[Skip to Content]
[Skip to Content Landing]
August 21, 1996

Acute Renal Failure and Mortality-Reply

Author Affiliations

Brigham and Women's Hospital Boston, Mass
Yale University School of Medicine New Haven, Conn

JAMA. 1996;276(7):529. doi:10.1001/jama.1996.03540070025025

In Reply.  —Dr Jacobson correctly points out that unmeasured factors can act as confounders in observational research by accounting for both the exposure (in our study, the development of renal failure) and the outcome (the associated increase in mortality). He suggests that a preliminary analysis should have focused on identifying the causes of renal failure, which could then be evaluated as potential confounders. The apparent causes of renal failure in our index subjects were advanced age, diabetes, sepsis, liver disease, congestive heart failure, and hypotension. These factors, which have been noted to predispose patients toward the development of acute renal failure in other studies,1-3 were more common in our study's index subjects than in their matched counterparts without renal failure. Each of these factors was examined as a potential confounder in the analysis. Many other known causes of acute renal failure, such as pigment injury or aminoglycoside exposure, occurred too infrequently