- Tutorial on propensity score analysis
- Measuring treatment effects Read►
- Matching on propensity scores
- Propensity scores and time to events
- Propensity scoring of cost data
MatchIT R Code►
Question 1: For this assignment you can use any statistical software you are
familiar with or use R. You can also use MatchIt or other software
designed to do propensity scoring. The objective is to find
response to citalopram for patients with different types of depression.
These data come from STAR*D experiment conducted by NIMH.
- Read about the study protocol.
- Download data. Use instructor's last name as password.
Must enter password twice.
Data 2010► Data
- Summarize the data. Describe different types of depression
- Select a set of variables and construct a
model to predict presence of a particular type of depression.
- Balance the data to remove the effects of other types of
co-occurring mental health diagnoses on remission/response to
treatment. Show visually that the
propensity scoring has been able to remove the effects of other
predictors of response to citalopram
- Estimate response to citalopram for the particular depression
- Describe what predicts success of citalopram.
- Describe how well the model predicts response to citalopram.
Question 2: In the following assignment, analyze the data provided by
Morgan and Harding and calculate the unconfounded impact of college
education. In this example, "Y is a measure of an individual’s economic
success at age 40, D is the indicator of receipt of a college degree,
and S is a mix of family background and preparedness-for-college variable
that completely accounts for the pattern of self-selection into college
that is relevant for lifetime economic success."
Solution by Morgan and Harding Read►
For additional information (not part of the required reading), please see the following links:
- Example of propensity scoring to understand if racial differences
exist in readmission
- Predictors of response to citalopram Read►
- Does citalopram help anxious depressions
This page is part of the course on
Comparative Effectiveness. Edited by Farrokh Alemi, Ph.D.