Definition
Selection bias occurs when selection into the study is related to both the intervention and the outcome. This type of bias concerns only selection into the study based on characteristics observed after the intervention.
Signalling Questions
- Was selection of subjects into the study based on subject characteristics observed after the start of the intervention?
- Were these post-intervention subject characteristics that influenced selection likely to be influenced by the outcome and the intervention?
Examples
In a retrospective study, the effect of perioperative effects of steroids for IVDD is used, but patients were included that already received steroids prior to surgery for IVDD.