1994 Annual Meeting of the SSC in Banff

BIOSTATISTICS WORKSHOP
Meta-Analysis
May 8, 1994
The Banff Conference Centre, Banff, Alberta

 

Professors Andy Oxman and Jesse Berlin will present a one day workshop the Sunday prior to the Annual Meeting of the Statistical Society of Canada, at the Banff Conference Centre, Banff, Alberta (Monday-Wednesday, May 9-11, 1994). The full day session will consist of a short course on the methods of meta-analysis.

Outline

In the first half of the workshop, the rationale and a general framework for meta-analysis will be presented. Key issues regarding meta-analytic methods will be addressed with respect to problem formulation, identification and selection of studies, validation of studies, data collection, analysis and inferences. The emphasis will be on protecting against bias. A small group, problem solving approach will be used, drawing on specific examples to illustrate important methodological issues. Particular attention will be given to the methods being used within the Cochrane Collaboration to prepare and maintain a database of up-to-date systematic reviews of randomized controlled trials of the effects of health care.

The second half of the workshop, using a similar format, will focus on the statistical methods generally applied in meta-analysis. Methods for both dichotomous and continuous outcomes will be addressed, with an emphasis on combining odds ratios, relative risks, risk differences, and means. Use of meta-analytic regressions will also be addressed, as will the meta-analysis of epidemiologic dose-response data. Several of the remaining questions in statistical methodology will be presented briefly, including the question of fixed versus random-effects, the analysis of rare events, and the use of cumulative meta-analysis.

About the Instructors

Dr. Andy Oxman is from the Departments of Clinical Epidemiology & Biostatistics and Family Medicine at McMaster University. He is the editor of the Cochrane Collaboration's 'Tool Kit' which includes guidelines and software to systematise and facilitate the preparation and updating of systematic reviews.

Dr. Jesse A. Berlin is from the Center for Clinical Epidemiology and Biostatistics at the University of Pennsylvania School of Medicine, where he is serving as the Interim Director of the Biostatistics Unit. He has published several methodologic papers in meta-analysis and has participated in numerous meta-analyses of clinical and epidemiologic studies.

Schedule

Following registration (8:30-9:30), the course will run from 9:00 to 5:00 with breaks at 10:30-10:45, 12:15-1:20, and 3:00-3:15.

Registration Fees (Before March 15)*
SSC member $80
non-member $100
student $30

* An additional conference center daily user fee of $21.40 to cover parking, refreshment breaks and lunch is payable by those not staying at the Banff Conference Centre. These services are included in the cost of accommodation for those staying at the Centre.

To register for this workshop, please use the registration form for the SSC Annual Meeting (whether or not you plan to attend the meeting). For further information on the SSC Annual Meeting, see this issue of Liason.

Sponsored by the Biostatistics Section of the Statistical Society of Canada