A multivariate model including all the significant predictors of log β2-microglobulin concentrations from table 2 is shown.

This model explains 62% of the variability of log β2-microglobulin concentrations and is thus not much more predictive than a model containing log eGFRBIS2, log age, and gender as independent variables. The partial correlation coefficients rpartial are given for significant predictors in the model and represent the correlation of each variable with log β2-microglobulin, adjusted for the effect of the other variables in the model.