Sample size calculation should be performed for design accuracy in diagnostic test studies

Area of Science:

• Biostatistics
• Medical Research Methodology
• Diagnostic Test Evaluation

Background:

• Published guidelines clarify requirements for diagnostic test studies, including case/control selection and reference standards.
• Accurate sample size determination is crucial for the validity of diagnostic test research.
• Existing methods for sample size calculation in this field require practical tools for researchers.

Purpose of the Study:

• To provide researchers with tables for determining appropriate sample sizes in diagnostic test studies.
• To facilitate accurate estimation of diagnostic test sensitivity and its confidence limits.
• To support the design of studies that meet established guidelines for medical literature.

Main Methods:

• Sample size calculations are based on the binomial distribution.
• Determines the minimum number of cases (Ncases) needed for a desired sensitivity precision.
• Calculates the minimum number of controls (Ncontrols) for population representativeness using the formula: Ncontrols = Ncases * [(1-Prev)/Prev].

Main Results:

• Presents tables with Ncases values for sensitivities ranging from 0.60 to 0.99.
• Includes values for acceptable lower 95% confidence limits from 0.50 to 0.98.
• Specifies a 5% probability for the estimated lower confidence limit falling below the acceptable level.

Conclusions:

• Sample size calculations are indispensable for diagnostic test studies.
• Performing these calculations ensures the accuracy and reliability of study designs.
• Adherence to sample size guidelines enhances the quality of medical literature on diagnostic tests.