When selection becomes selection bias

European Journal of Endocrinology, vol. 192, pp. E1–E3

Abstract

Selection in clinical research does not necessarily result in selection bias. To understand when selection leads to bias, we discuss collider-conditioning bias, which is a common and often self-inflicted type of selection bias. Collider-conditioning bias may be difficult to recognize, and paying more attention to the bias could therefore increase the quality of research. In this paper, we aim to increase awareness and understanding of the topic.

Authors 4

1. Kristina Laugesen corresponding

   Aarhus University · Aarhus University Hospital

   Affiliation as printed

   Department of Clinical Biochemistry, Aarhus University Hospital , 8200 Aarhus ,

   Department of Clinical Epidemiology, Aarhus University and Aarhus University Hospital , 8200 Aarhus ,

   Department of Clinical Biochemistry, Aarhus University Hospital, Aarhus, Denmark

   Department of Clinical Epidemiology, Aarhus University and Aarhus University Hospital, Aarhus, Denmark

2. Olaf M. Dekkers Aachen Department of Clinical Epidemiology

   Leiden University · Aarhus University · Leiden University Medical Center · Aarhus University Hospital

   Affiliation as printed

   Department of Clinical Epidemiology, Aarhus University and Aarhus University Hospital , 8200 Aarhus ,

   Department of Clinical Epidemiology, Leiden University Medical Centre , 2333 ZA Leiden ,

   Department of Clinical Epidemiology, Aarhus University and Aarhus University Hospital, Aarhus, Denmark

   Department of Clinical Epidemiology, Leiden University Medical Centre, Leiden, Netherlands

3. Rolf H. H. Groenwold Aachen Department of Clinical Epidemiology

   Leiden University · Leiden University Medical Center

   Affiliation as printed

   Department of Clinical Epidemiology, Leiden University Medical Centre , 2333 ZA Leiden ,

   Department of Clinical Epidemiology, Leiden University Medical Centre, Leiden, Netherlands

4. Nils Skajaa

   Aarhus University · Aarhus University Hospital

   Affiliation as printed

   Department of Clinical Epidemiology, Aarhus University and Aarhus University Hospital , 8200 Aarhus ,

   Department of Clinical Epidemiology, Aarhus University and Aarhus University Hospital, Aarhus, Denmark