Propensity score

[lsak]

Hello, 

I am trying to see if I can do a propensity score analysis and which kind with my data. 

I am looking at rural and urban communities and whether there is a relationship on where they live and Multi-drug resistant organisms. I am working with covariates as well:Age,Gender,Race/Ethnicity,Education level,Income level,Antibiotic use,Residence,Water,source,Insurance type,Number of people living in household

Can you guys help me with this? 

[webart999ARM]

Propensity score analysis is a statistical method that is used to control for confounders in observational studies. It is commonly used to estimate the effect of a treatment or intervention on an outcome, when there are differences between the treatment and control groups.

To conduct a propensity score analysis in SAS, you will need to use the PROC PSMATCH procedure. This procedure will allow you to estimate the propensity scores for each individual in your study, and then match individuals in the treatment and control groups based on their propensity scores.

To use the PROC PSMATCH procedure, you will need to specify the following:

1. The name of the data set that contains your data.

2. The name of the outcome variable that you are interested in analyzing.

3. The name of the treatment variable that indicates whether an individual received the treatment or not.

4. The name of the variables that you want to include in the propensity score model. These should include the covariates that you mentioned in your question, such as age, gender, race/ethnicity, education level, income level, antibiotic use, residence, water source, insurance type, and number of people living in the household.

Once you have specified these inputs, you can use the PROC PSMATCH procedure to estimate the propensity scores for each individual in your study, and then match individuals in the treatment and control groups based on their propensity scores. This will allow you to control for potential confounders and estimate the effect of the treatment on the outcome of interest.

here is an example of SAS code that you could use to conduct a propensity score analysis using the PROC PSMATCH procedure:

proc psmatch data=my_data;
  /* specify the name of the data set */

  outcome outcome_var;
  /* specify the name of the outcome variable */

  treat treatment_var;
  /* specify the name of the treatment variable */

  var age gender race_ethnicity education_level income_level antibiotic_use residence water_source insurance_type num_people_hh;
  /* specify the name of the variables to include in the propensity score model */

  /* additional options and statements go here */
run;

This code will use the PROC PSMATCH procedure to estimate the propensity scores for each individual in your study, and then match individuals in the treatment and control groups based on their propensity scores. You can then use additional options and statements in the PROC PSMATCH procedure to further analyze your data and interpret the results of the propensity score analysis.

For more information about the PROC PSMATCH procedure and how to use it, you can refer to the SAS documentation. You can also find additional examples and tutorials online that can help you learn how to conduct a propensity score analysis.