May 13, 2025
Explore causal inference methods and target trial emulation to enhance the design and analysis of observational studies and clinical trials in regulatory, payer, and HTA decision-making.
This course introduces the principles of causation in comparative effectiveness research, covering the use of causal diagrams (directed acyclic graphs; DAGs) to avoid common biases such as time-zero bias and immortal time bias. Participants will explore the "target trial" concept and apply a counterfactual approach using "replicates" to analyze big real-world datasets.
Technical Topics Include:
- Causal principles & causal diagrams – Understanding DAGs to mitigate common biases
- Target trial emulation – Designing observational studies that mimic clinical trials
- Causal methods for baseline confounding – Multivariate regression, propensity scores
- Time-varying confounding methods:
- G-formula
- Marginal structural models (MSMs) with inverse probability of treatment weighting (IPTW)
- Structural nested models with g-estimation
- Estimands for observational and clinical trial data – Addressing decision problems in the presence of treatment switching
- Applied case studies in health technology assessment (HTA):
- Single-arm trials with external control arms
- Trials affected by treatment switching
- HTA agency perspective: Acceptance and barriers in adopting causal inference methods
This Course Includes Practical Tools and Concepts That Can Be Immediately Applied, Including:
- Step-by-step guidance on causal modeling for confounding adjustment
- Best practices for applying target trial emulation to reduce self-inflicted biases
- Practical use of causal inference methods in real-world evidence (RWE) and clinical trial analyses
- Recommendations for integrating causal inference into HTA decision frameworks
The course consists of lectures, published case examples, and interactive discussions. The intended audience includes researchers from all substance matter fields, statisticians, epidemiologists, outcome researchers, health economists, and health policy decision-makers.
PREREQUISITE: Students are expected to have a basic knowledge in epidemiologic studies and methods (including the concept of confounding).
LEVEL: Advanced
TRACK: Real World Data & Information Systems
This short course is offered in-person at the ISPOR 2025 conference. Separate registration is required. Visit the ISPOR 2025 Program page to register and learn more.
FACULTY MEMBERS
Schedule:
LENGTH: 4 Hours | Course runs 1 day
Sunday, 13 May 2025 | Course runs 1 Day
1:00pm-5:00pm Eastern Daylight Time (EDT)
ISPOR short courses are designed to enhance knowledge and techniques in core health economics and outcomes research (HEOR) topics as well as emerging trends in the field. Short courses offer 4 or 8 hours of premium scientific education and an electronic course book. Active attendee participation combined with our expert faculty creates an immersive and impactful learning experience. Short courses are not recorded and are only available during the live course presentation.