- 1 Welcome
- 2 White Paper
- 3 Script Development Platform
- 4 Script Documentation
- 5 Validating Scripts
- 6 Mission
- 7 Leadership: Co-Leads
- 8 Conference Calls and Minutes
- 9 Approach
- 10 Charter
- 11 Scope
- 12 Draft Call for Scripts
- 13 FDA Disclaimer
- 14 Disclaimer for Others
This is the wiki home of the working group that is developing tools to support development and contributions of analysis scripts and programs.
Notice: Draft white paper on Tables, Flags and Listings associated with central tendency open for comments
Please provide comments in the Discussion Tab of the Working Group wiki by 14 December 2012. Please provide only content-related comments and suggested references, and not formatting, spelling, or typographical error comments. With each comment, please provide the document section number.
We realize the Discussion tab might not be an ideal method for providing and receiving comments. Alternative methods are being assessed and may be available for future reviews.
Script Development Platform
Working group 5 are pleased to announce that the initial platform for sharing and developing scripts (SAS, R, S+, STATA etc) is up and running at the following location:
Some team members have set up Tortoise SVN which provides a GUI for working with GoogleCode in a windows environment. The GoogleCode site is configured to use the MIT License.
Progress report on script sharing platform by Mike Carniello Platform for Standard Script Development: Progress as of September 18 2012
We are working on creating templates for specification and user guide documentation. Initial draft documents ready for review are here:
We are hoping to use the Good Programming Practice processes being developed by a seperate PhUSE group.
Here are the slides from the FDA workshop in march 2012 from subgroup 2, working on defining the process of creating and validating slides:
Similar slides were presented to PSDM (Pharmaceutical Statistics and Data Management), a Dutch network group of statisticians, clinical programmers and data managers working in and for the pharmaceutical industry. The aim of the PSDM is to facilitate professional conduct and development of its members and to provide a platform for the exchange of expertise and ideas by organizing scientific meetings, workshops and courses in the Netherlands. The slides were a personal reflection from Kevin Kane of his experiences at the conference:
The subgroup have started working on adding more details to the process of validating scripts. Here is the initial document for review:
This working group will identify potential standard scripts for data transformations and analyses across and/or within a therapeutic area and exploratory analysis. The goal will be to begin the process of standardizing analyses across the industry, but also include examples of what can be done with a standardized data set. This will be an interactive session using online wiki to create and add input that can be shared openly between industry and regulatory.
|Mary Nilsson||Eli Lillyemail@example.com|
|Sally Cassells||Next Stepfirstname.lastname@example.org|
|Hanming Tu||Octagon Research||HTu@OCTAGONRESEARCH.com|
|Joy Li||FDA - CDER||Joy.Li@fda.hhs.gov|
|Mat Soukup||FDA - CDERemail@example.com|
Steering Committee Liaison: Scott Getzin firstname.lastname@example.org
Conference Calls and Minutes
WG5 Subgroup1 F2F March 19-20 2012
File:WG5 subgroup 2 TC 7 Nov 12.docx
WG5 Subgroup3 F2F March 19-20
WG5 Subgroup3 TC June 7
Discussion Club: FDA/PhUSE CSS Working Group 5 Update Budapest 2012
Example Scripts: Current and Future State
Script Example - Posted: March 14, 2012:
Draft Call for Scripts
Here is draft text that is under review. It is planned to send this out to the statistical computing community before the March 2013 conference:
First Call for Scripts
The Computational Science Center of the FDA in combination with the Pharmaceutical Users Software Exchange (www.phuse.eu), have a new project to create an open source repository for scripts (tools, computer programs, resources) to help manage, summarise, explore, analyse and report clinical trial data. The scripts can be in any computer language (e.g. R/SAS/S+/STATA/C++/java).
We would like to invite you and your organisation to get involved in this project, by considering whether you could add any scripts to the repository. Ask yourself the following two questions:
- Do you or your organisation have existing scripts that could be shared with the statistical computing community, that could help increase the efficiency of clinical trials?
- Do you or your organisation have the time and/or resource to create new scripts?
It is hoped that in the future, the repository will be the home to an extensive variety of tools and resources that can be used in medical research. We are continuing to work to develop standards and processes relating to validation, so that scripts may be flagged as validated in the system. Additionally, we hope that the development of scripts may build on existing CDISC standards, and may also be used as a catalyst to standardise the way we report clinical trials data.
Once scripts are added into the system, they can be enhanced and improved by the statistical computing community. Issues with scripts can be tracked and fixed by the community of users. All scripts will be available publicly without any cost.
Finally, there are obvious advantage of having scripts that could be used both in the drug development process as well as by regulators in the review and approval process. We look forward to interacting with you in the script repository.
FDA comments are an informal communication and represent the individual's best judgment. These comments do not bind or obligate FDA. The contents of this wiki are from the individual contributors and do not necessarily reflect the view and/or policies of the Food and Drug Administration, the employers of the individuals involved or any of their staff.
Disclaimer for Others
Working drafts and comments provided as part of the FDA/PhUSE Working Group efforts are informal communications and represent the individual's best judgment. They do not necessarily reflect the view and/or policies of their company or institution, the employers of the individuals involved or any of their staff.
Note: Individuals participating in FDA/PhUSE Working Group discussions on the wiki are responsible for determining whether this disclaimer is sufficient for complying with any relevant procedures from their company or institution.
--- Last revision by Kevin Kane,01/16/2013