C. Mitterbauer(1), G. Heinze(2), A. Kainz(1,3), W.H. Hörl(1), R. Oberbauer(1,3)
1- Department of Nephrology, Medical University of Vienna, Austria
2- Core Unit of Medical Statistics and Informatics, Medical University of Vienna, Austria
3- Department of Nephrology, KH Elisabethinen, Linz, Austria
Assessment of working correlation structure.View PDF
Boxplot of potassium levels vs. year after engraftment, stratified for ACEI/ARB use.View PDF
Mixed linaer model with “sticky” group definitionView PDF
Number of ACEI/ARB user/nonusers vs. year after engraftment.View PDF
Testing for interaction between ACEI/ARB and year of use (sticky group definition)View PDF
Percentage of ACEI/ARB user/nonuser vs. year after engraftment.View PDF
Mixed linaer model without ACEI/ARB changers.
Subjects who switched groups were eliminated as suggested by the reviewer. The results are materially identical to the other models presented above and in the manuscript in table 3.
Residual diagnosticsView PDF
Test for interaction of ACEI/ARB and CNI use.View PDF
a,b: Residuals against predicted valuesView PDF
c,d: Absolute residuals against predicted values (to detect heteroscedasticity)View PDF