Bias and Confounding

Question 1

What is selection bias?

Answer 1

When a convenient group are used (first sheep, obesity in obesity clinic)
When individuals are systematically different to those not selected

Question 2

What is information bias?

Answer 2

Occurs when misclassification of animals and diseased and non-diseased or inaccurate measurements of study factor

Question 3

What effects information bias?

Answer 3

Test sensitivity and specificity

Question 4

What are the different examples of selection bias?

Answer 4

Healthcare bias
Spectrum bias- cases with certain signs more detected
Volunteer bias- certain people more likely to volunteer
Non-response bias- people who respond are different

Question 5

What is interpretive bias?

Answer 5

Occurs when different emphasis is given to same evidence when results are evaluated

Question 6

What are different exampled of interpretive bias?

Answer 6

Confirm bias- selectively reporting information information that agrees with your prior convictions
Rescue bias- discounting data by finding faults with a study and being less harsh on studies you agree with

Question 7

What is confounding?

Answer 7

Occurs when the apparent association between a study factor and the outcome is distorted by the effect of this variable, which is associated with both the outcome and the study factor
e.g carrying matches causes lung cancer

Question 8

How can confounding be controlled in study design?

Answer 8

Restriction- only study large herds
Matching- for each large/small herd with a fan include a large/small herd without
Randomisation- randomly select herd that have or not have fans

Question 9

How can confounding be controlled in analysis?

Answer 9

Restriction- only analyse large herds
Stratified analysis- analyse large herd and small herds separately
Multivariable analysis- to control for size of herd while examining effect of fans