« Using R as a scripting language with Rscript | Main | Study: Statisticians have awesome jobs »

January 14, 2009


Feed You can follow this conversation by subscribing to the comment feed for this post.

"Guidance for Industry - Computerized Systems Used in Clinical Investigations (2007)", the FDA document: more use could have been made of this. No matter which software you're talking about, this document states that its use should be validated by the end-user, and quotes examples of at least 10 SOPs that should be written. The need for large resources to do this is emphasized in the last slide of (I think) the Novartis link. End user validation is the guidance from the FDA, and I wonder what this means for small organisations, and for NHS statisticians ? Is R better than other software in facilitating this process ? There's the steep learning curve, and validation will often be done by someone who understands IQ/OQ/OP working with the statistician. Comments have been made that the documentation is....to the point...maybe part of that learning curve rather than helping to overcome it. The very thing that makes R so attractive to those who are able (to look at the code and change it to meet their needs) might ironically present a regulatory hurdle. You would not want ad hoc changes being made, and if changes are planned they would need change control (one of the SOPs listed by FDA). But I don't want to be accused of FUD: this can be controlled, allowing R to shine. My concerns have shifted away from R (barring yesterday's comment) to the situation overall for small-scale users of any software and the validity of their findings, given that the onus for obtaining and demonstrating accurate use of their software rests with them. Marc Schwartz contributed twice to the MedStats discussion. His last posting 26/03/2009 15:22 unfortunately has my name on the posting (I forwarded it on), but I recommend it to any statistician, and especially those in a small organisation. It can be found at
This (Analyzing Clinical Trial Data....) is a very useful article, and I recommend reading the links, which serve to support the enthusiastic message whilst honestly including the qualifications.

Your IQ/OQ/PQ criteria is very nice. In any company the documented should be maintained perfectly and It should not be shared with others. So you are doing everything perfectly including other methods you are following.

We are Speid & Associates partners with pharmaceutical companies to provide regulatory affairs
and drug development services to progress new chemical and other entities through the drug development process.


In the link to "Presentation from Frank Harrell", he says, "Classical software “validation” has little to do with quality, as most errors are committed when deriving variables, manipulating and analyzing data. Validation should be directed towards checking the analysis at hand." I hope this means that he is supporting the IQ/OQ/PQ approach. OQ and PQ address the issues he raises. I see software suppliers supporting their customers with IQ and OQ (so am somewhat puzzled by the Speid posting above) and end-users engaging in PQ ("checking the analysis at hand.") As the Novartis link says in the summary slide, this will involve considerable resources on behalf of the company doing the validation, usually one which expects FDA inspection. A software company will gain a considerable commercial advantage if it is known to be helpful in this respect.

hi friends,
I am new to R.I would like to know R-PLUS.Does any know where can I get the free training for R-PLUS.


There is no R-Plus. S-Plus (TIBCO Spotfire S+ now) is a commercial statistical programming package. You can find information about available training on Spotfire website: http://support.spotfire.com/training/pathways/s+pathway.asp . However it looks like new owner of S+ is more interested in Finance applications rather than in Life Science now.

My colleagues and I have just concluded one of India's most in-depth studies on "Clinical Trials in India". We have looked at various areas such as:
- Market Trends
- Growth Drivers
- Regulatory Bodies and Framework
- Major Players.
We interviewed over 200 individuals and firms to collect the data in what we believe is one of the most detailed study on the subject in India. If you are interested in a copy, you may email me at infoalcpl@gmail.com. It is a paid report.


The comments to this entry are closed.

R for the Enterprise

Got comments or suggestions for the blog editor?
Email David Smith.
Follow revodavid on Twitter Follow David on Twitter: @revodavid

Search Revolutions Blog