PhUSE Single Day Event in Boston: Short on time, long on content
By David Handelsman
Clinical research software was the hot topic at the PhUSE Single Day Event in Boston. SAS has a long history of sponsoring PhUSE (Pharmaceutical Users of Software Exchange), including the inaugural one-day event in Boston this year as well as the multi-day event held in Berlin.
The presentations and panels for the event focused on three key themes related to clinical research software: community-based software, standards and statistical computing environments. Speakers included biopharmaceutical companies, vendors and the FDA. For a short day, there was a lot of great content.
Here are some highlights:
The principle behind this approach is that the community as a whole is providing the knowledge and expertise to continue enhancing the toolset. As one of the panel speakers cleverly stated, this approach is not about being free, but about freedom to enhance and modify the software. Open-source still carries significant costs beyond the initial software download, and should never be considered "free."
Despite being a commercial company throughout its history, SAS has adopted a community-based approach for years. This approach, which can be traced back to the earliest SAS-L listserv discussions, is now being extended to public repositories of analysis scripts – including SAS programs – that are being prototyped by both PhUSE and the CTSA.
These SAS programs are based upon a community engagement model whereby industry experts develop, apply and refine specific SAS programming utilities (for example, to create a SAS-based Trellis plot). While the repositories are currently small, and longer-term it may make more sense to have a single harmonized repository, the goals are logical: clearly documented, validated standardized programming tools.
Standards continue to play a key role in emerging pharmaceutical software business users. The conversations have clearly cycled from “why?” to “what?” and from “how?” to “what next?”.
Most companies have active CDISC SDTM projects underway, and more and more these companies are looking for ways to push standards earlier into their business processes. Not coincidentally, community-sourced analysis scripts, such as those described above, will operate best and most successfully against standardized data.
Statistical computing environments
Every biopharmaceutical research company follows the same basic process in developing, testing and applying SAS programs to determine the safety and efficacy of a clinical trial.
Although the overall process may be similar, each company typically develops a highly specialized business process to manage the workflow and validation constraints for this process.
In recent years, the concept of a packaged “statistical computing environment” has emerged and matured. At the PhUSE meeting, there were several presentations regarding the concept of a statistical computing environment, and it’s clear that this is an area of increasing interest.
The SAS Drug Development solution was being used in production before the term "statistical computing environment" existed, and we're excited to continue to see our number of production customers grow around the world.
David Handelsman is an Advanced Analytics Strategist for SAS Health and Life Sciences.