IMR Press / FBL / Volume 24 / Issue 8 / DOI: 10.2741/4785

Frontiers in Bioscience-Landmark (FBL) is published by IMR Press from Volume 26 Issue 5 (2021). Previous articles were published by another publisher on a subscription basis, and they are hosted by IMR Press on as a courtesy and upon agreement with Frontiers in Bioscience.

Analysis of repeated measures data in nutrition research
Show Less
1 Department of Statistics, Texas A and M University, College Station, TX, 77843, USA
2 Department of Animal Science, Texas A and M University, College Station, TX, 77843, USA
3 School of Mathematical and Physical Sciences, University of Technology, Sydney, Australia
*Correspondence: (Guoyao Wu)
Front. Biosci. (Landmark Ed) 2019, 24(8), 1377–1389;
Published: 1 June 2019

Amino acid nutrition studies often involve repeated measures data. An example is that the concentrations of plasma citrulline in steers are repeatedly measured from the same animals. The standard repeated measures ANOVA method does not detect significant time changes in the concentrations of plasma citrulline within 6 hours after steers consumed rumen-protected citrulline, while a graphical analysis indicates that there exists a time effect. Here we describe three mixed model analyses that capture the time effect in a statistically significant way, while accounting for the correlations of measurements over time from the same steers. First, we allow flexible variance-covariance structures on our model. Second, we use baseline measurements as a covariate in our model. Third, we use percent-change from baseline as a data normalization method. In our data analysis, all these three approaches can lead to meaningful statistical results that oral administration of rumen-protected citrulline enhances the concentrations of plasma citrulline over time in ruminants. This supports the notion that rumen-protected citrulline can bypass the rumen to effectively enter the blood circulation.

Baseline Covariates
Mixed effects model
Percent-change from baseline
Repeated measures data
Variance-covariance structures
Figure 1.
Back to top