Turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

- Home
- /
- Analytics
- /
- Stat Procs
- /
- How to plot survival estimates from the model / excess hazard ratio

Options

- RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Mute
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content

Posted 06-11-2019 05:29 PM
(448 views)

Hello folks,

I'm trying to reproduce the graph below (Woods et al.,2012) attached to this post. And I read and produced the cumulative relative risk tables following Dr. Dickman's resources for http://www.pauldickman.com/survival/sas/relative_survival_using_sas.pdf. Now I'm getting into the Chapter: Modelling relative survival hoping that I'm pursuing the right direction to find a way to reproduce the graph.

My question is:

1. Do you recognize this graph as a SAS output? or you think that authors might have created this plot from any output estimates data?

2. Does proc phreg produce "excess hazard ratio" as used on the y-axis int he plot below over a survival time in weeks?

I know these authors mostly use stata software. But I can't imagine that SAS wouldn't have a way to produce something close. I keep reading Dr. Dickman's manual. In the meantime, I greatly appreciate to hear from your educated suggestions as to how I should approach to produce a similar plot like below. I use comparable data to that of Woods et al., and datasets that Dr. Dickman used in his tutorial.

1 REPLY 1

- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content

It appears that Dickman suggested to use poisson regression to model and output relative excess risks. Please let me know if this might be the way to obtain estimates to graph in SAS. I'll keep testing in the meantime. Thanks.

```
proc genmod data=&grouped(where=(fu le 5)) order=formatted;
title2 'Poisson error model fitted to collapsed data';
title3 'Main effects model (follow-up, sex, age, and dgnyear)';
fwdlink link = log(_MEAN_-d_star);
invlink ilink= exp(_XBETA_)+d_star;
class fu sex age yydx;
model d = fu sex yydx age / error=poisson offset=ln_y type3;
format fu fu. age age. yydx yydx.;
run;
```

**Available on demand!**

Missed SAS Innovate Las Vegas? Watch all the action for free! View the keynotes, general sessions and 22 breakouts on demand.

What is ANOVA?

ANOVA, or Analysis Of Variance, is used to compare the averages or means of two or more populations to better understand how they differ. Watch this tutorial for more.

Find more tutorials on the SAS Users YouTube channel.