I am reviewing a paper where two different sensors which measure the same variable are being compared. Over the course of a year, each month a patient will wear both sensors for one week to gather data (so 12 weeks total data). From the 12 weeks of data the averages of different metrics can be computed. The analysis then uses the averaged data from the 44 patients to construct a Bland-Altman plot, which is then used to draw conclusions.

It seems to me that instead of averaging the data over the entire year for each patient, each of the 12 measurements from a patient constitute repeated measures, and so the appropriate analysis is to compute the metrics from each week of data and then use the methods described in the Bland and Altman paper where there are multiple measurements per patient (https://doi.org/10.1080/10543400701329422). Does this seem correct?



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.