Medical statistics plays an essential role in all stages of a quantitative health care research project from design through to analysis and interpretation. This intensive course covers the essential principles and methods required. Emphasis is on study design, appropriate analysis, and interpretation of results. The underlying concepts of statistical analysis as well as basic and some more advanced analysis techniques are covered. Sessions include lectures and practical work, both computer based (using Stata) and using small workshops for discussion. The course has been running for more than 30 years and has earned an international reputation.

Please note that the Early Bird Rates are only available until the 30th April 2027.

Abstract

Applied health and population health researchers are often asked to answer causal questions using empirical data: does an exposure affect later health, does a treatment work outside a trial, or does a policy or service change improve outcomes? These questions are difficult to answer using observational data alone, because associations may reflect confounding, selection bias, missing data, measurement error or inappropriate adjustment.

This three-day in-person course provides practical training in causal inference for researchers working with cohort studies, electronic health records, routinely collected data, trials with observational follow-up, administrative data and other population health datasets. The course introduces key causal inference frameworks and methods, including potential outcomes, directed acyclic graphs, target trial emulation, propensity score methods, regression adjustment, instrumental variable analysis and sensitivity analyses.

Each approach will be explained using applied examples, with emphasis on defining clear causal questions, making assumptions explicit, choosing appropriate analytical strategies, assessing robustness and interpreting results cautiously. Computer practicals will be provided in both R and Stata and will focus on implementing methods in realistic applied research settings.