Creating different groups having different exposures

ak2011 — Mon, 17 Aug 2020 02:52:25 GMT

Hello,

1. I am creating 2 different groups; refgroupA and refgroup B.I have been able to create refgroupA but I would need assistance for refgroup B. Refgroup B in the results(attached) is incorrect.

There are 3 agents (a1exp,a2exp and a3exp), their levels and categories of exposures. 0=unexposed, 1 is exposed with levels 1=high and 2=low.

Refgroup A are ids(cases and controls(cont) not exposed to agent 1(a1exp): ie. a single reference group used for the evaluated agents (a2exp and a3exp).

Refgroup B are ids exposed to agent 1(a1exp) but not to agent under analysis ie. a2exp and a3exp.[Actually there are many agents but I am using only 3 of them as a test].

The idea of these refgroups is to check of the ORs are lower when using refgroup A in the analysis, in comparison with ORs obtained with regroup B.

refgroupB created is not right because ids exposed to a1exp not a2exp are os2, os6 and os12;

Ids exposed to a1exp not a3exp are os4,os6 and os13. Help needed here, please for the refgroupB.

2. Help also needed when using refgroupA and refgroupB in analysis to be able to fill in the OR table (Table 3) attached.

? After creating the refgroups, do I have to merge the regroup results with original data(expinfo) to be able to fill in the table? I have no clue. Please help.

Thanks. ak.


 data expinfo;
input id$ 1-4  a1exp 6-7 a1exp_level$ 8-17  a1exp_cat 19-20  
a2exp 26-27 a2exp_level$ 28-32 a2exp_cat 39-40 a3exp 46-47 a3exp_level$ 48-50 a3exp_cat 59-60  
lung$ 68-72;
datalines;
os1  0 unexposed  0      1 high       1      0 unexposed  0        case  
os2  1 low        2      0 unexposed  0      1 low        2        case  
os3  0 unexposed  0      0 unexposed  0      0 unexposed  0        case
os4  1 high       1      1 low        2      0 unexposed  0        cont 
os5  0 unexposed  0	     0 unexposed  0      1 high       1        cont 	   
os6  1 low        2      0 unexposed  0	     0 unexposed  0        case
os7  1 low        2      1 high       1      1 low        2        case  
os8  0 unexposed  0      0 unexposed  0      0 unexposed  0        case  
os9  0 unexposed  0      0 unexposed  0      0 unexposed  0        cont
os10 1 high       1      1 low        2      1 low        2        cont 
os11 0 unexposed  0	     0 unexposed  0      1 high       1        case 
os12 1 low        2      0 unexposed  0	     1 high       1        case
os13 1 high       1      1 low        2      0 unexposed  0        case  
os14 0 unexposed  0      0 unexposed  0      0 unexposed  0        cont  
os15 0 unexposed  0      0 unexposed  0      0 unexposed  0        cont
os16 1 high       1      1 low        2      1 high       1        cont 
os17 0 unexposed  0	     1 high       1      1 low        2        cont 
os18 0 unexposed  0      0 unexposed  0	     0 unexposed  0        cont
;
run;



proc freq data=expinfo; where a1exp=0;
     tables lung;
title 'Table 1:Subjects unexposed to a1exp';
run;

proc sql;
create table t as 
  select
    id, a1exp, a2exp,a3exp,lung,
    a1exp=0 as refgroupA,
    
   a1exp ne 0 and a2exp=0 and a1exp ne 0 and a3exp=0 as refgroupB
 
     from expinfo 
     ;
     quit;
 
proc print data=t; 
   title 'Table 2: original variables and refgroupA and refgroupB;
   refgroupB is not correct';
run;

1 OPTIONS NONOTES NOSTIMER NOSOURCE NOSYNTAXCHECK;

74 data expinfo;

75 input id$ 1-4 a1exp 6-7 a1exp_level$ 8-17 a1exp_cat 19-20

76 a2exp 26-27 a2exp_level$ 28-32 a2exp_cat 39-40 a3exp 46-47 a3exp_level$ 48-50 a3exp_cat 59-60

77 lung$ 68-72;

78 datalines;

NOTE: The data set WORK.EXPINFO has 18 observations and 11 variables.

NOTE: DATA statement used (Total process time):

real time 0.01 seconds

cpu time 0.02 seconds

97 ;

98 run;

100

101

102 proc freq data=expinfo; where a1exp=0;

103 tables lung;

104 title 'Table 1:Subjects unexposed to a1exp';

105 run;

NOTE: There were 10 observations read from the data set WORK.EXPINFO.

WHERE a1exp=0;

NOTE: PROCEDURE FREQ used (Total process time):

real time 0.26 seconds

cpu time 0.25 seconds

106

107 proc sql;

108 create table t as

109 select

110 id, a1exp, a2exp,a3exp,lung,

111 a1exp=0 as refgroupA,

112

113 a1exp ne 0 and a2exp=0 and a1exp ne 0 and a3exp=0 as refgroupB

114

115 from expinfo

116 ;

NOTE: Table WORK.T created, with 18 rows and 7 columns.

117 quit;

NOTE: PROCEDURE SQL used (Total process time):

real time 0.01 seconds

cpu time 0.01 seconds

118

119 proc print data=t;

120 title 'Table 2: original variables and refgroupA and refgroupB;

121 refgroupB is not correct';

122 run;

NOTE: There were 18 observations read from the data set WORK.T.

NOTE: PROCEDURE PRINT used (Total process time):

real time 0.35 seconds

cpu time 0.35 seconds

123

124 OPTIONS NONOTES NOSTIMER NOSOURCE NOSYNTAXCHECK;

136

topic Creating different groups having different exposures in SAS Studio

Creating different groups having different exposures