SWE-REG: Statistical Methods for Register-based Research

High-quality registers alone are not enough to advance medical science. Statistical methods that allow for the best use of registry data are also needed.

SWE-REG is short for the Swedish network for registry-based research. This is a research network funded by the Swedish Research Council to create an environment for research on the development and evaluation of statistical methods for register-based research.

Our research is in four primary areas:

  1. Extend methods for survival and multi-state models to use rich register data better when modelling complex time-to-event outcomes;
  2. Develop methods for causal inference from register data in the presence of unmeasured confounding;
  3. Adapt machine learning methods to survival outcomes for register data
  4. Develop and apply statistical methods for the analysis of ‘big’ register data

SWE-REG is short for the Swedish network for registry-based research. If you are interested in contributing to our research or believe that our methods may be of use in your application area, please contact us.

Contact person

MEB Partners

Erin Gabriel

Affiliated to research

Arvid Sjölander

Principal researcher

Mark Clements

Lecturer senior

Michael Sachs


National and International Partners

Els Goetghebeur

Sven Ove Samuelsen

Paul Lambert

Current Ph.D. students and Postdocs

Adam Brand

PhD student

Gustav Jonzon

PhD student


Causal bounds for outcome-dependent sampling in observational studies

Journal of the American Statistical Association

Gabriel EE, Sachs MC, Sjölander A.

Content reviewer: