topic Re: Funnel plot for standardized rate in SAS Programming

Funnel plot for standardized rate

Isevr — Thu, 28 Apr 2016 16:02:02 GMT

Hy,

I raise an unresolved issue posted in 2013.

I am wondering if anyone has had experience working with indirectly standardized rates.

Standardized rates is the product of the target crude rate and the ratio of observed to expected values from the relevant regression model.

In short, I try to compute the control limits by using the formulas in Appendix A.2.1.in Spiegelhalter's 2005 paper on funnel plots.

I'm working on SAS 9.3.

Thanks,

Re: Funnel plot for standardized rate

Cynthia_sas — Sun, 01 May 2016 16:58:13 GMT

Hi:
Look at these 2 blog posts. One of them mentions the Spiegelhalter paper:
http://blogs.sas.com/content/iml/2011/04/15/funnel-plots-an-alternative-to-ranking.html
http://blogs.sas.com/content/iml/2011/11/23/funnel-plots-for-proportions.html

cynthia

Re: Funnel plot for standardized rate

Isevr — Mon, 02 May 2016 07:14:42 GMT

Thank you for answering.

These papers are for binomial and normal variables, I'm looking for the application for poisson variable.

Isabelle

Re: Funnel plot for standardized rate

Rick_SAS — Tue, 03 May 2016 14:31:00 GMT

Yes, but Spieghalter show in Appendix A.2 how to compute the limits for the Poisson case. The second blog post uses the formulas from Appendix A.1, but the idea is the same. The important programming detail is that you can use the QUANTILE function to invert the cumulative Poisson distribution (F^{-1} in the paper). For the interpolation, mimic what the blog post does when it interpolates the binomial proportions.

Re: Funnel plot for standardized rate

Isevr — Tue, 03 May 2016 14:55:28 GMT

I try to.It seems to work for smr (target theta=1) but for standardized ratio with target not equal to 1, it is not so simple.

Thank you anyway.

Re: Funnel plot for standardized rate

sasseln — Thu, 19 Oct 2017 13:06:45 GMT

Hi Isevr,

I am having the same problem as you. Did you manage to write a sytnax for the intervals for SMR using the poisson distribution? Would you mind sharing it? That would be so helpful!

Thanks a lot!