Treatment-competing events in dynamic regimes.

Area of Science:

• Biostatistics
• Clinical Trial Design
• Health Services Research

Background:

• Dynamic treatment regimes adapt medical interventions based on patient status over time.
• Chronic diseases often require complex, longitudinal treatment strategies.
• Treatment-competing events can interrupt or alter planned treatment sequences.

Purpose of the Study:

• To develop a methodology for dynamic treatment regimes that accounts for treatment-competing events.
• To integrate treatment-competing events into a dynamic infusion policy framework.
• To propose an estimator for a relevant causal estimand in the presence of competing events.

Main Methods:

• Utilized counting processes to model treatment-competing events within dynamic regimes.
• Developed a dynamic infusion policy framework accommodating competing events.
• Applied the methodology to a dataset of patients undergoing coronary stent implantation.

Main Results:

• Demonstrated the incorporation of treatment-competing events into dynamic infusion policies.
• Provided an illustration of how the methodology can suggest a causal estimand estimator.
• Successfully exemplified the methods in a real-world clinical scenario.

Conclusions:

• The proposed framework effectively integrates treatment-competing events into dynamic treatment regimes.
• The methodology offers a robust approach for analyzing complex treatment strategies in chronic diseases.
• The study provides valuable insights for clinical decision-making and causal inference.