Testing equality and interval estimation in binary responses when high dose cannot be used first under a three-period...

Area of Science:

• Biostatistics
• Clinical Trial Design
• Pharmacometrics

Background:

• Crossover trial designs are efficient for comparing treatments, especially in longitudinal studies.
• Sequential dose administration constraints (e.g., low dose before high dose) present unique analytical challenges.
• Accurate statistical methods are needed to evaluate treatment effects under such constraints.

Purpose of the Study:

• To develop and evaluate statistical procedures for analyzing dichotomous outcomes in three-period crossover trials.
• To address the specific scenario where the high drug dose cannot precede the low dose.
• To provide methods for testing treatment equality and estimating relative treatment effects.

Main Methods:

• Utilized a random-effects logistic regression model for dichotomous response data.
• Derived asymptotic test procedures for treatment comparisons.
• Developed interval estimators for quantifying relative treatment effects.
• Employed Monte Carlo simulations to assess the performance of the proposed methods.

Main Results:

• The study successfully derived and evaluated novel asymptotic test procedures and interval estimators.
• Monte Carlo simulations demonstrated the performance of these statistical tools under various conditions.
• The methods are applicable to clinical trials with ordered dose administration.

Conclusions:

• The proposed statistical methods provide a robust framework for analyzing three-period crossover trials with dose-order restrictions.
• These methods enable reliable assessment of drug efficacy and treatment effects in complex trial designs.
• The approach was illustrated using data from a trial on primary dysmenorrhea.