I am trying to run the following command:
psmatch2 intervention1 age1 sex_n1 tearsize_n1 bmi1 smoking1 workerscomp1 diabetes1 pre_ases1, out(revision1)
Where intervention1 is my study intervention (yes/no variable) and revision1 is my outcome (yes/no variable). The control and treatment arm of the study both have 47 patients pre-match.
Age1, bmi1, and pre_ases1 are all continuous variables with measurable means.
The rest of the variables are either yes/no or multiple choice.
- Sex_n1 (male/female converted to 1/0)
- tearsize_n1 (small/medium/large/massive converted to 1/2/3/4/)
- smoking1 (no/yes converted to 0/1)
- workerscomp1 (no/yes converted to 0/1)
- diabetes1 (no/type1/type2 converted to 0/1/2)
After I run the psmatch2 command, I get a total of 83 observations (40 control, 43 treatment). I want to be able to make an excel table with the # of patients in each cohort that have each variable that I matched on. For example, post-match 5 patients in the treatment arm are smokers and 4 in the control arm are smokers, etc.
After I run the command:
pstest age1 sex_n1 tearsize_n1 bmi1 smoking1 workerscomp1 diabetes1 pre_ases1
I get the mean and p-values for all variables. This is fine for continuous variables like age1, bmi1, and pre_ases1. But how do I get individual counts for each of the yes/no or multiple choice variables so I can say how many of each were included in the final match? For example I want to be able to say that post-match, the breakdown of the tear sizes between each cohort was x small tears, y medium tears, z large tears, and xx massive tears. How can I get this information?
psmatch2 intervention1 age1 sex_n1 tearsize_n1 bmi1 smoking1 workerscomp1 diabetes1 pre_ases1, out(revision1)
Where intervention1 is my study intervention (yes/no variable) and revision1 is my outcome (yes/no variable). The control and treatment arm of the study both have 47 patients pre-match.
Age1, bmi1, and pre_ases1 are all continuous variables with measurable means.
The rest of the variables are either yes/no or multiple choice.
- Sex_n1 (male/female converted to 1/0)
- tearsize_n1 (small/medium/large/massive converted to 1/2/3/4/)
- smoking1 (no/yes converted to 0/1)
- workerscomp1 (no/yes converted to 0/1)
- diabetes1 (no/type1/type2 converted to 0/1/2)
After I run the psmatch2 command, I get a total of 83 observations (40 control, 43 treatment). I want to be able to make an excel table with the # of patients in each cohort that have each variable that I matched on. For example, post-match 5 patients in the treatment arm are smokers and 4 in the control arm are smokers, etc.
After I run the command:
pstest age1 sex_n1 tearsize_n1 bmi1 smoking1 workerscomp1 diabetes1 pre_ases1
I get the mean and p-values for all variables. This is fine for continuous variables like age1, bmi1, and pre_ases1. But how do I get individual counts for each of the yes/no or multiple choice variables so I can say how many of each were included in the final match? For example I want to be able to say that post-match, the breakdown of the tear sizes between each cohort was x small tears, y medium tears, z large tears, and xx massive tears. How can I get this information?
Comment