Editing Epidemiology (section)

==== Confounding ====
[[Confounding]] has traditionally been defined as bias arising from the co-occurrence or mixing of effects of extraneous factors, referred to as confounders, with the main effect(s) of interest.<ref name=Rothman2002/><ref name=Greenland>{{cite journal |vauthors=Greenland S, Morgenstern H | s2cid = 4647751 | year = 2001 | title = Confounding in Health Research | journal = Annu. Rev. Public Health | volume = 22 | pages = 189–212 | doi=10.1146/annurev.publhealth.22.1.189| pmid = 11274518 | doi-access =  }}</ref> A more recent definition of confounding invokes the notion of ''counterfactual'' effects.<ref name=Greenland/> According to this view, when one observes an outcome of interest, say Y=1 (as opposed to Y=0), in a given population A which is entirely exposed (i.e. exposure ''X''&nbsp;=&nbsp;1 for every unit of the population) the risk of this event will be ''R''<sub>A1</sub>. The counterfactual or unobserved risk ''R''<sub>A0</sub> corresponds to the risk which would have been observed if these same individuals had been unexposed (i.e. ''X''&nbsp;=&nbsp;0 for every unit of the population). The true effect of exposure therefore is: ''R''<sub>A1</sub>&nbsp;−&nbsp;''R''<sub>A0</sub> (if one is interested in risk differences) or ''R''<sub>A1</sub>/''R''<sub>A0</sub> (if one is interested in relative risk). Since the counterfactual risk ''R''<sub>A0</sub> is unobservable we approximate it using a second population B and we actually measure the following relations: ''R''<sub>A1</sub>&nbsp;−&nbsp;''R''<sub>B0</sub> or ''R''<sub>A1</sub>/''R''<sub>B0</sub>. In this situation, confounding occurs when ''R''<sub>A0</sub>&nbsp;≠&nbsp;''R''<sub>B0</sub>.<ref name=Greenland/> (NB: Example assumes binary outcome and exposure variables.)

Some epidemiologists prefer to think of confounding separately from common categorizations of bias since, unlike selection and information bias, confounding stems from real causal effects.<ref name="Hernán2004">{{cite journal |last1=Hernán |first1=M. A. |last2=Hernández-Díaz |first2=S. |author-link2=Sonia Hernández-Díaz |last3=Robins |first3=J. M. |year=2004 |title=A structural approach to selection bias |journal=Epidemiology |volume=15 |issue=5 |pages=615–25 |doi=10.1097/01.ede.0000135174.63482.43 |pmid=15308962 |s2cid=1373077 |doi-access=free}}</ref>