to-day variability or the use of multiple reagent lo
ts. Those different random effects must be modeled as crossed\, nested or
a combination of both.
\n
\nConfidence\, prediction and toleranc
e intervals in linear mixed models will be detailed with a focus on the in
terpretation of statistical results. Their relationships will be discussed
together with the POOS (out-of-specification probability). Two real datas
ets from assay validation study during vaccine development are used to ill
ustrate the statistical intervals and the POOS.
\n
Speaker \n | \n Biography \n |
\n
| \n Bernard G Francq is Lead Statistic ian with GSK Biologicals\, driving statistical innovation for CMC projects worldwide. He holds a PhD in Statistics (UCLouvain\, 2013). \nHis work on errors-in-variables (EIV) regressions in method compari son studies has been awarded Best MSc Thesis Biostatistics (Qué\;tel et 2008\, Belgium)\, Best Chemometrician Prize (Chimiomé\;trie 2009\ , Paris) and Best Young Researcher (Agrostat 2012\, Paris). His communicat ion skills have been recognized with the Greenfield Challenge Award (ENBIS 2012\, Ljubljana). Recent work on tolerance intervals in bridging studies was awarded Best GSK Statistical Paper (2020). He lectures at UCLouvain a nd regularly offers trainings to statisticians in the (bio)pharmaceutical industry. \n |