- Posted By: Susan Dutton
The bi-annual meeting took place at the Sheffield Clinical Trials Unit and brought together statisticians from many of the registered trials units. This was a great opportunity to receive information about current research projects being undertaken by the group and associated members.
Current ongoing projects
* Guidelines on Statistical Analysis Plans – this is currently in the final stages and the results will hopefully be piloted in mid-November with publication expected early next year.
* Statistician Specific Good Clinical Practice Training – a survey was completed at the meeting and the need and potential content of a course will be decided after review.
* Statistical Expertise for DSMC and TSC – This project is currently being tested and will go live before the end of the year. It will provide a portal for trials to find statisticians with relevant experience and availability to sit on these committees and also for more junior statisticians to take on roles with support, or as observers so that we can develop expertise for future trials.
The morning session discussed adaptive designs, difficulties, perceived or actual of getting these funded and elaborating that there is help from the Adaptive Design Subgroup of the methodology Hub Network with planning and developing adaptive designs for funding. Email Thomas Jaki - t.jaki@lancaster.ac.uk or Philip Pallmann - p.pallmann@lancaster.ac.uk for further details.
The afternoon session discussed progression rules for internal pilot studies monitoring recruitment rates and designing and reporting internal and external pilot studies. A new CONSORT extension will be available soon for reporting external feasibility and pilot studies and a new journal has been launched: http://www.pilotfeasibilitystudies.com where these can be submitted.
Lisa Hampson talked about her research into developing some statistical techniques for choosing the progression rules for internal pilot studies. These Internal Pilot Stop/Go rules are often a requirement of the funding body. Lisa’s work is based on a 2-stage progression pre-specified progression rules developed using optimal characteristics and is based on the Bayesian approach to accommodate uncertainty in the recruitment parameters. Lisa’s work is currently being written up for publication and is an excellent first step in formalising the current ad-hoc choice of these criteria.