Bias Confounding and the Role of Chance Principles

Bias, Confounding and the Role of Chance Principles of Epidemiology Lecture 5 Dona Schneider, Ph. D, MPH, FACE

To Show Cause We Use n Koch’s Postulates for Infectious Disease n Hill’s Postulates for Chronic Disease and Complex Questions n Strength of Association – Tonight’s entire lecture n Biologic Credibility n Specificity n Consistency with Other Associations n Time Sequence n Dose-Response Relationship n Analogy n Experiment n Coherence Epidemiology (Schneider)

To Show a Valid Statistical Association n We need to assess: n Bias: whether systematic error has been built into the study design n Confounding: whether an extraneous factor is related to both the disease and the exposure n Role of chance: how likely is it that we found is a true finding Epidemiology (Schneider)

BIAS Systematic error built into the study design Selection Bias Information Bias

Types of Selection Bias n Berksonian bias – There may be a spurious association between diseases or between a characteristic and a disease because of the different probabilities of admission to a hospital for those with the disease, without the disease and with the characteristic of interest Berkson J. Limitations of the application of fourfold table analysis to hospital data. Biometrics 1946; 2: 47 -53 Epidemiology (Schneider)

Types of Selection Bias (cont. ) n Response Bias – those who agree to be in a study may be in some way different from those who refuse to participate n Volunteers may be different from those who are enlisted Epidemiology (Schneider)

Types of Information Bias n Interviewer Bias – an interviewer’s knowledge may influence the structure of questions and the manner of presentation, which may influence responses n Recall Bias – those with a particular outcome or exposure may remember events more clearly or amplify their recollections Epidemiology (Schneider)

Types of Information Bias (cont. ) n Observer Bias – observers may have preconceived expectations of what they should find in an examination n Loss to follow-up – those that are lost to follow-up or who withdraw from the study may be different from those who are followed for the entire study Epidemiology (Schneider)

Information Bias (cont. ) n Hawthorne effect – an effect first documented at a Hawthorne manufacturing plant; people act differently if they know they are being watched n Surveillance bias – the group with the known exposure or outcome may be followed more closely or longer than the comparison group Epidemiology (Schneider)

Information Bias (cont. ) n Misclassification bias – errors are made in classifying either disease or exposure status Epidemiology (Schneider)

Types of Misclassification Bias n Differential misclassification – Errors in measurement are one way only n Example: Measurement bias – instrumentation may be inaccurate, such as using only one size blood pressure cuff to take measurements on both adults and children Epidemiology (Schneider)

Misclassification Bias (cont. ) True Classification Exposed Nonexposed Cases Controls Total 100 50 150 50 50 100 150 100 250 OR = ad/bc = 2. 0; RR = a/(a+b)/c/(c+d) = 1. 3 Differential misclassification - Overestimate exposure for 10 cases, inflate rates Exposed Nonexposed Cases Controls Total 110 40 150 50 50 100 160 90 250 OR = ad/bc = 2. 8; RR = a/(a+b)/c/(c+d) = 1. 6

Misclassification Bias (cont. ) True Classification Exposed Nonexpose d Cases Controls Total 100 50 150 50 50 100 250 OR = ad/bc =1502. 0; RR = a/(a+b)/c/(c+d) = 1. 3 Differential misclassification - Underestimate exposure for 10 cases, deflate rates Exposed Nonexpose d Cases Controls Total 90 50 140 60 50 110 150 100 250 OR = ad/bc = 1. 5; RR = a/(a+b)/c/(c+d) = 1. 2

Misclassification Bias (cont. ) True Classification Exposed Nonexpose d Cases Controls Total 100 50 150 50 50 100 250 OR = ad/bc =1502. 0; RR = a/(a+b)/c/(c+d) = 1. 3 Differential misclassification - Underestimate exposure for 10 controls, inflate rates Exposed Nonexpose d Cases Controls Total 100 40 140 50 60 110 250= 1. 6 OR = ad/bc 150 = 3. 0; RR =100 a/(a+b)/c/(c+d)

Misclassification Bias (cont. ) True Classification Cases 100 50 150 Exposed Nonexposed Controls 50 50 100 Total 150 100 250 OR = ad/bc = 2. 0; RR = a/(a+b)/c/(c+d) = 1. 3 Differential misclassification - Overestimate exposure for 10 controls, deflate rates Exposed Nonexpose d Cases Controls Total 100 60 160 50 40 90 150 100 250 OR = ad/bc = 1. 3; RR = a/(a+b)/c/(c+d) = 1. 1

Misclassification Bias (cont. ) n Nondifferential (random) misclassification – errors in assignment of group happens in more than one direction n This will dilute the study findings BIAS TOWARD THE NULL Epidemiology (Schneider)

Misclassification Bias (cont. ) True Classification Exposed Nonexpose d Cases Controls Total 100 50 150 50 50 100 250 OR = ad/bc 150 = 2. 0; RR = a/(a+b)/c/(c+d) = 1. 3 Nondifferential misclassification - Overestimate exposure in 10 cases, 10 controls – bias towards null Exposed Nonexpose d Cases Controls Total 110 60 170 40 40 80 250= 1. 3 OR = ad/bc 150 = 1. 8; RR = 100 a/(a+b)/c/(c+d)

Controls for Bias n Be purposeful in the study design to minimize the chance for bias n n Define, a priori, who is a case or what constitutes exposure so that there is no overlap n n Example: use more than one control group Define categories within groups clearly (age groups, aggregates of person years) Set up strict guidelines for data collection n Train observers or interviewers to obtain data in the same fashion n It is preferable to use more than one observer or interviewer, but not so many that they cannot be trained in an identical manner

Controls for Bias (cont) n Randomly allocate observers/interviewer data collection assignments n Institute a masking process if appropriate n n Single masked study – subjects are unaware of whether they are in the experimental or control group n Double masked study – the subject and the observer are unaware of the subject’s group allocation n Triple masked study – the subject, observer and data analyst are unaware of the subject’s group allocation Build in methods to minimize loss to followup