Extending inferences from a randomized trial to a new target population

Area of Science:

• Biostatistics
• Epidemiology
• Causal Inference

Background:

• Randomized trials are crucial for estimating treatment effects.
• Treatment effect modifiers can bias trial generalizability if they influence participation.
• Extending causal inferences from trials to target populations is essential for real-world application.

Purpose of the Study:

• To present and compare methods for extending causal inferences from randomized trials to target populations of nonparticipants.
• To address situations where treatment effect modifiers influence trial participation.
• To provide practical guidance on implementing these methods in software.

Main Methods:

• Utilized methods based on modeling outcome expectations, participation probabilities, or both (doubly robust methods).
• Employed data from a completed randomized trial and baseline covariate data from a target population sample.
• Conducted a simulation study to compare the performance of different methods.

Main Results:

• Demonstrated methods for extending causal inferences from trial participants to nonparticipants.
• Showcased practical implementation of these causal inference techniques.
• Applied methods to a real-world case study comparing surgical versus medical therapy for coronary artery disease.

Conclusions:

• Methods exist to generalize trial findings to target populations, even with non-random participation.
• Careful consideration of issues arising in applied analyses is necessary for valid causal inference.
• The presented techniques enhance the utility of randomized trial data for broader public health and clinical decision-making.