Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
Contents
4.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . 10
4.3 Confounding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
4.3.1 Simpsons Paradox . . . . . . . . . . . . . . . . . . . . . . . 11
4.3.2 Definition of Confounding . . . . . . . . . . . . . . . . . . . 12
4.3.3 Diagnosing Confounding
. . . . . . . . . . . . . . . . . . . 15
CONTENTS
4.6 Graphical Methods for Visualizing Effect Modification, Confounding, and Precision . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.6.1 Effect Modifiers . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.6.2 Confounders . . . . . . . . . . . . . . . . . . . . . . . . . . 26
4.6.3 Precision Variables . . . . . . . . . . . . . . . . . . . . . . . 28
Chapter 4
4.1
Overview
Scientific questions
Most often scientific questions are translated into comparing the distribution of some response variable across groups of interest
Groups are defined by the predictor of interest (POI)
Categorical predictors of interest: Treatment or control, knockout or
wild type, ethnic group
Continuous predictors of interest: Age, BMI, cholesterol, blood pressure
Covariates other than the Predictor of Interest are included in the model as...
Effect modifiers
Confounders
Precision variables
Not necessarily mutually exclusive
4.2
Effect Modification
The association between the Response and the Predictor of Interest differs
in strata defined by the effect modifier
Statistical term: Interaction between the effect modifier and the POI
Effect modification depends on the measure of effect that you choose
Choice of summary measure: mean, median, geometric mean, odds,
hazard
Choice of comparisons across groups: differences, ratios
4.2.1
-13 0
0.90 1.00
-13 -9
0.90 0.93
-0.29 -0.43
0.71 0.72
Median
Women Men
0.7 1.1
1.0 1.5
-0.3 -0.4
0.70 0.73
Proportion
Women Men
No Smoke
0.10 0.16
Smoke
0.16 0.26
Diff
Ratio
Odds
Women Men
0.03 0.19
0.19 0.35
-0.06 -0.10
0.62 0.62
-0.16 -0.16
0.16 0.54
Odds
None CVD
0.02 0.50
0.04 1.00
0.13 0.02
3.60 1.08
0.17 0.03
4.17 1.11
Odds
Women Men
0.19 0.33
0.19 0.35
0.00 -0.01
1.00 0.96
0.00 -0.02
1.00 0.95
10
4.2.2
11
If effect modification exists, an analysis adjusting only for the third variable
(but no interaction) will tend toward a weighted average of the stratum specific effects
Hence, an association in one stratum and not the other will make an
adjusted analysis look like an association (provide the sample size is
large enough)
4.3
Confounding
4.3.1
Simpsons Paradox
12
4.3.2
Definition of Confounding
The association between a predictor of interest and the response is confounded by a third variable if
The third variable is associated with the predictor of interest in the sample, AND
The third variable is associated with the response
Causally (in truth)
In groups that are homogeneous with respect to the predictor of interest
Not in the causal pathway of interest
We must consider our belief about the causal relationships among the measured variables
There is no statistical test for causality
Inference about causation comes only from the study design
BUT, consideration of the causal relationships helps us to decide which
statistical questions to answer
Classic confounder
13
Causal pathways
A variable in the causal pathway of interest is not a confounder
We would not adjust for such a variable
If we did adjust, we would lose ability to detect associations between
the POI and the outcome
Example: Second hand smoke (POI), stunted growth (confounder),
FEV1 (outcome)
Scientific question is about the impact of smoking on lung function
Stunted growth addresses lung anatomy, not lung function, which
we dont care about it
Directed Acyclyic Graph:
14
15
Greenland,
Pearl, and Robins. Causal Diagrams for Epidemiologic Research. Epidemiology. (1999)
http://www.jstor.org/stable/3702180
4.3.3
Diagnosing Confounding
16
Example: Suppose you have a sample where 50% of the subjects die
What is the variability?
We can reduce this variability by changing p, the probability of death
17
4.4
Precision Variables
4.4.1
Overview
=
Typically: se()
18
V
n
se()
4.4.2
V = (r + 1)( r1 + 22 )
=
se()
V
n
12
n1
22
n2
19
Might control for some variable in order to decrease the within group variability
Restrict population sampled
Standardize ancillary treatments
Standardize measurement procedure
Y |X
V = Var
(X)
=
se()
v
u
u
t
Y2 |X
nVar(X)
Independent continuous outcome associated with covariate (X) and precision variable (W )
ind Yi |Xi , Wi (0 + 1 Xi + 2 Wi , Y2 |X,W ), i = 1, . . . , n
= 1 , = 1 from LS regression
Y2 |X,W
V = Var(X)(1r2 )
X,W
=
se()
v
u
u
t
20
Y2 |X
2
nVar(X)(1rX,W
)
V = (r + 1)( r1 + 22 )
=
se()
V
n
12
n1
22
n2
When analyzing odds (a nonlinear function of the mean), adjusting for precision variables results in more extreme estimates
21
Odds = 1pp
Odds using average of stratum specific p is not the average of stratum specific odds
4.5
2.00
2.00
1.75
Diagnosing Confounding
4.5.1
In order of
22
Adjustment for covariates changes the question being answered by the statistical analysis
Adjustments can be made to isolate associations that are of particular
interest
When consulting with a scientist, it is often difficult to decide whether the
interest in an additional covariate is due to confounding, effect modification, or precision
The distinction is important because I tend to treat these variable differently in the analysis
Often the scientific question dictates inclusion of particular predictors
Predictor of interest: The scientific parameter of interest can be modeled by multiple predictors (e.g. dummy variables, polynomials, splines)
Effect Modifiers: The scientific question relates to the detection of effect modification
Confounders: The scientific question may be state in terms of adjusting for known (or suspected) confounders
4.5.2
Confounder Detection
Unanticipated confounding
Some times we must explore our data to assess whether our results were
confounded by some variable
Goal is to assess the independent effect of the predictor of interest on
the outcome
23
Confounders
Variables (causally) predictive of the outcome, but not in the causal pathway
Best method: Think about the scientific problem beforehand (perhaps
draw DAG)
Using data, often assessed in the control group
Variables associated with the predictor of interest in the sample
Note that statistical significance is not relevant because this tells us
about associations in the population
Detection of confounding ultimately must rely on our best knowledge
about the possible scientific mechanisms
4.6
24
Precision variables
4.6.1
Effect Modifiers
25
Effect Modification
4
Unadjusted
2
1
0
0.0
0.2
0.4
0.6
X
0.8
1.0
0.0
0.2
0.4
0.6
X
0.8
1.0
4.6.2
Confounders
26
27
Unadjusted
Adjusted Confounding
0.0
0.2
0.4
0.6
X
0.8
1.0
0.0
0.2
0.4
0.6
X
0.8
1.0
4.6.3
28
Precision Variables
29
0.0
0.2
0.4
10
0.6
X
10
15
15
Adjusted Precision
20
20
Unadjusted Precision
0.8
1.0
0.0
0.2
0.4
0.6
X
0.8
1.0