Agreement between methods of measurement with multiple observations per individual

Area of Science:

• Biostatistics
• Medical Statistics
• Quantitative Methods

Background:

• Limits of agreement assess agreement between measurement methods.
• Standard calculations assume independent observations, which is not always the case.
• Repeated measurements introduce complexity due to potential changes in the true value or the underlying quantity.

Purpose of the Study:

• To develop and describe methods for analyzing clustered agreement data.
• To address situations with repeated measurements on the same subject.
• To account for scenarios where the underlying quantity may be changing or constant.

Main Methods:

• Analysis of clustered observations in agreement studies.
• Methods for handling repeated pairs of measurements.
• Statistical approaches for changing and unchanging underlying quantities.

Main Results:

• The paper details analytical methods for clustered agreement data.
• The proposed methods are applicable to various repeated measurement designs.
• The techniques are suitable for both dynamic and static underlying quantities.

Conclusions:

• New methods are provided for analyzing agreement with clustered data.
• These methods extend the application of limits of agreement to complex designs.
• The study offers practical solutions for researchers dealing with repeated measurements.