Linear mixed model better than repeated measures analysis

Abstract

We have some criticism regarding some technical issues. Mixed models have begun to play a pivotal role in statistical analyses and offer many advantages over more conventional analyses regarding repeated variance analyses. First, they allow to avoid conducting multiple t-tests; second, they can accommodate for within-patient correlation; third, they allow to incorporate not only a random coefficient, but also a random slope, typically ‘linear’ time in longitudinal case series when there are enough data and patients’ trajectories vary a lot and improving model fit.

Reply to Dr Işıl Kutluturk Karagoz letter to the editor

“…We agree that other statistical tests might have been more suitable for our study, such as a linear mixed model, recommended more recently by some biostatisticians to analyze data in longitudinal studies. 2 However, we disagree that correlation and regression analysis were of no significant benefit at all in our research. Measures of correlation ask, what is the relationship between variable A and variable B?…”

“…We also disagree with the statement: “Correlation between FAF degrees and functional and morphologic parameters section does not match the values given in figure 4.2.” 2 The determination coefficient described in the results section is 0.29, and in figure 4 (there is no 4.2 figure) is 0.28. A difference of 0.01 in a determination coefficient in this scenario is insignificant and not worth mentioning.…”