Randomized and observational studies in nephrology: how strong is the evidence?

Abstract

The strength of the evidence that can be obtained from observational studies and randomized clinical trials (RCTs) has been widely discussed. This report critically examines 2 perspectives of this discussion in the setting of clinical studies in nephrology that have the objective of determining the efficacy of therapeutic interventions. Emphasizing the fundamental role of randomization in supporting causal inference, the first perspective asserts the strict superiority of RCTs for showing efficacy and emphasizes the susceptibility of observational studies to bias that cannot be corrected without invoking untestable assumptions. We also examine counterarguments to this strictly hierarchical perspective that support a greater role for observational research. The counterarguments cite limitations of RCTs, systematic reviews of the concordance between results of observational studies and RCTs, and methodological advances that have strengthened the inferences that can be made from observational data. Ultimately, we support an integrative approach that targets the use of observational studies and RCTs at different stages of the research process based on their respective strengths and weaknesses and seeks to maximize the information gained by joint evaluation of both types of evidence.