RT Journal
A1 Antonelli M, Giunta F, Ronco C
T1 POlymyxin b hemoperfusion and mortality in abdominal septic shock—reply
JF JAMA
JO JAMA
YR 2009
FD November 11
VO 302
IS 18
SP 1968
OP 1970
DO 10.1001/jama.2009.1609
UL http://dx.doi.org/10.1001/jama.2009.1609
AB 
In Reply: Dr Vincent and Dr Amaral raise questions about the analytic approach used in our study. An unadjusted description of the data (eg, χ2 analysis) is sufficient if the study goal is to explore the survival of populations. However, if the goal is to investigate the relationship between exposure to a treatment and clinical outcomes, investigators aim to account for confounding factors as well. For EUPHAS, it was appropriate to adjust for the confounding effect of the Sequential Organ Failure Assessment. We used Cox proportional hazard analysis, a valid method used commonly in the literature.1- 2