• Mashup Score: 27

    Collider-conditioning bias is a common concern in clinical epidemiology and can be difficult to recognize. Collider-conditioning bias is defined as the alteration in the association between an exposure and an outcome under study when conditioning on (via restriction, stratification or regression adjustment) a third factor influenced by both exposure and outcome. Collider-conditioning bias may also be induced by conditioning on a third factor influenced by causes of both the exposure and outcome of primary interest.

    Tweet Tweets with this article
    • RT @JClinEpi: Key concepts in clinical epidemiology: Collider-conditioning bias - Journal of Clinical Epidemiology https://t.co/zw7iuPfInR