advertisement

19 Confounding 2006

70 %
30 %
advertisement
Information about 19 Confounding 2006
Education

Published on January 23, 2008

Author: Miranda

Source: authorstream.com

advertisement

Analytical epidemiology:  Analytical epidemiology Disease frequency Study design: cohorts & case control Choice of a reference group Biases Impact Causal inference Alain Moren, 2006 Stratification - Effect modification - Confounding Matching Significance testing Multivariable analysis Slide2:  Exposure Outcome Third variable Two main complications:  Two main complications (1) Effect modifier (2) Confounding factor - useful information - bias Slide4:  To analyse effect modification To eliminate confounding Solution = stratification stratified analysis Create strata according to categories inside the range of values taken by third variable Effect modifier:  Variation in the magnitude of measure of effect across levels of a third variable. Effect modification is not a bias but useful information Effect modifier Happens when RR or OR is different between strata (subgroups of population) Effect modifier:  Effect modifier To identify a subgroup with a lower or higher risk To target public health action To study interaction between risk factors Vaccine efficacy:  AR NV - AR V VE = ----------------------------- AR NV VE = 1 - RR Vaccine efficacy Vaccine efficacy:  Vaccine efficacy VE = 1 - RR = 1 - 0.28 VE = 72% Vaccine efficacy by age group:  Vaccine efficacy by age group Effect modification:  Effect modification Different effects (RR) in different strata (age groups) VE is modified by age Test for homogeneity among strata (Woolf test) Oral contraceptives (OC) and myocardial infarction (MI):  Oral contraceptives (OC) and myocardial infarction (MI) Case-control study, unstratified data OC MI Controls OR Yes 693 320 4.8 No 307 680 Ref. Total 1000 1000 Physical activity and MI:  Physical activity and MI Effect function:  * * * * * 40 50 60 70 80 1 2 3 4 5 6 Relative risk (RR) of dying from coronary heart disease for smoking physicians, by age groups, England & Wales, RR Age Doll et Hill, 1966 30 20 10 Effect (OR or RR) is a function of the effect modifier Effect function Any statistical test to help us?:  Any statistical test to help us? Breslow-Day Woolf test Test for trends: Chi square Heterogeneity Confounding:  Confounding Distortion of measure of effect because of a third factor Should be prevented Needs to be controlled for Simpson’s paradox:  Simpson’s paradox Slide19:  Second table Slide20:  Day 2, one table only Slide23:  Birth order Age or mother Down syndrom Confounding:  Confounding Exposure Outcome Third variable To be a confounding factor, 2 conditions must be met: Be associated with exposure - without being the consequence of exposure Be associated with outcome - independently of exposure To identify confounding:  To identify confounding Compare crude measure of effect (RR or OR) to adjusted (weighted) measure of effect (Mantel Haenszel RR or OR) Are Mercedes more dangerous than Porsches?:  Are Mercedes more dangerous than Porsches? 95% CI = 1.3 - 1.8 Slide28:  Crude RR = 1.5 Adjusted RR = 1.1 (0.94 - 1.27) Slide29:  Car type Accidents Confounding factor: Age of driver Slide30:  Age Porsches Mercedes < 25 years 550 (55%) 300 (30%) >= 25 years 450 700 Chi2 = 127.9 Age Accidents No accidents < 25 years 370 (44%) 480 >= 25 years 130 (11%) 1020 Chi2 = 270.7 Slide31:  Exposure Outcome Hypercholesterolaemia Myocardial infarction Third factor Atheroma Any factor which is a necessary step in the causal chain is not a confounder Slide32:  Salt Myocardial infarction Hypertension Any statistical test to help us?:  10 - 20 % Any statistical test to help us? When is ORMH different from crude OR ? How to prevent/control confounding?:  How to prevent/control confounding? Prevention Restriction to one stratum Matching Control Stratified analysis Multivariable analysis Mantel-Haenszel summary measure:  Mantel-Haenszel summary measure Adjusted or weighted RR or OR Advantages of MH Zeroes allowed Examples of stratified analysis:  Examples of stratified analysis Slide38:  Effect modifier Belongs to nature Different effects in different strata Simple Useful Increases knowledge of biological mechanism Allows targeting of PH action Confounding factor Belongs to study Weighted RR different from crude RR Distortion of effect Creates confusion in data Prevent (protocol) Control (analysis) How to conduct a stratified analysis:  How to conduct a stratified analysis Perform crude analysis Measure the strength of association List potential effect modifiers and confounders Stratify data according to potential modifiers or confounders Check for effect modification If effect modification present, show the data by stratum If no effect modification present, check for confounding If confounding, show adjusted data If no confounding, show crude data How to define strata:  How to define strata In each stratum, third variable is no longer a confounder Stratum of public health interest If 2 risk factors, we stratify on the different levels of one of them to study the second Residual confounding ? Logical order of data analysis:  Logical order of data analysis How to deal with multiple risk factors: Crude analysis Multivariate analysis 1. stratified analysis 2. modelling linear regression logistic regression Slide42:  A train can mask a second train A variable can mask another variable What happened?:  What happened? Tables % Fitting Hat Colour Hat fitting higher in Table I (83%) vs table II (13%) Blue and red hats not evenly distributed between the 2 tables - table I, 33 % blue - table II, 66 % blue

Add a comment

Related presentations

Related pages

Confounding - Wikipedia, the free encyclopedia

Causal definition. The concept of confounding must be defined, and managed, in terms of the data generating model (as in the Figure above). Specifically ...
Read more

Michael Haneke: The director on his film Hidden | Film ...

The enigmatic French thriller is the water cooler film of the year, confounding audiences with its ambiguous plot. Jason Solomons asked the director to ...
Read more

Talk:Confounding - Wikipedia, the free encyclopedia

Talk:Confounding WikiProject ... 19, 13 August 2007 (UTC) ... 9 May 2006 (UTC) No, Confounding factor deals with the statistical use of the term, ...
Read more

PARTICULATE AIR POLLUTION AND INFANT MORTALITY

October 19, 2006 Air Resources Board ... n Controlled for maternal confounding factors “Fine Particulate Matter (PM 2.5) Air Pollution and Selected
Read more

Lecture 11: Blocking and Confounding in 2 - stat.purdue.edu

Spring 2006 Lecture 11: Blocking and Confounding in 2 k design Montgomery: Chapter 7 March, 2006 Page 1. Statistics 514: Design and Analysis of Experiments ...
Read more

The Confounding Claims of the Cross19mar06

The Confounding Claims of the Cross John 2: 13-22 & 1 Corinthians 1: 18-25 Third Sunday in Lent/ 19th March 2006 It’s 1943. Antwerp, Belgium.
Read more

Confounding and Control - Friday, May 05, 2006

Confounding and Control, Friday, May 05, 2006, 14:45 Galton Lecture Theatre (Room 115), 1-19 Torrington Place, London WC1, http:// , Contact: ...
Read more

Confounding and Control of Confounding in Nonexperimental ...

Confounding and Control of Confounding in Nonexperimental Studies of Medications in Patients With CKD. ... 2006; 35: 337–344 ... 19 Greenland, S. and ...
Read more

Confounding and Control - Demographic Research

Confounding and control . ... a confounding factor is far from universal and there exist different methodological ... and Kromhout 2006), ...
Read more