R versus SAS as analysis software for Clinical trials

R is a high-level, interpreted programming language-based software which is known for statistical analysis and graphical reporting. It allows data analysis, manipulation, graphical presentation in easy and effective way. The use of R increased exponentially over last decade considering its easy access and popularity. R (the successor of S) was created in 1992 by Ross Ihaka and Robert Gentleman at University of Auckland. Now it is developed by the R Development Core Team. SAS and R are different primarily considering data management, analysis with wide-ranging options available.


R is easily available open source software and SAS is licensed software with significant cost attached to it. Private sectors including enterprises prefer SAS considering long history of ease of doing businesses, evolved interfaces and extended functionalities over the time. Since long back, effective, standard and statistically strong quality codes are result of hours of paid resources. SAS offers great customer service and support. R may need time to conceivably match the level as there is no paid support for free version. However, as discussed in earlier sections, there are groups supporting R commercially, such as Revolution analytics, RStudio(commercial).


SAS operates at observation level whereas R offers vectors-based coding. SAS is generally not case sensitive whereas R is case sensitive. R has come a long way and analysts are aware of it. R is very suited for medical research and hence, many of the top pharmaceutical companies building strengths within the team.


SAS is a part of pharmaceutical industry since early ages i.e more than 30 years. Analysis done in past or codes prepared are compatible/readable in SAS only. It would take few years of dedicated resources to convert all those legacy codes to R. This calls for investment as well or else legacy data would not be available for further reference or reproducing results. This is also one of the reasons that R is not capturing the pharmaceutical market very fast.


Preparing datasets which are analysis friendly needs lots of peek into intermittent datasets to evaluate the programming ongoing for conversion. SAS has user friendly interface to inspect datasets stored in directory. Multiple commands for filtering datasets are available on the interface itself. On the contrary, it is slightly difficult to inspect datasets on R interface.

Sas and R

GCE is working on building a team which could support our client in switching their analysis related activities to R and make submissions possible. GCE is already working on creating tools which are primarily using R at the backend for statistical analysis.