Individualized treatment effect was predicted best by modeling baseline risk in interaction with treatment assignment

TL;DR

This study compares risk-based methods for predicting treatment effects, finding that modeling the interaction between baseline risk and treatment improves prediction accuracy, especially with larger samples and non-linear effects.

Contribution

It demonstrates that modeling the interaction between baseline risk and treatment assignment enhances treatment effect prediction accuracy, with specific methods performing best under different scenarios.

Findings

Linear interaction models perform well across many scenarios.

RCS models excel with strong non-linear effects and larger samples.

Adaptive methods need larger samples for optimal performance.

Abstract

Objective: To compare different risk-based methods for optimal prediction of treatment effects. Methods: We simulated RCT data using diverse assumptions for the average treatment effect, a baseline prognostic index of risk (PI), the shape of its interaction with treatment (none, linear, quadratic or non-monotonic), and the magnitude of treatment-related harms (none or constant independent of the PI). We predicted absolute benefit using: models with a constant relative treatment effect; stratification in quarters of the PI; models including a linear interaction of treatment with the PI; models including an interaction of treatment with a restricted cubic spline (RCS) transformation of the PI; an adaptive approach using Akaike's Information Criterion. We evaluated predictive performance using root mean squared error and measures of discrimination and calibration for benefit. Results: The linear-interaction model displayed optimal or close-to-optimal performance across many simulation scenarios with moderate sample size (N=4,250 patients; ~ 785 events). The RCS-model was optimal for strong non-linear deviations from a constant treatment effect, particularly when sample size was larger (N=17,000). The adaptive approach also required larger sample sizes. These findings were illustrated in the GUSTO-I trial. Conclusion: An interaction between baseline risk and treatment assignment should be considered to improve treatment effect predictions.