1、There are known knowns. These are things we know that we know. There are known unknowns. That is to say, there are things that we know we dont know. But there are also unknown unknowns. There are things we dont know we dont know.Donald Rumsfeld,Bias,A systematic error (caused by the investigator or
2、the subjects) that causes an incorrect (over- or under-) estimate of an association.,Bias,Protective effect No Difference Increased risk,0 1.0 10,Relative Risk,True Effect,True value,Precise & Accurate,Null,Suppose a study was conducted multiple times in an identical way.,Errors Affecting Validity,C
3、hance (Random Error; Sampling Error)Bias (Systematic Errors inaccuracies)Selection biasLoss to follow-up bias Information bias Nondifferential (e.g. simple misclassification) Differential Biases (e.g., recall bias, interviewer bias) Confounding (Imbalance in Other Factors),Consider:,Selection Bias,O
4、ccurs when selection, enrollment, or continued participation in a study is somehow dependent on the likelihood of having the exposure of interest or the outcome of interest.,Selection bias can cause an overestimate or underestimate of the association.,Selection bias can occur in several ways:Selecti
5、on of a comparison group (“controls“) that is not representative of the population that produced the cases in a case-control study. (Control selection bias) Differential loss to follow up in a cohort study, such that the likelihood of being lost to follow up is related to outcome status and exposure
6、 status. (Loss to follow-up bias) Refusal, non-response, or agreement to participate that is related to the exposure and disease (Self-selection bias) Using the general population as a comparison group for an occupational cohort study (“Healthy worker“ effect) Differential referral or diagnosis of s
7、ubjects,MGH 100 Hospital Cases,Selection bias can occur in a case-control study if controls are more (or less) likely to be selected if they have the exposure.,Do women of low SES have higher risk of cervical cancer?,Selection Bias in a Case-Control Study,200 Controls: Door-to-door survey of neighbo
8、rhood around the hospital during work day.,200 Controls: Door-to-door survey of neighborhood around the hospital during work day.,Problems: SE status of people living around the hospital may generally be different from that of the population that produced the cases. The door-to-door method of select
9、ing controls may tend to select people of lower (or higher) SE status.,Selection bias is not caused by differences in other potential risk factors (confounding). It is caused by selecting controls who are more (or less) likely to have the exposure of interest.,Selection bias can occur in a case-cont
10、rol study if controls are more (or less) likely to be selected if they have the exposure.,OR=3.0,Selection Bias in a Case-Control Study,OR=2.0,Exp.,YN,Exp.,YN,Dis.,Y N,Dis.,Y N,Control Selection Bias,True,Control Selection Bias The “Would” Criterion,Are the controls a representative sample of the po
11、pulation that produced the cases?If a control had developed cervical cancer, would she have been included in the case group? (“Would” criterion),You should try to fulfill the “would” criterion: if a control patient had had the disease being studied, is it likely that they would have ended up in the
12、case group? If the answer is “not necessarily,” then there is likely to be a problem with selection bias.,2,000,000 women age 20 in MA, & about 200 cases of cervical cancer per year.,If low SES were associated with cervical cancer with OR=3.0, MA would look like this.,OR = (75/25) = 2.0(120/80),(Bia
13、sed),Referred Cases,Are mothers of children with hemifacial microsomia more often diabetic?,Cases are referred, but what if controls are selected from the general pediatrics ward at MGH?,Could mothers of controls be more or less likely to be diabetic than the cases (regardless of any association bet
14、ween diabetes and microsomia)?,Referral mechanism of controls might be very different from that of the cases with microsomia.,How would you select controls for this study?,Selection bias can be introduced into case-control studies with low response or participation rates if the likelihood of respond
15、ing or participating is related to both the exposure and outcome.Example: A case-control study explored an association between family history of heart disease (exposure) and the presence of heart disease in subjects. Volunteers are recruited from an HMO. Subjects with heart disease may be more likel
16、y to participate if they have a family history of disease.,Self- Selection Bias in a Case-Control Study,Self-Selection Bias in a Case-Control Study,OR=2.25,OR=3.0,Exp.,YN,Exp.,YN,Dis.,Y N,Dis.,Y N,Self-Selection Bias,True,Best solution is to work toward high participation (80%) in all groups.,In a r
