SSC 1996 Annual Meeting

June 2 - 5, 1996
University of Waterloo, Waterloo, Ontario

Biostatistical Issues in Pharmacoepidemiology

Organizer and Chair:
Charles H. Goldsmith, McMaster University

Speakers:
Michael Edwardes*, McGill University & Samy Suissa, McGill University
Sampling Within Cohorts to Assess Risk

Richard J. Cook, University of Waterloo
Monitoring Multiple Outcomes in Randomized Trials

Andrew R. Willan*, McMaster University, Bernie J. O'Brien, McMaster University & Deborah J. Cook, McMaster University
Benefit-Risk Ratios in the Assessment of the Clinical Evidence of a New Therapy

Biostatistical Applications of Neural Nets

Organizer and Chair:
Rollin F. Brant, University of Calgary

Speakers:
Tony Plate*, British Columbia Cancer Agency and University of British Columbia, Pierre Band, British Columbia Cancer Agency, Joel Bert, University of British Columbia & John Grace, University of British Columbia
Neural-Network Models of the Relationship Between Tobacco and Alcohol Consumption and Cancer

Jack V. Tu, Harvard Medical School and University of Toronto
Advantages and Disadvantages of Using Artificial Neural Networks versus Logistic Regression for Predicting Outcomes

Penny M. A. Brasher*, Alberta Cancer Board & Maire A. Duggan, Foothills Hospital
PAPNET: An Application of Neural Networks in Cervical Cancer Screening

Discussant:
Rob Tibshirani, University of Toronto

Statistics in Resource Management Controversies

Organizer and Chair:
Rick Routledge, Simon Fraser University

Speakers:
Don Russell, Canadian Wildlife Service
Oil and the Porcupine Caribou Herd - Do Data Really Matter?

Valerie LeMay, University of British Columbia
Forest Land Inventory - Estimation under Constant Change

Rick Routledge, Simon Fraser University
Fishy Statistics: The Case of the Missing Salmon

Estimating Functions in Biostatistics

Organizer:
Mary L. Lesperance, University of Victoria

Chair:
John Neuhaus, University of California, San Francisco

Speakers:
Charmaine Dean*, Simon Fraser University & Ying C. MacNab, Ministry of Health, British Columbia
Empirical Bayes Methods for Local Smooth Mapping of Infant Mortality Rates in British Columbia

Jerry Lawless, University of Waterloo
Estimating Functions for Longitudinal Studies

Biostatistics in Epidemiology, Dealing with Interaction and Confounding

Organizer:
Karen Campbell, University of Western Ontario

Chair:
Neil Klar, University of Western Ontario

Speakers:
W. Douglas Thompson, University of Southern Maine
Conceptual and Practical Issues in the Epidemiologic Evaluation of Interactions

John Neuhaus, University of California, San Francisco
Bias and Efficiency Loss Due to Omitted Covariates in General Linear Models

Jack Siemiatycki, Institut Armand-Frappier
An Empirical Evaluation of the Magnitude of Confounding Bias