Pre-conference: Causal inference and simulation in public health: evaluating real-world policies with real-world data (ECON)

Tuesday 11 November, 13:30 – 17:00 Helsinki time
(Starting with network lunch at 12:30)


Organised by

EUPHA Public Health Economics section, Organisation for Economic Co-operation and Developement (OECD)

Introduction

Public health decision-making increasingly relies on robust evidence to predict the impacts of interventions before their implementation. Causal inference seeks to determine the effect of an intervention on an outcome by carefully addressing confounding, measurement, and selection biases. Since most public health data are observational—and thus non-randomized—strong assumptions must be articulated and justified to draw valid conclusions. Furthermore, simulation models allow researchers and policymakers to explore “what-if” scenarios under varying structural assumptions, generating forecasts of hypothetical policy impacts.

This pre-conference will dive into the causal inference concept, methods and recent examples. You will learn their potential and limitations in the context of public health research. Through methodological sessions and applied case studies, participants will gain insight into how to critically design, assess, and communicate evidence to support health system decisions.

Objectives

By the end of this workshop, participants will be able to:

  • Understand the foundational frameworks and assumptions underpinning causal inference in public health.
  • Construct and interpret causal diagrams (DAGs) to guide study design and analysis.
  • Explain a variety of causal inference methods (e.g., regression adjustment, instrumental variables, regression discontinuity, difference-in-differences, mediation).

Programme

13:30 Introduction to session and topics - João Vasco Santos 
13:45 Frameworks in causal inference & development of direct acyclic diagram - Eduardo Barcho Montes de Oca 
14:30 G computation - Vanessa Gorasso 
14:45 Mediation analysis - Eduardo Barcho Montes de Oca
15:00 Coffee break
15:30 Design based approaches (DiD, regression discontinuity and instrumental variables) - Karl Emmert-Fees
16:30 Synergies between quasi-experimental research and simulation methods - Michele Cecchini
16:45 Closing

Registration

The registration fee is EUR 80. Networking lunch and refreshments are included.