Editing Epidemiology (section)

=== Systematic error ===
A systematic error or bias occurs when there is a difference between the true value (in the population) and the observed value (in the study) from any cause other than sampling variability. An example of systematic error is if, unknown to you, the [[pulse oximeter]] you are using is set incorrectly and adds two points to the true value each time a measurement is taken. The measuring device could be [[Accuracy and precision|precise but not accurate]]. Because the error happens in every instance, it is systematic. Conclusions you draw based on that data will still be incorrect. But the error can be reproduced in the future (e.g., by using the same mis-set instrument).

A mistake in coding that affects ''all'' responses for that particular question is another example of a systematic error.

The validity of a study is dependent on the degree of systematic error. Validity is usually separated into two components:
* [[Internal validity]] is dependent on the amount of error in measurements, including exposure, disease, and the associations between these variables. Good internal validity implies a lack of error in measurement and suggests that inferences may be drawn at least as they pertain to the subjects under study.
* [[External validity]] pertains to the process of generalizing the findings of the study to the population from which the sample was drawn (or even beyond that population to a more universal statement). This requires an understanding of which conditions are relevant (or irrelevant) to the generalization. Internal validity is clearly a prerequisite for external validity.

==== Selection bias ====
[[Selection bias]] occurs when study subjects are selected or become part of the study as a result of a third, unmeasured variable which is associated with both the exposure and outcome of interest.<ref name="Hernán2004"/> For instance, it has repeatedly been noted that cigarette smokers and non smokers tend to differ in their study participation rates. (Sackett D cites the example of Seltzer et al., in which 85% of non smokers and 67% of smokers returned mailed questionnaires.)<ref name="Sackett D. Bias in analytic research. J Chron Dis 1979; vol. 32:51–63.">[http://www.epidemiology.ch/history/PDF%20bg/Sackett%20DL%201979%20bias%20in%20analytic%20research.pdf] {{Webarchive|url=https://web.archive.org/web/20170829193522/http://epidemiology.ch/history/PDF%20bg/Sackett%20DL%201979%20bias%20in%20analytic%20research.pdf|date=29 August 2017}} 24</ref> Such a difference in response will not lead to bias if it is not also associated with a systematic difference in outcome between the two response groups.

==== Information bias ====
[[Information bias (epidemiology)|Information bias]] is bias arising from systematic error in the assessment of a variable.<ref name=Rothman2002>{{cite book|last1=Rothman|first1=K.|title=Epidemiology: An Introduction|url=https://archive.org/details/epidemiology00kenn|url-access=registration|date=2002|publisher=[[Oxford University Press]]|location=Oxford|isbn=978-0195135541}}</ref> An example of this is recall bias. A typical example is again provided by Sackett in his discussion of a study examining the effect of specific exposures on fetal health: "in questioning mothers whose recent pregnancies had ended in fetal death or malformation (cases) and a matched group of mothers whose pregnancies ended normally (controls) it was found that 28% of the former, but only 20% of the latter, reported exposure to drugs which could not be substantiated either in earlier prospective interviews or in other health records".<ref name="Sackett D. Bias in analytic research. J Chron Dis 1979; vol. 32:51–63." /> In this example, recall bias probably occurred as a result of women who had had miscarriages having an apparent tendency to better recall and therefore report previous exposures.

==== Design-related bias ====
Next to sample- and variable-related bias, bias can also arise from an imperfect study design. One example is immortal time bias, where during study period, there is some interval during which the outcome event cannot occur (making these individual "immortal").<ref>{{Cite journal |last1=Yadav |first1=Kabir |last2=Lewis |first2=Roger J. |date=2021-02-16 |title=Immortal Time Bias in Observational Studies |url=https://jamanetwork.com/journals/jama/article-abstract/2776315 |journal=JAMA |volume=325 |issue=7 |pages=686–687 |doi=10.1001/jama.2020.9151 |pmid=33591334 |issn=0098-7484}}</ref><ref>{{Cite journal |last1=Lévesque |first1=Linda E. |last2=Hanley |first2=James A. |last3=Kezouh |first3=Abbas |last4=Suissa |first4=Samy |date=2010-03-12 |title=Problem of immortal time bias in cohort studies: example using statins for preventing progression of diabetes |url=https://www.bmj.com/content/340/bmj.b5087 |journal=BMJ |language=en |volume=340 |pages=b5087 |doi=10.1136/bmj.b5087 |issn=0959-8138 |pmid=20228141}}</ref>

==== 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>