Help with multi series interrupted time series analysis with two interventions

sms1891 — Fri, 16 Feb 2024 15:42:38 GMT

Hi experts,

I have a monthly rate data for two variables, community onset rate of infections and hospital-onset rate of infections. I need to evaluate if hospital-onset rate changed over time due to interventions (Intervention 1 and 2) compared to pre-intervention period after controlling for community-onset rate. I would appreciate any advice on this. I have read the examples of simple ITS and comparative ITS from Joe Caswell, however, I am confused.

I also tried to rearrange the data (stack them as one rate variable with different dummy variables)as demonstrated in the article link above, but not sure how to approach this with two interventions and two rates.

proc model data =TimeSeries2;
parms b0 b1 b2 b3 b4 b5 b6 b7;
  rate = b0+(b1*t)+(b2*x)+(b3*tx)+(b4*z)+(b5*zt)+(b6*zx)+(b7*ztx);
fit rate / covb ;
run; quit;

Thanks a lot!

Re: Help with multi series interrupted time series analysis with two interventions

sbxkoenk — Fri, 16 Feb 2024 17:59:57 GMT

I have moved your question / topic to the "Forecasting & Econometrics" - board.

Koen

topic Re: Help with multi series interrupted time series analysis with two interventions in SAS Forecasting and Econometrics

Help with multi series interrupted time series analysis with two interventions

Re: Help with multi series interrupted time series analysis with two interventions