Biotech Values

[Biowatch]

>>ex post facto analyses<<

I think there was concern about "subgroup analyses in clinical trials" at the ODAC meeting on RSR 13. I did not see it, but there was a presentation by this title there, given by Stephen George, PhD, Director of Biostatistics at Duke.

The last slide concluded:

>>
- Pre-planning is key
- Larger studies required for proper subgroup analyses
- Exploratory analyses are good for hypothesis generating but are not convincing alone
- More than one study important for validation
<<

------
Searching on "George SL" at PubMed turns up some interesting articles,
http://www.ncbi.nlm.nih.gov/entrez/query.fcgi

including articles on accelerated approval and whether statisticians on data monitoring committees should be independent of clinical trial sponsors.

http://jncicancerspectrum.oupjournals.org/cgi/content/full/jnci;95/18/1351