Fax +41 61 306 12 34 E-Mail karger@karger.ch www.karger.com Kidney Disease and Population Health Nephron Clin Pract 2009;113:c337–c342 DOI: 10.1159/000237143 The Randomized Controlled Trial Vianda S. Stel a Carmine Zoccali c Friedo W. Dekker a, b Kitty J. Jager a a European Renal Association – European Dialysis and Transplant Association Registry, Department of Medical Informatics, Academic Medical Center, University of Amsterdam, Amsterdam, and b Department of Clinical Epidemiology, Leiden University Medical Center, Leiden, The Netherlands; c Department of Clinical Epidemiology and Pathophysiology of Renal Diseases and Hypertension, Institute of Biomedicine and Molecular Immunology, National Research Council, Renal and Transplantation Unit, Ospedali Riuniti, Reggio Calabria, Italy Introduction For a research question on the effect of a therapy, the randomized controlled trial (RCT) is the most appropri- ate study design [1]. Nevertheless, because of several dif- ficulties in conducting an RCT [2], until recently high- quality RCTs have rarely been performed in nephrology [3, 4]. The purpose of this article is to describe several is- sues related to the design and quality of RCTs in their conduct, analysis, interpretation, and reporting of the re- sults. Isssues Related to the Design of RCTs Randomization To explain the concept of ‘randomization’, we use the hemodialysis (HEMO) trial of Eknoyan et al. [5] as an ex- ample. The HEMO trial was designed to determine wheth- er increasing the dose of dialysis or using a high-flux dia- lyzer membrane alters survival (or morbidity) among pa- tients undergoing hemodialysis. This RCT comprised a 2 ! 2 factorial design in which patients were randomly as- signed to a standard or high dose of dialysis and to a low- or high-flux dialyzer ( fig. 1 ). The investigators tested the effect of flux intervention (or Kt/V intervention) on mor- Key Words Epidemiology  Patient selection  Randomized controlled trials  Study design Abstract Until recently, high-quality randomized controlled trials (RCTs) have rarely been performed in nephrology, because of several difficulties in conducting an RCT. The purpose of this article is to describe several issues related to the design and to the quality of RCTs in their conduct, analysis, interpre- tation, and reporting of the results. The advantage of an RCT is that, as a result of randomization, selection by prognosis by the clinician is prevented. However, not all RCTs provide a definitive answer on the research question that the inves- tigators try to answer because of potential problems in their design, conduct and analysis. For example, the results of an RCT could be biased if the physicians and patients in the ex- perimental and control groups were not sufficiently ‘blind- ed’. Nevertheless, if conducted properly, RCTs remain the gold standard for studying the effects of therapy and other interventions. Copyright © 2009 S. Karger AG, Basel Published online: September 11, 2009 Vianda S. Stel, PhD Department of Medical Informatics, Academic Medical Center ERA-EDTA Registry, J1b 113.1, POB 22700 1100 DE Amsterdam (The Netherlands) Tel. +31 20 566 7637, Fax +31 20 691 9840, E-Mail v.s.stel@amc.uva.nl © 2009 S. Karger AG, Basel 1660–2110/09/1134–0337$26.00/0 Accessible online at: www.karger.com/nec