Hello, I'm doing a difference-in-difference analysis to show the impact of removing financial incentives for the achievement quality indicators.
If incentives are removed, the indicator is 'treated' (Treated dummy =1); if the incentives aren't removed, the indicator is a control (Treated =0)
The 'denominator' is the number of eligible patients for a quality indicator, the 'numerator' the number of patients actually treated (in each practice).
I'm using the following dataset:
Year Clinic Indicator Numerator Denominator Achievement Treated Time
2016 W95009 AST004 30 35 85.714286 1 0
2016 W95009 COPD002 93 97 95.876289 1 0
2016 W95009 COPD003 282 294 95.918367 1 0
2016 W95009 COPD005 77 77 100 1 0
etc....
I can find the DiD coefficient for the difference in mean clinic 'achievement' (in %), between the treated and controls by doing:
reg Achievement Time Treated Time##Treated
However, I'd also like to look at the mean number of treated patients overall (e.g. sum of the numerators, divided by sum of the denominators) - how would I do that in Stata? e.g. to see whether the overall proportion of treated patients is affected (rather than the mean practice achievement). Many thanks.
Related Posts with Difference-in-difference analysis
Help in creating grouped bar chartHi, I am using Stata version 17 and trying to create a grouped bar chart like this: Array I use t…
How to get the first 3 digit of a variable ?I have a variable receiving 3 types of results: 4-digit number, 3-digit number, and NA For example …
graphicHallo, I am trying to understand and figure out how to actually code/test between one item (Investme…
Skipping and Restarting A LoopHi, I have the following data. I am trying to compute the mean for each occupation category in the …
Missing values in Panel DataHello everyone, I am working with panel data describing company code as id variable and financial y…
Subscribe to:
Post Comments (Atom)
0 Response to Difference-in-difference analysis
Post a Comment