- From: Forsberg, Kerstin L <Kerstin.L.Forsberg@astrazeneca.com>
- Date: Sun, 11 Feb 2007 23:22:33 +0100
- To: <public-hcls-dse@w3.org>
Hi Stephen and Eric, Stephen's example excel-files will give us a great start for a test bed of actual, but faked, clinical data. One thing I think is important to recognise is that also the variables in the demog file as sex, race, age, weight, unitbeer etc. are basically also observation results/values. In the same way as it is possible to add qualifiers to blood pressure results/values and to adverse events text/codes, it must be possible to add qualifiers also for observations recorded as variables in a file of demographic data. Furthermore, what is to be considered as a subject characteristics can differ from different studies, diseases and projects. In most cases sex and age are considered to be non time dependent during a study and therefore treated as fixed subject characteristics obsererved only once. While in other cases other types of observations results/values, or aggregations of such, such as weight, had-a-SAE, unitbeer, mean-weight etc. can be interesting subject characteristics. The above discussion is also related to the unfortunate implementation of the DM domain in SDTM with fixed horizontal variables for e.g. sex and race. I look forward to how the N3 implementation of this would look like. Cheers Kerstin
Received on Sunday, 11 February 2007 22:23:13 UTC