Bias and Confounding

Question 1

how did the Cochrane Handbook define bias?

Answer 1

systematic error, meaning multiple replications of the same study would reach the wrong answer on average

Question 2

what is bias in terms of validity

Answer 2

lack of internal validity or incorrect assessment of the association between exposure and effect in target population

Question 3

what is external validity

Answer 3

generalisation of the results observed in one population to others

Question 4

how can bias arise in review and publication

Answer 4

biases of rhetoric
positive results bias (especially peer-reviewed journals)
‘hot stuff’ bias
‘alls well’ literature bias

Question 5

what is selection bias

Answer 5

arises from procedure of selecting subjects and factors affecting participation

Question 6

what is the consequence of selection bias

Answer 6

relation between exposure and disease is DIFFERENT for study population and target population

Question 7

what is the target population

Answer 7

those eligible for inclusion in the study but not included

Question 8

what are the 4 types of selection bias

Answer 8

• ASCERTAINMENT BIAS = inappropriate definiton of study population
• NON-RANDOM SAMPLING = lack of accuracy in sampling frame
• uneven diagnosis in target population
• during study implementation e.g withdrawal, loss to follow up,

Question 9

what is attrition bias (selection)

Answer 9

systematic differences between groups on withdrawals from a study

Question 10

how can we try to prevent selection bias?

Answer 10

chose study pop with extreme care - representative of the target population
use validated disease registers
maximise TAKE UP and FOLLOW UP among subjects using standardised protocol
compare characteristics of study population to people who didnt respond/others

Question 11

when does information bias arise

Answer 11

during data collection
focus is misclassification bias

Question 12

what is misclassification bias (information)

Answer 12

bias in classifying either exposure or outcome

Question 13

what is observer (interviewer bias) bias
(information)

Answer 13

an interviewers knowledge may influence structure of questions and manner of presentation
which may influence responses

Question 14

what is recall bias - info bias

Answer 14

those with particular outcome/exposure may remember events more clearly or amplify their recollections

Question 15

what is reporting bias - info bias

Answer 15

participants may give answers they perceive to be of interest to investigators

Question 16

what is the Hawthorne effect - info bias

Answer 16

people act differently if they know they are being watched

Question 17

what is detection bias - info bias

Answer 17

if greater effort is given to diagnosing disease in exposed group than the unexposed group

Question 18

what is protopathic bias

Answer 18

exposure is influenced by early stages of disease
e.g drug is prescribed for early manifestation of disease that has not yet been diagnosed

Question 19

how can we deal with information bias

Answer 19

BLINDING - participants/those assessing exposure/ those assessing disease status
use standardised definition of exposure and disease
under standardised data collection instruments
maximise response to questionnaires

Question 20

what is ‘mistaken identity bias’

Answer 20

where strategies meant to alter patient behaviour may also alter clinician behaviour

Question 21

what is cognitive dissonance bias

Answer 21

belief in given mechanism may increase rather than decrease in face of contradictory evidence

Question 22

what is significance bias

Answer 22

statistical vs. clinical significance

Question 23

how to avoid analysis bias

Answer 23

state hypotheses to be tested in advance
follow analysis plan

Question 24

how to avoid interpretation bias

Answer 24

treat all interpretations with caution
look for evidence in systematic reviews

Question 25

what is confounding

Answer 25

the effect of extraneous factor is mistaken for or mixed with effect of the exposure under investigation

Question 26

what is the Simpsons paradox

Answer 26

correlation present within different groups may disappear or even be reversed when the groups are combined

Question 27

how can we control for confounding factors in the study design

Answer 27

restriction (exclude people)
matching (select people to minimise correlations with confounders)
randomisation

Question 28

how can we control for confounding factors in study analysis

Answer 28

stratification (make comparisons within specific levels of confounders e.g gender, age)
standardisation

Question 29

what is the low birth weight paradox an example of

Answer 29

children of smoking mothers more likely to be LBW, and LBW children have higher infant mortality rate
BUT
LBW children of smoking mothers have LOWER infant mortality rate
because of Simpsons paradox

Question 30

what studies is confounding particularly important for

Answer 30

non-randomised studies