Course info

This is course is co-organized with




Instructor: Miguel Hernán (Harvard University)
Dates: 25-29 August 2025
Hours: 9:30 to 13:00 CEST
Format: In person

Intended for

Researchers who work with large databases of individual-level data, health policy practitioners.

Prerequisites

Working knowledge of study design and regression analysis, interest in evaluation of policy interventions for public health and medicine.

Overview

The course introduces a general-purpose causal inference framework that integrates methods for both experimental and non-experimental data. The framework has two steps: 1) specification of the (hypothetical) target experiment or trial that would answer the causal question of interest, and 2) emulation of the target trial using the available data. The course explores key challenges for target trial emulation and critically reviews methods proposed to overcome those challenges. The methods are presented in the context of the evaluation of the comparative effectiveness of health interventions using existing databases of administrative and clinical data. At the end of the course students should be able to:

• Formulate sufficiently well-defined causal questions
• Specify the protocol of the target trial
• Design analyses of observational data that emulate the target trial
• Identify key assumptions for a correct emulation of the target trial

Topics

• Causal inference as a key component of decision making
• Target trial emulation as a unifying concept for causal inference
• Target trial emulation to avoid self-inflicted biases in causal inference
• Point interventions vs. sustained policies
• G-methods to evaluate sustained policies