In the quest for "evidence-based" medicine, an accepted hierarchy of evidence has been proposed. This hierarchy places in vitro studies and animal data at the base, and puts systematic reviews, meta-analyses, and randomized controlled trials at the pinnacle. However, when clinical medicine faces questions that have not yet been studied by the "gold standard" methods, how is one to proceed? Often, the best evidence at hand falls short of randomized controlled trials and meta-analyses. Using this framework, a review of the evidence supporting the hypothesis that population-wide diuretic use is directly associated with end-stage renal disease in the United States is presented. Publications pertaining to diuretic use in recent clinical trials are also discussed.