Click here for file(31K, DOC)Additional file 5:Lengths of ICU stay after reaching maximum RIFLE class in nonsurvivors with and without renal replacement therapy (RRT).Click here for file(31K, DOC)Additional file 6:Association of renal replacement therapy (RRT) with hospital mortality in multivariate conditional logistic regression according to timing of RRT and maximum RIFLE class reached during the ICU stay (model 1).Click here for file(36K, DOC)Additional file 7:Association of renal replacement therapy (RRT) with hospital mortality in multivariate conditional logistic regression according to timing of RRT and maximum RIFLE class reached during the ICU stay (model 2).Click here for file(36K, DOC)Additional file 8:Association of renal replacement therapy (RRT) with hospital mortality in multivariate conditional logistic regression (matched patients) according to timing of RRT: results of sensitivity analyses including only patients with a normal serum creatinine value measured on ICU admission.Click here for file(33K, DOC)NotesSee related commentary by Joannidis et al., http://ccforum.com/content/17/2/125AcknowledgementsWe are indebted in the persons listed below for their participation in the Outcomerea study group:Scientific committeeJean-Fran?ois Timsit (H?pital Albert Michallon and INSERM U823, Grenoble, France), Elie Azoulay (Medical ICU, H?pital Saint Louis, Paris, France), Yves Cohen (ICU, H?pital Avicenne, Bobigny, France), Ma?t�� Garrouste-Orgeas (ICU H?pital Saint-Joseph, Paris, France), Lilia Soufir (ICU, H?pital Saint-Joseph, Paris, France), Jean-Ralph Zahar (Microbiology Department, H?pital Necker, Paris, France), Christophe Adrie (ICU, H?pital Delafontaine, Saint Denis, France), and Christophe Clec’h (ICU, H?pital Avicenne, Bobigny, and INSERM U823, Grenoble, France).Biostatistical and informatics expertiseJean-Francois Timsit (H?pital Albert Michallon and INSERM U823, Grenoble, France), Sylvie Chevret (Medical Computer Sciences and Biostatistics Department, H?pital Saint-Louis, Paris, France), Corinne Alberti (Medical Computer Sciences and Biostatistics Department, Robert Debr��, Paris, France), Adrien Fran?ais (INSERM U823, Grenoble, France), Aur��lien Vesin INSERM U823, Grenoble, France), Christophe Clec’h (ICU, H?pital Avicenne, Bobigny, and INSERM U823, Grenoble, France), Frederik Lecorre (Supelec, France), and Didier Nakache (Conservatoire National des Arts et M��tiers, Paris, France).