Causal research in modern methodological framework

Programme

Day 1, Monday 15.9, Place: P117
9.00-10.20 Lecture I: Causality and causal questions in health sciences. Potential outcomes and counterfactuals. Elements and rules of causal diagrams.
10.20-10.40 Break
10.40-12.10 Practical session I: Critical appraisal of research questions and answers on interesting causal effects in selected published articles.
12.10-13.15 Lunch break
13.15-15.00 Lecture II: Causal estimands and their identification. Confounding and its control. Standardization or g-formula.

Day 2, Tuesday 16.9, Place: 289B
9.00-10.20 Lecture III: Causal estimands (continued). Inverse probability weighting. Time-to-event outcomes. Types of experimental and observational study designs. Target population and study population, issues of selection.
10.20-10.40 Break
10.40-12.10 Practical session II: Analysis of causal diagrams and identification of causal effects.
12.10-13.15 Lunch break
13.15-15.00 Lecture IV: Randomized trials, observational cohort studies, and target trial emulation (TTE).

Break of two weeks

Day 3: Monday 29.9, Place: P117
09.00-10.20 Lecture V: Recap of causal diagrams and study designs. Introduction to variants of case-control studies.
10.20-10.40 Break
10.40-12.10 Practical session III: Critical appraisal of published cohort studies and TTE studies
12.10-13.15 Lunch break
13.15-15.00 Lecture VI: Identifying causal estimands in case-control studies. Selection bias and information bias in causal diagrams.

Day 4, Tuesday 30.9, Place: 286A-B
09.00-10.20 Lecture VII: Introduction to sensitivity and bias analyses. Key issues in statistical inference on causal estimands.
10.20-10.40 Break
10.40-12.10 Practical session IV: Critical appraisal of published case-control studies
12.10-13.15 Lunch break
13.15-15.00 Lecture VIII: Statistical modelling and analysis of causal estimands from empirical data.

After successful completion of the course, the student is able to define the basic concepts and rules of causal diagrams, how to specify precisely causal effect of a single time-fixed exposure and identify it using causal diagrams in the presence confounding. The student can also characterize the major types of bias with the help of these diagrams, the main features of the common observational study designs used in causal research and the key principles of statistical analysis and inference on causal estimands.

Basics in statistical methods. Some knowledge on linear, logistic and Cox regression is useful.

Course leader: Professor Jouko Miettunen, Research Unit of Population Health (jouko.miettunen@oulu.fi)
Teacher: Emeritus professor Esa Läärä, Research Unit of Mathematical Sciences (esa.laara@oulu.fi)

Recommended reading before the course:

Digitale JC et al. Tutorial on directed acyclic graphs. J Clin Epidemiol 2022;142:264-267. https://doi.org/10.1016/j.jclinepi.2021.08.001,

plus at least one of the following:

Igelström P et al. Causal inference and effect estimation using observational data. JECH 2022;76:960-966. https://doi.org/10.1136/jech-2022-219267

van Amsterdam WAC et al. Causal inference in oncology: Why, what, how, and when. Clin Oncol 2024 (in press). https://doi.org/10.1016/j.clon.2024.07.002

Akingubke AA et al. Directed Acyclic Graphs for Oral Disease Research. J Dent Res 2016, Vol. 95(8) 853–859. https://doi.org/10.1177/0022034516639920

Requirements: Active participation in lectures, and practical sessions. Each day the participants are given homework exercises mostly based on selected articles.

Evaluation: Pass/Fail.