We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

stat.ME

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Statistics > Methodology

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

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.
Subjects: Methodology (stat.ME)
Cite as: arXiv:2205.01717 [stat.ME]
  (or arXiv:2205.01717v4 [stat.ME] for this version)

Submission history

From: Alexandros Rekkas [view email]
[v1] Tue, 3 May 2022 18:30:19 GMT (20456kb,D)
[v2] Thu, 5 May 2022 13:03:02 GMT (7202kb,D)
[v3] Wed, 25 May 2022 07:33:32 GMT (4154kb,D)
[v4] Mon, 4 Jul 2022 16:00:51 GMT (4154kb,D)

Link back to: arXiv, form interface, contact.