Bias and Confounding Flashcards Preview

RUSVM Epidemiology 2016 > Bias and Confounding > Flashcards

Flashcards in Bias and Confounding Deck (28):
1

Validity

Relates to the absence of systematic error in a study

2

a valid measure of association

a study will have the same value as the true measure in the source population, except for error due to random variation

3

Bias

The extent to which a measure of association from a study differs from the true measure of association in the source population

4

What causes bias?

Systematic error

5

Internal Validity

the study result is valid with respect to the population under study

6

External Validity

the study result is valid to a wider population beyond the study population and/or source population

7

What is also called generalizability?

External Validity

8

Study Population

the subjects in the study

9

Source Population

Population from which the subjects were drawn

10

target population

population to which we may want to generalize our results

11

Non-differential bias

A bias that equally affects groups

12

Differential bias

a bias that affects one group or more than another

13

What are two general sources of bias?

Selection bias
Information bias

14

Selection bias

error in selection of study subjects

15

Information bias

error in measurement

16

Misclassification bias

Misclassify animals
Animals with no disease are classified as diseased

17

Confounding

an unknown factor distorts the relationship between the exposure and outcome

18

Self Selection bias

studies based upon volunteers- may not be representative of the population as a whole

19

Diagnostic bias

Diagnostic of disease may be influenced by the vet's knowledge of the exposure and their expectation of disease

20

Observer variation

Different observes interpret information differently

21

What are some examples of Information Bias?

Observer variation
Deficiency of tools and technical errors
Recall bias
Reporting Bias

22

Recall bias

Those affected by a disease have a greater sensitivity for recalling exposure

23

Reporting Bias

Individuals with severe disease tend to have more complete medical records

24

Selection Bias Examples

Diagnostic Bias
Self-selection bias

25

Three condition a confounder must meet

Associated with exposure
Associated with the outcome
Not in the causal pathway between the exposure and the outcome

26

How can you reduce confounding variables?

Match the study so the confounding factor is equally respresented
Restrict animals with the confounding factor
Randomize

27

Stratify

Partition the results based on the confounding factor

28

How do you reduce the confounding variable after the study has been completed?

Stratify