Lecture 7 (Bias + Confounding)

Question 1

Bias (def)

Answer 1

Systematic errore in design/conduct of study –> mistaken asso
-Major prob in obsv epi studies

Question 2

Selection bias (2 main reasons, major issue in ____)

Answer 2

• Error resulting f/selection procedures or factors that influence participation
• Major issue in case-control studies

Question 3

Info Bias (def)

Answer 3

-Error due to collection of incorrect info about subjects (misclassification)

Question 4

Compensating for selection bias (case-control)

Answer 4

-Equalizing selection bias by selecting cases/control using same selection process

Question 5

Minimizing selection bias

Answer 5

• Carefully consider selection criteria

- Maximize participation rate

Question 6

Selection bias and cohort studies (when is likely/unlikely)

Answer 6

• Unlikely w/internal comp grps
• External comp grps more prone (diff pops, might have diff risks aside f/exposure)
• Healthy workers effect: issue when general pop = external comp grp, healthier
• Loss to follow is an issue

Question 7

Nondiff exposure/disease misclassification (bias towards of away)

Answer 7

Biases asso towards Ho (no asso)

Question 8

Diff misclassification (related to both exposure and disease)

Answer 8

-Biases towards/away Ho (depends)

Question 9

Nondiff exposure missclass (cohort)

Answer 9

• Biases RR toward Ho if exposure misscl is unrelated to the future development of disease
• Not likely
• Won’t bias if happens to same degree among those who develop/don’t

Question 10

Disease missclass (cohort)

Answer 10

• Issue when based on self-report, esp for more subjective things
• Concern if those collecting data are aware of subjects’ exposures

Question 11

Exposure missclass in case-controls (what’s the effect)

Answer 11

• Important source of this bias
• Bias towards null if missclass unrelated to disease stat (non diff)
• Bias either direction if it missclass depends on disease status (diff)

Question 12

Types of info bias

Answer 12

• Recall bias
• Reporting bias
• Observer bias (sub: interviewer bias, abstractor bias)

Question 13

Reducing bias

Answer 13

• Carefully define things
• Choose valid measurement methods
• Train ppl
• Quality control
• Maximize participation
• Apply methods in same way/detail to everyone
• Don’t improve quality of data in one but not other

Question 14

Detection bias

Answer 14

=surveillance bias

-When close med surveillance –> higher prob of detection of disease

Question 15

Most common confounder

Answer 15

Age

Question 16

Difference btwn confounding and bias

Answer 16

confounding is real, bias is not

Question 17

Checking for confounding

Answer 17

• Compare unadjusted and adjusted ORs (weighted average of stratum-specific ORs)
• If difference is more than 10%: confounding

Question 18

Confounding (def)

Answer 18

• Distortion of asso due to another extraneous exposure associated w/both disease and exposure
• Cannot be intermediate variable in causal pathway though

Question 19

Positive confounding

Answer 19

-Overestimation of asso

Question 20

Negative confounding

Answer 20

-Underestimation of asso

Question 21

Qualitative confounding

Answer 21

-Inversion of direction of asso

Question 22

Prevention of confounding (design) (3)

Answer 22

• Randomization
• Restriction (single category of confounding exposure)
• Matching

Question 23

Controlling for confounding in analyses

Answer 23

Stratification methods:

• Direct adj (cohort)
• Indirect (occupational retro cohort)
• Mantel haenszel method (most common, case control/cohort)

Multivariate models:

• Logistic reg (case-control)
• Cox proportional hazards model (cohort)
• Poisson regression (cohort)

Question 24

Limitations of controlling for confounding using stratification

Answer 24

• Can only adjust for 1-2 w/small # of categories each
• Or else data would be sparse
• Only categorical variables

Question 25

Residual confounding (def and sources)

Answer 25

• Confounding that remains after adjustment

- Sources: categories in adj too broad; variable was imperfect surrogate