17、etrospective cohort study selection bias occurs if selection of exposed & non-exposed subjects is somehow related to the outcome.,Selection Bias in a Retrospective Cohort Study,What will be the result if the investigators are more likely to select an exposed person if they have the outcome of intere
18、st?,Example: Investigating occupational exposure (an organic solvent) occurring 15-20 yrs. ago in a factory.Exposed & unexposed subjects are enrolled based on employment records, but some records were lost.,Selection Bias in a Retrospective Cohort Study,Suppose there was a greater likelihood of reta
19、ining records of those who were exposed & got disease.,RR=2.0,Selection Bias in a Retrospective Cohort Study,RR=2.42,Exp.,YN,Exp.,YN,Dis.,Y N,Dis.,Y N,True,20% of employee health records were lost or discarded, except in “solvent” workers who reported illness (1% loss).,Workers in the exposed group
20、were more likely to be included if they had the outcome of interest.,Differential “referral” or diagnosis of subjects,General Population,vs.,Mortality Rates?,The general population is often used in occupational studies of mortality, since data is readily available, and they are mostly unexposed.,The
21、 main disadvantage is bias by the “healthy worker effect.” The employed work force (mostly healthy) generally has lower rates of mortality and disease than the general population (with healthy & ill people).,The “Healthy Worker” Effect,Can be considered a form of selection bias because the general p
22、opulation controls have a higher probability of getting the outcome (death).,Enrollment into a prospective cohort study will not be biased by the outcome, because the outcome has not occurred at enrollment. However, prospective cohort studies can have selection bias if the exposure groups have diffe
23、rential retention of subjects with the outcomes of interest. This can cause either an over- or under- estimate of association,Differential Retention (Loss to Follow Up) in Prospective Cohort Studies,OR=2.0,Selection Bias in a Prospective Cohort Study,RR=1.0,Exp.,YN,Exp.,YN,Dis.,Y N,Dis.,Y N,Loss to
24、Follow Up Bias,True,More events lost in one exposure group,Differential loss to follow up in a prospective cohort study on oral contraceptives (OC) & thromboembolism (TE).,If OC were associated with TE with RR=2.0 (TRUTH), the 2x2 for all of MA would look like this.,There is 40% loss to follow up ov
25、erall, but a greater tendency to loose OC users with TE results in a de facto selection.,RR = (8/5988) = 1.0(8/5998),If OC users with TE are more likely to be lost than non-OC-users with TE,(Biased),Observation Bias (Information Bias),Systematic errors due to incorrect categorization.,The Correct Cl
26、assification,Non-differential Misclassification (random): If errors are about the same in both groups, it tends to minimize any true difference between the groups (bias toward the null).,Misclassification Bias,Differential Misclassification (non-random): If information is better in one group than an
27、other, the association maybe over- or underestimated.,Subjects are misclassified with respect to their risk factor status or their outcome, i.e., errors in classification.,=,Non-Differential Misclassification,Difficulty remembering exposures (equal in both groups) Example: Case-control study of hear
28、t disease and past activity: difficulty remembering your specific exercise frequency, duration, intensity over many years Recording and coding errors in records and databases. Example: ICD-9 codes in hospital discharge summaries. Using surrogate measures of exposure: Example: Using prescriptions for
29、 anti-hypertensive medications as an indication of treatment Non-specific or broad definitions of exposure or outcome. Example: “Do you smoke?” to define exposure to tobacco smoke.,When errors in exposure or outcome status occur with approximately equal frequency in groups being compared.,Random err
30、ors in classification of risk factors or outcome (i.e., error rate about the same in all groups).,Non-Differential Misclassification,Example: When patients are discharged, the MD dictates a summary which is transcribed. Diagnoses and procedures noted on the summary are encoded (ICD-9 codes) and sent
31、 to the MA Health Data Consortium. MDs dont list all relevant diagnoses. Coders assign incorrect codes (they arent MDs).Errors occur in 25-30% of records.,Random errors in classification of risk factors or outcome (i.e., error rate about the same in all groups).,Non-Differential Misclassification,Te
32、nds to minimize differences, generally causing an underestimate of effect.,Effect:,CAD Controls Diabetes 40 10 No diabetes 60 90 OR= 40x90 = 6.0 10x60,CAD Controls Diabetes 20 5 No diabetes 80 95OR= 20x95 = 4.755x80,True Relationship,With Nondifferential Misclassification,Example: A case-control stu
33、dy comparing CAD cases & controls for history of diabetes. Only half of the diabetics are correctly recorded as such in cases and controls.,When there are random errors in classification of risk or outcome, i.e. errors occur with equal frequency in both groups.,Non-Differential Misclassification,Eff
