Bias

Question 0

Confounding definition

Answer 0

The mixing of effects between the exposure, the disease, and a 3rd variable termed a cofounder. Cofounders distort the relationship

Question 1

Bias definition

Answer 1

systematic error in the design or conduct of study that leads to an erroneous association between the exposure and the disease

results in an incorrect or invalid estimate of the measure of association

Question 2

Random error

Answer 2

probability that the observed result is due to chance, an uncontrollable force that seems to have no assignable cause

Question 3

2 main types of bias

Answer 3

selection bias - occurs during selection and follow-up

information/observation bias - occurs during data collection

Question 4

common selection bias studies

Answer 4

case-control
retrospective cohort

can occur in cohort and exp. studies from diff loss to follow-up

Question 5

nondifferential loss to follow up

Answer 5

relative measure of association unaffected

absolute measure biased towards the null

Question 6

differential loss to follow-up

Answer 6

can bias relative and absolute measures of association upward or downward

Question 7

information bias

Answer 7

systematic difference in the way that the exposure or outcome is measured between compared groups

• occurs after subjects enter study
• pertains to how data is collected
• results in incorrect classification of participants as either exposed/unexposed or diseased/not diseased

Question 8

healthy worker effect

Answer 8

selection bias that occurs in cohort studies (PMR and SMR)

healthy workers compared to general population

Question 9

Recall bias

Answer 9

differential level of accuracy in the information provided by compared groups

most common in case-control studies

Question 10

To minimize recall bias:

Answer 10

select a diseased control group

design structured questionnaires

Question 11

Interviewer bias

Answer 11

systematic difference in soliciting, recording, or interpreting information that occurs in studies using in-person or telephone interviews

Question 12

Misclassification

Answer 12

measurement error, most common

Question 13

How to reduce misclassification?

Answer 13

improve accuracy of collected data
validation
use better information source for the exposure or disease
define exposure/disease using sensitive and specific criteria

Question 14

Neyman bias

Answer 14

incidence-prevalence bias, arises when a gap in time occurs between exposure and selection of study participants

case group not representative of cases in the community