A new preference-based analysis for randomized trials can estimate treatment acceptability and effect in compliant...

Area of Science:

• Clinical Trials
• Biostatistics
• Health Services Research

Background:

• Traditional intention-to-treat (ITT) analysis in randomized trials has limitations.
• ITT analysis may not reflect treatment benefits for patients adhering to therapy.
• Per-protocol and as-treated analyses can introduce biases.

Purpose of the Study:

• To develop a novel analytical method for assessing treatment acceptability and benefit among compliers.
• To provide a more accurate estimation of treatment effects for patients committed to a specific therapy.
• To address the need for understanding treatment benefits beyond ITT in clinical trials.

Main Methods:

• Characterizing trial participants into five preference groups: compliers, two types of preferers, and two types of insisters.
• Developing a framework to estimate the proportions of patients within these defined preference groups.
• Proposing and illustrating two sets of constraints necessary for estimating outcome rates and treatment effects in compliers.

Main Results:

• The proposed framework allows for the estimation of patient proportions across all five preference groups.
• Constraints are identified as necessary for estimating outcome rates and treatment effects among compliers.
• Numerical examples are provided to illustrate the application of the proposed constraints.

Conclusions:

• The new preference-based analysis offers an estimate of treatment benefit specifically for patients committed to therapy.
• This method provides valuable information often sought by physicians and patients regarding adherence.
• It complements traditional analyses by offering a more relevant measure of benefit for compliant patients.