34、ect: With a dichotomous exposure, it minimizes differences & causes an underestimate of effect, i.e. “bias toward the null.”,Nondifferential Misclassification of Exposure #1,Nondifferential Misclassification of Exposure #2,Validation to Identify Random Misclassification in a Prospective Cohort Study
35、,Obesity & heart disease in women (questionnaires):Guessing at weight?,“Self-reported weights were validated in a subsample of 184 NHS participants living in the Boston, MA area and were highly correlated with actual measured weights (r = 0.96).”,Cho E, Manson JE, et al.: A Prospective Study of Obes
36、ity and Risk of Coronary Heart Disease Among Diabetic Women. Diabetes Care 25:11421148, 2002.,Differential Misclassification,Differences in accurately remembering exposures (unequal) Example: Mothers of children with birth defects will remember the drugs they took during pregnancy better than mother
37、s of normal children (maternal recall bias). Interviewer or recorder bias. Example: Interview has subconscious belief about the hypothesis. More accurate information in one of the groups. Example: Case-control study with cases from one facility and controls from another with differences in record ke
38、eping.,When there are more frequent errors in exposure or outcome classification in one of the groups.,Recall Bias,People with disease may remember exposures differently (more or less accurately) than those without disease.,To Minimize: Use a control group that has a different disease (unrelated to
39、the disease under study). Use questionnaires that are constructed to maximize accuracy and completeness. Ask specific questions. More accuracy means fewer differences. For socially sensitive questions, such as alcohol and drug use or sexual behaviors, use a self-administered questionnaire instead of
40、 an interviewer. If possible, assess past exposures from biomarkers or from pre-existing records.,(If the groups have the same % of errors based on faulty memory, thats non-differential misclassification.),(Differential),Systematic difference in soliciting, recording, or interpreting information.,In
41、terviewer Bias (& Recorder Bias in Chart Reviews),Minimized by: Blinding the interviewers if possible. Using standardized questionnaires consisting of closed-end, easy to understand questions with appropriate response options. Training all interviewers to adhere to the question and answer format str
42、ictly, with the same degree of questioning for both cases and controls. Obtaining data or verifying data by examining pre-existing records (e.g., medical records or employment records) or assessing biomarkers.,(Differential),Interviewer bias,Recall Bias,Differential Misclassification,Non-Differentia
43、l Misclassification,Bias to Null,These are differential and can bias toward or away from null.,Selection bias,Effects of Bias,Misclassification of Outcome Can Also Introduce Bias, but it usually has much less of an impact than misclassification of exposure, because:Most of the problems with misclass
44、ification occur with respect to exposure status, not outcome. There are a number of mechanisms by which misclassification of exposure can be introduced, but most outcomes are more definitive and there are few mechanisms that introduce errors in outcome. Most outcomes are relatively uncommon. Misclas
45、sification of outcome will generally bias toward the null, so if an association is demonstrated, if anything the true effect might be slightly greater.,Any concerns?,A study is conducted to see if serum cholesterol screening reduces the rate of heart attacks. 1,500 members of an HMO are offered the
46、opportunity to participate in the screening program, & 600 volunteer to be screened. Their rates of MI are compared to those of randomly selected members who were not invited to be screened. After 3 years of follow-up rates of MI are found to be significantly less in the screened group.,No Different
47、ial misclassification Interviewer bias Recall bias Selection bias,Background Information on Abdominal Aortic Aneurysms,Usually asymptomatic (surgery if 5 cm.)Discovered during routine abdominal exam by palpation, orSeen on x-ray or ultrasound of abdomen (done for other reasons).,Diagnosis of AAA,Kno
48、wn risk factors:AgeMale genderSmokingHypertension,Costa & Robbs: Br. J. Surg. 1986 Abdominal Aneurysms.,A vascular surgery (referral) service in So. Africa reviewed records of elective peripheral vascular surgery.,Conclusion:AAA uncommon in Blacks and more often due to infections.,OR = 0.12 (0.09 0.
49、15),320 1,862,Other: a variety of readily apparent conditions.,Was there selection bias?,Other: variety of readily apparent conditions.,Yes No,Was there selection bias?,Other: variety of readily apparent conditions.,A possibility of misclassification?,“AAA in blacks are more often due to infectious causes.”,
