Determining the cutoff based on a continuous variable to define two populations with application to vaccines.

Area of Science:

• Biostatistics
• Clinical Research Methodology
• Data Analysis

Background:

• Dichotomizing continuous variables in clinical research aids interpretation but lacks standardized methods.
• Continuous variable distributions can vary between disease populations, complicating dichotomization.
• Optimal cutoff point determination is crucial for reliable dichotomous variables.

Purpose of the Study:

• To develop and evaluate a novel methodology for determining optimal cutoff points for dichotomizing continuous variables.
• To enhance the utility of dichotomous variables as endpoints in clinical studies.
• To provide a robust method applicable even when outcome status is not fully identified.

Main Methods:

• Developed a methodology to find the optimal cutoff by maximizing correlation between continuous variable distributions and the dichotomous outcome.
• Recommended the Expectation-Maximization (EM) algorithm for parameter estimation in scenarios with incomplete outcome data.
• Investigated method performance across various continuous variable distributions (normal, log-normal, exponential).

Main Results:

• The proposed method effectively determines optimal cutoff points for dichotomization.
• The EM algorithm integration addresses real-world data challenges with missing outcome status.
• The methodology demonstrated robust performance across tested distributions.

Conclusions:

• The developed methodology offers a statistically sound approach for dichotomizing continuous variables in clinical research.
• This method improves the interpretability and utility of dichotomous variables as study endpoints.
• Application to a varicella vaccine case study validates the method's practical utility.