Selection Bias

An error due to selection of cases and controls based on differing criteria that are related to exposure status, or selection (or follow-up) of exposed and unexposed individuals in a way that is related to the development of the outcome. (Aschengrau & Seage, p. 263) Different types of selection bias include:

  1. Control Selection bias: A result of selecting controls from a different source population than the cases. (Aschengrau & Seage, pp. 264-266)
    # Self-selection bias: A type of bias which can result from differential rates of participation between cases and eligible controls. (Aschengrau & Seage, p. 266)
  2. Differential surveillance, diagnosis, or referral bias: can result from a tendency to hospitalize patients differentially based on their exposure status. Ex. Oral contraceptives and thromboembolism. (Aschengrau & Seage, p. 267)
  3. Losses to follow-up: When a subject whose outcome in unknown can no longer be followed in the course of a study, that individual is considered lost to follow-up.