Cochran-Mantel-Haenszel Statistics using Proc FREQ
Page 1 of 1
Cochran-Mantel-Haenszel Statistics using Proc FREQ
The Cochran-Mantel-Haenszel test statistic is one of the many statistical measures used in case control studies. To obtain this useful statistic, which compares the odds ratio, you can declare the CMH option on the TABLE statement in PROC FREQ.
General Syntax
PROC FREQ DATA = <data>;
TABLE X*Y/CMH;
RUN;
Example:
The data set Migraine contains hypothetical data for a clinical trial of migraine treatment. Subjects of both genders receive either a new drug therapy or a placebo. Their response to treatment is coded as 'Better' or 'Same'. The data are recorded as cell counts, and the number of subjects for each treatment and response combination is recorded in the variable Count.
data Migraine;
input Gender $ Treatment $ Response $ Count @@;
datalines;
female Active Better 16 female Active Same 11
female Placebo Better 5 female Placebo Same 20
male Active Better 12 male Active Same 16
male Placebo Better 7 male Placebo Same 19
;
Run;
The following PROC FREQ statements create a multiway table stratified by Gender, where Treatment forms the rows and Response forms the columns. The CMH option produces the Cochran-Mantel-Haenszel statistics. For this stratified table, estimates of the common relative risk and the Breslow-Day test for homogeneity of the odds ratios are also displayed. The NOPRINT option suppresses the display of the contingency tables.
proc freq data=Migraine;
tables Gender*Treatment*Response / cmh;
weight Count;
title 'Clinical Trial for Treatment of Migraine Headaches';
run;
For a stratified table, the three CMH statistics displayed in Output above test the same hypothesis. The significant -value (0.004) indicates that the association between treatment and response remains strong after adjusting for gender.
The CMH option also produces a table of relative risks, as shown in next Output. Because this is a prospective study, the relative risk estimate assesses the effectiveness of the new drug; the "Cohort (Col1 Risk)" values are the appropriate estimates for the first column (the risk of improvement). The probability of migraine improvement with the new drug is just over two times the probability of improvement with the placebo.
The large -value for the Breslow-Day test (0.2218) in Output 3.7.3 indicates no significant gender difference in the odds ratios.
General Syntax
PROC FREQ DATA = <data>;
TABLE X*Y/CMH;
RUN;
Example:
The data set Migraine contains hypothetical data for a clinical trial of migraine treatment. Subjects of both genders receive either a new drug therapy or a placebo. Their response to treatment is coded as 'Better' or 'Same'. The data are recorded as cell counts, and the number of subjects for each treatment and response combination is recorded in the variable Count.
data Migraine;
input Gender $ Treatment $ Response $ Count @@;
datalines;
female Active Better 16 female Active Same 11
female Placebo Better 5 female Placebo Same 20
male Active Better 12 male Active Same 16
male Placebo Better 7 male Placebo Same 19
;
Run;
The following PROC FREQ statements create a multiway table stratified by Gender, where Treatment forms the rows and Response forms the columns. The CMH option produces the Cochran-Mantel-Haenszel statistics. For this stratified table, estimates of the common relative risk and the Breslow-Day test for homogeneity of the odds ratios are also displayed. The NOPRINT option suppresses the display of the contingency tables.
proc freq data=Migraine;
tables Gender*Treatment*Response / cmh;
weight Count;
title 'Clinical Trial for Treatment of Migraine Headaches';
run;
For a stratified table, the three CMH statistics displayed in Output above test the same hypothesis. The significant -value (0.004) indicates that the association between treatment and response remains strong after adjusting for gender.
The CMH option also produces a table of relative risks, as shown in next Output. Because this is a prospective study, the relative risk estimate assesses the effectiveness of the new drug; the "Cohort (Col1 Risk)" values are the appropriate estimates for the first column (the risk of improvement). The probability of migraine improvement with the new drug is just over two times the probability of improvement with the placebo.
The large -value for the Breslow-Day test (0.2218) in Output 3.7.3 indicates no significant gender difference in the odds ratios.
pallav- Posts : 98
Join date : 2012-03-14
Location : Ahmedabad
Similar topics
» order option in PROC FREQ to find out duplicate records
» What is sampling in statistics..?
» Descriptive statistics Problem
» Basic statistics ( Interactive learning)
» Collapsing across observations using proc sql
» What is sampling in statistics..?
» Descriptive statistics Problem
» Basic statistics ( Interactive learning)
» Collapsing across observations using proc sql
Page 1 of 1
Permissions in this forum:
You cannot reply to topics in this forum
|
|
Mon Jun 10, 2013 2:57 pm by Admin
» SAS Enterprise Guide: Explore Data before you start Analysing
Mon Jun 10, 2013 2:53 pm by Admin
» Predictive Analytics
Fri May 31, 2013 3:30 pm by Admin
» Creating Frequency Report and Creating User Defined styles in SAS Add-In for Microsoft Office
Tue Oct 23, 2012 2:11 pm by pallav
» Prompted Reports in BI Dashboard
Thu Oct 11, 2012 4:54 pm by kushal
» Free SAS Webinar : Getting Started with SAS Hash Object
Wed Oct 10, 2012 6:08 pm by pallav
» Which SAS procedure changes the name of a permanent format for a variable stored in a SAS data set?
Tue Oct 02, 2012 12:47 am by rajin_kumar@yahoo.com
» What is written to SAS Log?
Tue Oct 02, 2012 12:39 am by rajin_kumar@yahoo.com
» SAS Portal and BI Dashboard: Customizing the Tabs
Tue Oct 02, 2012 12:33 am by kushal