Flashcards in Classes #11-#12: Confounding & Effect Modification Deck (15)

Loading flashcards...

1

## What is a confounding variable?

###
A 3rd variable that distorts an observed relationship (association; RR/OR/HR) between the exposure and the outcome (disease).

***Looking at exposure and outcome***

2

## What characteristics must a 3rd variable have to be a confounder?

###
The 3rd variable must be associated (related/correlated) with the exposure and the outcome of interest, yet independent of both...

-but not directly in the hypothesized causal-pathway between the exposure and the outcome

3

## What are the 2 main ways that confounders impact the study?

###
1) Intensity/Magnitude/Strength

-produces an association more or less extreme than true association

2) Direction

-produces an association that moves true association in a positive or negative direction

>towards or away from a null association (RR/OR/HR = 1.0)

4

## How do you go about knowing if confounding is present?

###
Step 1) Calculate crude outcome measure of association (OR/RR) between exposure and outcome

-commonly called "unadjusted" association

Step 2) Re-calculate outcome measure of association (OR/RR) between exposure and outcome while statistically controlling the effects of the confounder

-commonly called "adjusted" association

Step 3) compare the crude vs. adjusted measures of association between the exposure and outcome

-the crude and adjusted estimate (RR/OR) of the association will be different by 20% if there is confounding present

5

## What is the purpose for controlling for confounders?

### To get a more precise (accurate) estimate of the true association between the exposure and the disease/outcome.

6

## What are the ways that confounding can be controlled for?

###
1) Study Design Stage

-randomization (blocked or stratified)

-restriction

-matching

2) Analysis of Data Stage

-stratification

-multivariate statistical analysis

7

## What are the strength and weakness for randomization?

###
Strength:

-with sufficient sample size (N), randomization will likely be successful in serving its purpose (making groups 'equal')

Weakness:

-sample size may not be large enough to control for all known and unknown confounders

-randomization process does not guarantee successful, equal allocation between all intervention groups for all known and unknown confounders

-practical only for interventional studies

8

## What is randomization in controlling for confounders?

### Randomization technique hopefully allocates an equal number of subjects with the known (and unknown) confounders into each intervention group.

9

## What is restriction in controlling for confounding?

### Study participation is restricted to only subjects who do not fall within pre-specified category(ies) of confounder.

10

## What is matching in controlling for confounding?

### Study subjects selected in matched-pairs related to the confounding variable to equally distribute confounder among each of the study groups.

11

## What is stratification in controlling for confounding?

### Statistical analysis of the data by evaluating the association between the exposure and disease within the various strata (layers) within the confounding variable(s).

12

## What is multivariate analysis in controlling for confounding?

### Statistical analysis of the data by mathematically factoring out the effects of the confounding variable(s).

13

## What is an effect modification?

###
a 3rd variable, that when present, modifies the magnitude of effect of an association by varying it within different levels or a 3rd variable (effect modifier).

-if an interaction is present, the researcher must report the measures of association for each strata individually.

>so, unlike confounding, an effect modifying variable should be described and reported at each level of the variable, rather than controlled-for.

14

## How do you figure out if an effect modification is present?

###
Step 1) Calculate crude outcome measure of association between exposure and outcome (OR/RR).

Step 2) Calculate crude outcome measure of association (OR/RR) between exposure and outcome for each strata (layers) of the effect-modifying variable.

Step 3) Compare the stratum-specific measure of associations (for each strata of the rd variable between the exposure and outcome (OR/RR)

- the point estimate (RR/OR) for the association will be different by 20% between the lowest and highest strata (layers) of the effect-modifying variable if there is effect modification (interaction) present.

15