[Systat, NCSS, MedCalc and Prism] Software validation strategy in regulated area

Dear All,

I wonder if anyone can help me please with what may sound an odd request? I work in an area regulated by Good Laboratory Practice and use the above software packages for some data analysis. My Quality Assurance people keep going on about having these programmes validated, which is fine but it's a bit of a "how long is a piece of string?" issue.

Is there anyone who has been in a similar position who would be willing to share how they got around this? I apologise if this has been discussed before, but I couldn't find anything similar. The obvious way round this is to cull examples from textbooks and show that the answers are the same (within reason), but a pre-defined test set would be better.

Many thanks in advance.