Lecture #13 Bias & Misclassification Flashcards
Bias
- systematic (non-random) error in study design or conduct leading to erroneous results
- distorts the relationship between exposure and outcome
- Nothing can be done to fix a bias once it has already occurred
- prospective consideration/adjustment can minimize impact of bias
Components of Bias
evaluate 3 components:
1. source/type
- magnitude/strength- could count entirely for weak association but not likely to account entirely for a very strong association
- direction- bias can over or under estimate true measure of association (has enhancing or minimizing effect on the true measure of association towards or away form 1)
Measure Related Biases (Informational/Observational)
- any aspect in the way the researcher collects information or measures/observes subjects (and their characteristics/variables) which creates a systematic difference between groups in the quality/accuracy of their information
- errors in measurement can cause a resultant error in patient classification (misclassification)
Selection Related Biases
- any aspect in the way the researcher selects or acquires study subjects (cases/controls or exposed/unexposed; even study sample) which creates a systematic difference in the composition between groups
- commonly seen when comparative groups not coming from the same population or not being represented of the full population or even differently selected
Healthy Worker Bias
- selection bias
- can easily be seen in prospective cohort studies
ex) an exposer impacts a patients health and you only look at the healthy people in the office and not the ones sick at home or dead from the exposure (makes exposure impact look smaller)
Self-Selection/Participant Responder Bias
- selection bias
- those that wish to participate may be different in some way to those that don’t volunteer to participate
ex) most people drop their head when see a surveyor, the ones who run up and want to take the survey may be different
Control Selection Bias
- selection bias
- can easily be seen in case-control studies
ex) calling people in the middle of the day to survey, som people don’t have a phone and some people can’t answer during the day
Subject Related Variation
- measurement bias done by the patient or subject
- recall bias
- contamination bias
- compliance/adherence bias
- lost follow up bias
Recall (Reporting) Bias
- differential level of accuracy/detail in provided info between study groups
- exposed subjects may have greater sensitivity for recalling their history or amplify their responses
- individuals can report their effects of exposure differently because they are part of a study
- “Hawthorne Effect” some people want to be overly helpful and overly positive/encouraging, give you more than you need
Contamination Bias
- members of the control group accidentally or outside of the study protocol receive treatment or are exposed to the intervention being studied
ex) ASA vs placebo but someone accidentally takes Advil or naproxen and it contaminates results b/c they were supposed to be on placebo
Compliance/Adherence Bias
- groups being interventionally studied have different compliances
ex) some people have trouble adhering to the rules
Lost to Follow-up Bias
- groups being studied have different withdrawal or lost to follow-up rates OR there are other differences between those that stay in the study and those that withdraw
- differential vs non-differential
ex) move, die, lose them. ex- want pain killer most surgery but on placebo and demand the pain killer or i will leave
ex) drop out b/c of side effects of medication
Observer Related Variation
-measurement bias
- interviewer bias
- diagnosis/surveillance (expectation) bias
Interviewer (Proficiency) Bias
- systematic difference in soliciting, recording, or interpreting on the part of the researcher
- interviewers knowledge may influence structure or tone of questions which may influence response from the subject
- treatments not applied equally between groups due to skill or training differences or differences in procedure compliance by staff
- conscious or unconscious
Diagnosis/Surveillance (Expectation) Bias
- similar to interviewer bias
- different evaluation, classification, dx, or observation between groups
- observers may have preconceived expectations of what they should find in examination, evaluation or follow-up
- “Hawthorne like effect” from the researchers perspective, researcher overly nice/positive really explains questions, etc.
