This paper reports the multivariate analysis of experimental results from more than one hundred FRC prismatic specimens tested under sustained flexural loads for at least 90 days, collected from previously published sources. Principal Component Analysis was used to minimise the number of variables in the modelling process while compromising the minimum amount of information. The creep parameters analysed were the creep coefficients at 14, 30, and 90 days and the corresponding crack opening rates. They were related to the following factors: concrete compressive strength, residual load-bearing capacity in flexure, fibre material, and load. Multiple linear regression was used for the modelling of these relationships. Higher levels of flexural toughness were found to significantly reduce the variability of all creep parameters. Differences in fibre material were detected to introduce important differences in interaction with other factors, especially the load ratio, which was attributed to the association between the range of flexural toughness to be expected and the type of fibres used.