logistic regression analysis of matched case-control data- part 2

Logistic Regression Logistic Regression Analysis of Matched Analysis of Matched Case-Control Data-Case-Control Data-

Part 2Part 2

EVW LOGISTIC MODEL FOR MATCHED DATA

Logit P(X) = + E + 1iV1i + 2iV2i

+ EkWk

OR = exp [ + kWkk

]

USING SAS’S PHREG PROCEDURE TO CARRY

OUTCONDITIONAL ML

ESTIMATION OF MATCHED DATA USING LOGISTIC

REGRESSON

USING STATA’S CLOGIT PROCEDURE TO CARRY

OUTCONDITIONAL ML

ESTIMATION OF MATCHED DATA USING LOGISTIC

REGRESSON

. clogit case est gall [fweight=wgt], strata(stratum)

. clogit case est gall [fweight=wgt], strata(stratum) or

. clogit case est [fweight=wgt], strata(stratum)

. clogit case est [fweight=wgt], strata(stratum) or

SPSS

• SPSS does not perform conditional logistic regression except in the special case in which there is only one case per stratum, with one or more controls.

• The SPSS survival analysis procedure COXREG can be used to obtain coefficient estimates equivalent to running a conditional logistic regression.

• The process is similar to that demonstrated for SAS with PROC PHREG and the GALL dataset, although SAS is not limited to the special case.

SPSS (continued)

As with SAS, a time variable must be created in the data, coded to indicate that all cases had the event at the same time, and all controls were censored at a later time.

In the GALL dataset, this time variable is named survt( SPSS dataset Is called gall.sav.)

SPSS (continued)

The program code for the gall.sav dataset (assuming one line of data for each subject, I.e., no weight variable) :

GET FILE='A:gall.sav'.COXREG survt /STATUS=case(1) /STRATA=stratum /METHOD=ENTER est gall /CRITERIA=PIN(.05) POUT(.10) ITERATE(20) .

The model statement contains the time variable (survt) followed by a backslash and the case status variable (case) with the value for cases (1) in parentheses.