Skekkjur, confounding, aðferðir

Question 1

Definition of bias

Answer 1

Any systematic error in an epidemiological study resulting in an incorrect estimate of association between exposure and risk of disease

Question 2

Hawthorne effect

Answer 2

People act differently if they know they are being watched

Question 3

Differential vs. non differential misclassification

Answer 3

• differential: errors in measurement are one way only (t.d. measurement bias)
• nondifferential: = random: errors in assignment of group happens in more than one direction -¨bias toward the null

Question 4

Confounding factor - skilyrði

Answer 4

1. It is an independent risk factor (cause) of disease
2. It is unevenly distributed among exposed and unexposed
3. It is not on the causal pathway between exposure and disease

Question 5

To be a confounding factor, two conditions must be met:

Answer 5

be associated with exposure
- without being the consequence of exposure
be associated with outcome
- independently of exposure (not an intermediary)

Question 6

Hugtak: accuracy

Answer 6

lack of error

Question 7

Hugtak: bias

Answer 7

systematic error (independent of study size)

Question 8

Hugtak: error

Answer 8

discrepancy between the observed results and the true value

Question 9

Hugtak: precision

Answer 9

absence of random error

Question 10

Hugtak: reliability

Answer 10

repeatability of a measure

Question 11

Hugtak: validity

Answer 11

absence of bias (absence of all error)

Question 12

Validity: internal vs. external

Answer 12

internal: í hvað miklu mæli eru skekkjur fjarverandi
external: í hvað amæli er hægt að heimfræa niðurstöðurnar varðandi rannsóknarhópinn yfir á aðra hópa? -> generalization

Question 13

Ways of dealing with confounding

1. in study design
2. in the analysis

Answer 13

1. • randomization
   • restriction (takmörkun/einskorðun)
   • matching (pörun)
   • collect information on confounders
     2.
   • stratification (lagskipting)
   • standardization (stöðlun)
   • multivariate models (fjölbreytumódel)

Question 14

Measure of confounding

Answer 14

RR(crude) - RR(MH) = effect of confounding

10-20% change from RR(crude) to RR(MH) required to consider confounding

Question 15

Effect modification

Answer 15

stærð sambands milli áreitis og útkomu er breytilegt eftir því hvaða gildi e-r þriðja breyta tekur
- áhættuhlutfallið (RR) breytist..