Loading in 5 sec....

Conditional Logistic Regression for Matched Data HRP 261 02/25/04 reading: Agresti chapter 9.2PowerPoint Presentation

Conditional Logistic Regression for Matched Data HRP 261 02/25/04 reading: Agresti chapter 9.2

- By
**Jims** - Follow User

- 379 Views
- Uploaded on

Download Presentation
## PowerPoint Slideshow about 'Conditional Logistic Regression for Matched Data HRP 261 02' - Jims

**An Image/Link below is provided (as is) to download presentation**

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.

- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -

Presentation Transcript

### Conditional Logistic Regression for Matched DataHRP 261 02/25/04reading: Agresti chapter 9.2

### Part II: Rater agreement: controls. Typically this is just the value 0.Cohen’s KappaAgresti, Chapter 9.5

Recall: Matching

- Matching can control for extraneous sources of variability and increase the power of a statistical test.
- Match M controls to each case based on potential confounders, such as age and gender.

Recall: Agresti example, diabetes and MI

Match each MI case to an MI control based on age and gender.

Ask about history of diabetes to find out if diabetes increases your risk for MI.

No Diabetes

9

37

16

82

MI controls

MI cases

46

Diabetes

No diabetes

98

25

119

144

=the probability of observing a case-control pair with only the case exposed

=the probability of observing a case-control pair with only the control exposed

P(“favors” case/discordant pair) =

No Diabetes

9

37

16

82

MI controls

MI cases

46

Diabetes

No diabetes

98

25

119

144

odds(“favors” case/discordant pair) =

Logistic Regression for Matched Pairs option 1:the logistic-normal model

- Mixed model; logit=i+x
- Where irepresents the “stratum effect”
- (e.g. different odds of disease for different ages and genders)
- Example of a “random effect”

- Allow i’s to follow a normal distribution with unknown mean and standard deviation
- Gives “marginal ML estimate of ”

option 2: Conditional Logistic Regression

The conditional likelihood is based on….

The conditional probability (for pair-matched data):

P(“favors” case/discordant pair) =

The Conditional Likelihood: each discordant stratum (rather than individual) gets 1 term in the likelihood

Note: the marginal probability of disease may differ in each age-gender stratum, but we assume that the (multiplicative) increase in disease risk due to exposure is constant across strata.

Each age-gender stratum has the same baseline odds of disease; but these baseline odds may differ across strata

The conditional likelihood=Conditional Logistic Regression disease; but these baseline odds may differ across strata

Example: MI and diabetes disease; but these baseline odds may differ across strata

Conditional Logistic Regression disease; but these baseline odds may differ across strata

Example: disease; but these baseline odds may differ across strataPrenatal ultrasound examinations and risk of childhood leukemia: case-control study BMJ 2000;320:282-283

- Could there be an association between exposure to ultrasound in utero and an increased risk of childhood malignancies?
- Previous studies have found no association, but they have had poor statistical power to detect an association.
- Swedish researchers performed a nationwide populationbased case-control study using prospectively assembled data onprenatal exposure toultrasound.

Example: disease; but these baseline odds may differ across strataPrenatal ultrasound examinations and risk of childhood leukemia: case-control study BMJ 2000;320:282-283

- 535 cases: all children born and diagnosed as having myeloid leukemia between 1973 and 1989 in Swedish registers of birth, cancer, and causesof death.
- 535 matched controls: 1 control was randomly selectedfor each case from the Swedish Birth Registry, matched by sex and year and month of birth.

Ultrasound disease; but these baseline odds may differ across strata

No Ultrasound

Myeloid leukemia controls

Leukemia cases

200

Ultrasound

No ultrasound

335

215

320

535

115

85

100

235

But this type of analysis is limited to single dichotomous exposure…

- Used conditional logistic regression to look at dose-response with number of ultrasounds:
- Results:
- Reference OR = 1.0; no ultrasounds
- OR =.91 for 1-2 ultrasounds
- OR=.64 for >=3 ultrasounds
- Conclusion: no evidence of a positive association between prenatal ultrasound and childhood leukemia; even evidence of inverse association (which could be explained by reasons for frequent ultrasound)

Extension: 1:M matching dose-response with number of ultrasounds:

- Each term in the likelihood represents a stratum of 1+M individuals
- More complicated likelihood expression! See: 02/02/04 lecture

Conditional Logistic Regression in SAS: Please read Ray’s slides at:

- http://www.stanford.edu/class/hrp223/2003/Lecture15/Lecture15_223_2003.ppt
Available here:

-SAS tips, explanations and code

-SAS macro that generates automatic logit plots (under “Lecture 15” at: http://www.stanford.edu/class/hrp223/) to check if predictor is linear in the logit.

Put the values in the IsOUTCOME variable here that are the controls. Typically this is just the value 0.

This is the switch requesting a m:n CLR.

This is the m:n matching variable.

M:N Matching Syntax- The basic syntax is shown here.
procphregdata=BLAH;

model WEIRD*IsOUTCOME(Censor_v)= PREDICTORS /ties=discrete;

strata STRATA_VARS;

run;

Courtesy: Ray Balise

Cohen’s Kappa controls. Typically this is just the value 0.

Actual agreement = sum of the proportions found on the diagonals.

Cohen: Compare the actual agreement with the “chance agreement” (which depends on the marginals).

Normalize by its maximum possible value.

Rating by supervisor 2 controls. Typically this is just the value 0.

Rating by supervisor 1

Authoritarian

Democratic

Permissive

Totals

Authoritarian

17

4

8

29

Democratic

5

12

0

17

Permissive

10

3

13

26

Totals

32

19

21

72

Ex: student teacher ratingsExample: student teacher ratings controls. Typically this is just the value 0.

Null hypothesis: Kappa=0 (no agreement beyond chance)

Interpretation: achieved 36.2% of maximum possible improvement over that expected by chance alone

Example: student teacher ratingsNull hypothesis: Kappa=0 (no agreement beyond chance)

Download Presentation

Connecting to Server..