Abstract
There have only been two randomized controlled trials studying outcome as a function of dose in hemodialysis (HD). The first was the National Cooperative Dialysis Study which showed that adequate dialysis was achieved with spKt/V >1.00. The second study was HEMO which was originally designed to study spKt/V 1.2 compared to spKt/V 1.45. Unfortunately by the time HEMO was started, observational studies (OS) had convinced the nephrology community that the minimum adequate dose of spKt/V was 1.40, so the lower target was increased to 1.4 and the upper target to 1.7. The study showed no difference in outcome, although OS have now demonstrated that outcome improves up to spKt/v 2.00. Analysis of HEMO as treated showed that there is a fundamental flaw in dose-targeted OS in that the optimal dose always, but spuriously, increases as the studied dose increases due to dose-targeting bias. Similar flaws exist in the association of treatment time to outcome.