WEEK 5: confounding

Question 1

What is confounding?

Answer 1

biased exposure –> outcome relation, when an effect of an extaneous factor is mistaken for/mixed with actual exposure effect. Threat for internal validity

(e.g. drinking -> diabetes, but people who drink tend do smoke a lot as well)

Question 2

What is effect measure modification?

Answer 2

Measure modification: association which differs across levels of a third factor (e.g. sex is an effect modifier) -> treat for external validity

= fact of nature, life, society or culture

Question 3

How to avoid confounding in observational studies?

Answer 3

1. Prevention in the design
   ● Restriction
   ● Matching
2. During data-analysis
   ● Stratification
   ● Multivariable logistic regression analysis (statistical technique)

Question 4

How does confounding restriction in the case-control design work? Limitations?

Answer 4

Do your study in participants with or without the bias only, not mixed

E.g. in a study of drinking -> disease with confounder: smoking, choose to do your study in non-smokers or smokers only

Limitation: difficulty in attaining required sample size

Question 5

How does confounding matching in the case- control design work? Limitations?

Answer 5

Match cases and controls on the confounding habit/confounder

>

 Select study subjects in such a way that major known confounders are equally distributed across cases and controls

Limitation: may be very difficult to find individuals that match multiple characteristics

Question 6

How can you select study subjects in such a way that major known confounders are equally distributed across cases and controls?

Answer 6

• On individual basis, with one or more controls matched to each case (e.g. same sex, birth date) one-to-one pair matching
• Frequency matching: distribution of confounders is equal in the group of cases and controls
• Select a population of controls such that the overall characteristics of the group match the overall characteristics of the cases.
  ● e.g. if 15% of cases are under age 20, 15% of the controls are also

Question 7

What variables should you not match? Why should you be critical?

Answer 7

do not match on variables that could be in the causal pathway of the exposure-disease association

If you match on a variable you cannot study its impact anymore

Question 8

What is stratification? Limitation?

Answer 8

When both smokers and non-smokers are in your study population and no matching was performed, you adjust for the confounder by dividing into two groups: smokers vs non-smokers.

Odds ratio from whole group is different from small groups

Limitation:
With multiple stratification variables, there will be many empty strata (stratification tables?)

Question 9

What is the “intention-to-treat” principle?

Answer 9

 Effect of a treatment is assessed on the basis of the planned treatment rather than the actual treatment given → All randomized subjects are included in the data-analysis

Question 10

Why would you use the “intention-to-treat” principle?

Answer 10

Eliminating data from non-complint/lost subjects from analysis may yield biased results

Question 11

What is a drawback of “intention-to-treat”?

Answer 11

Including non-compliant subjects may weaken the effect of treatment

Question 12

What is the ‘per protocol’ method?

Answer 12

Compares those who actually received treatment with those who did not

Question 13

Why use the ‘per protocol’ method?

Answer 13

Then you look directly at the effect of treatment

Question 14

Drawback of the per protocol treatment?

Answer 14

Susceptible to bias if those who adhere to treatment differ from those who do not

Question 15

When is double blinding not possible?

Answer 15

e.g. weight loss trials, dietary interventions

Question 15

When is double blinding not possible?

Answer 15

e.g. weight loss trials, dietary interventions

Question 16

Confounding: criteria

Answer 16

1. Risk factor for outcome (among unexposed)
2. Associated with exposure
3. No intermediate step in the causal pathway between exposure and outcome

Question 17

What is the GPC (Good clinical practice) for RCT’s?

Answer 17

international and scientific quality standard for designing, conducting, recording and reporting human trials

Question 18

How is the transparency and high quality of work validated in observational studies?

Answer 18

STROBE

Question 19

What is external validity?

Answer 19

Given sufficient internal validity, does the observed measure of interest also apply to broader populations and settings?

Question 20

What is internal validity?

Answer 20

Does the observed measure of interest reflect the true measure of interest in the population?

Question 21

What is a necessary cause?

Answer 21

Any component cause that is required for the development of the disease

Question 22

What is a component cause?

Answer 22

Each component in a sufficient cause is called a component cause. It is an individual factor that contributes to cause disease

Question 23

What is a sufficient cause?

Answer 23

A complete causal mechanism that produces disease

Question 24

What is a sufficient cause?

Answer 24

A complete causal mechanism that produces disease

Question 25

What is reverse causation?

Answer 25

When the outcome influences the exposure

Question 26

What is a negative association?

Answer 26

A variable increases as the values of the other variable decrease

Question 27

What is the difference between an association and a causal relation?

Answer 27

Association: two or more variables are related
Causation: one variable has a direct influence on the other (cause must precede the effect)

Question 28

In cohort studies, mostly .. is used to avoid confounding

Answer 28

Data-analysis:

stratification, regression analyses