turn on suggestions

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

Showing results for

Find a Community

- Home
- /
- SAS Programming
- /
- SAS Procedures
- /
- Difference between adjusted exponentiated means in...

Topic Options

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

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

10-14-2015 01:51 PM

Hello,

I am trying to find the difference between adjusted exponentiated means using proc genmod. I have two sets of patients: 1 set took drug A (DrugA=1), 1 set did not (DrugA=0). The outcome is the total number of times they visited the doctor, so count variable so I would like the distribution to be Poisson and the link to be log. Here is the code:

PROC genmod DATA=MYDATA;

CLASS covariate1 DRUGA/DESC;

model DocVisits=DrugA covariate1 covariate2 /link=log dist=poisson ;

lsmeans DrugA / exp diff cl ;

run;

And here is the output:

SAS Output

DrugA Least Squares Means | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|

DrugA | Estimate | Standard Error | z Value | Pr > |z| | Alpha | Lower | Upper | Exponentiated | Exponentiated Lower |
Exponentiated Upper |

1 | 4.2798 | 0.01154 | 370.78 | <.0001 | 0.05 | 4.2572 | 4.3024 | 72.2270 | 70.6113 | 73.8796 |

0 | 4.3194 | 0.009114 | 473.94 | <.0001 | 0.05 | 4.3016 | 4.3373 | 75.1464 | 73.8160 | 76.5008 |

Differences of DrugA Least Squares Means | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|

DrugA | DrugA | Estimate | Standard Error | z Value | Pr > |z| | Alpha | Lower | Upper | Exponentiated | Exponentiated Lower |
Exponentiated Upper |

1 | 0 | -0.03963 | 0.01419 | -2.79 | 0.0052 | 0.05 | -0.06743 | -0.01182 | 0.9611 | 0.9348 |
0.9883 |

Thanks so much in advance for any advice!