History of the Cochrane Statistical Methods Group
The Statistical Methods Group (SMG), (Cochrane Methods Statistics) was created following a meeting at the UK Cochrane Centre in Oxford in the summer of 1993. It merged with the Reporting and Quality Assessment of RCT’s Methods Group in 1997 and retained the name SMG. Membership has been steadily increasing over the years from a membership of 80 in 2001 to 275 as of May 2016 from 33 countries.
The group has evolved in response to the demands for methods to resolve key statistical issues in review methodology. In 2000, for example nine topic action groups were formed to address issues such as Bayesian methods of meta-analysis, meta-analysis of continuous data, computing study weights in meta-analysis, and heterogeneity, subgroups and meta-regression. Special meetings have been organised to cover specialised statistical concepts and methods, e.g. analysis of ordinal data and analysis of cluster randomized trials. Advanced statistical training has also been supported by Cochrane to increase statistical expertise within the Cochrane Review Groups. Over the years other groups have evolved to specialise in particular aspects of methodology such as bias, individual participant data and prospective meta-analysis.
The SMG is a forum where all statistical issues related to the work of Cochrane are discussed. Membership of the SMG is currently through membership of the group's email discussion list. The list is used for discussing all issues of importance for the group, whether research, training, software or administration. All statisticians working with Collaborative Review Groups (CRGs) are strongly encouraged to join the SMG. Communications within the SMG are mainly via the email discussion list (details available here). The SMG meets at each Cochrane Colloquium and occasionally at other times.