topic PROC ARIMA with binary response in SAS Programming

PROC ARIMA with binary response

jp134711 — Sun, 12 May 2013 23:35:41 GMT

Is it possible to combine a binary response variable (e.g. whether or not a patient is readmitted) with a time series model using PROC ARIMA?

Thanks!

Re: PROC ARIMA with binary response

Reeza — Sun, 12 May 2013 23:37:07 GMT

No idea, but it sounds more like survival analysis.

What doesn't survival analysis cover that you would want in ARIMA, some sort of seasonality?

Re: PROC ARIMA with binary response

jp134711 — Sun, 12 May 2013 23:47:17 GMT

I'm interested in investigating the effect of an intervention on readmission rate, after controlling for patient-level covariates. I have about 3 years of historical/pre-intervention data and 1.5 years of post-intervention data. I want to incorporate time trend to account for changes in medical practice over time and its relation with readmission.

Re: PROC ARIMA with binary response

SteveDenham — Mon, 13 May 2013 13:45:05 GMT

That sounds, at least to me, more like a survival analysis with a time-dependent covariate, as proposed by @Reeza. Think about what the ARIMA model would be fitting--a long string of zeroes, a single 1, perhaps some more 1's (if you model as still admitted), then another long string of zeroes. That is not a good dataset for fitting an ARIMA model. Instead, time to re-admission, with a covariate that describes the intervention status, strikes me as something that would work. Check out PROC PHREG.

Steve Denham

Re: PROC ARIMA with binary response

Reeza — Mon, 13 May 2013 15:04:17 GMT

I would add some indicator variables, possible time dependent to account for the changes in practice.

You'll have to be careful with the pre-intervention/post-intervention data to make sure they're handled appropriately, but survival analysis is what you're looking for.

Re: PROC ARIMA with binary response

SteveDenham — Mon, 13 May 2013 15:23:42 GMT

, that's what I was trying to say. I think a record would look like:

subjid date admission_status intervention_status covariate1 covariate2 (other covariates of interest).

That should set it up for a survival analysis, as per Example 67.7 Time-Dependent Repeated Measurements of a Covariate.

Steve Denham