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
- /
- Learn SAS
- /
- Analytics U
- /
- Cochran-Mantel-Haenszel Odds Ratios

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

10-18-2016 05:13 AM

I am completing a meta-analysis for my graduate thesis in Epidemiology and am attempting to find a CMH odds ratio for the relationship between Schizophrenia and 3q29 deletions. I have written the code as:

proc freq data=schizophrenia3q292;

tables study*case*deldupcat / relrisk CMH;

title "3q29 Deletions and Duplications";

run;

with data in the form of:

studycat $ case x n

with case=(0=control, 1=case), x=number of individuals with deletion and n=total number of individuals (cases or controls).

I followed that up with:

data schizophrenia1p362(drop=i);

length studycat $20;

set schizophrenia1p36;

do i=1 to n;

if i<= x then deldupcat=1;

if i > x then deldupcat=0;

status=2-deldupcat;

output;

end;

run;

in order to create an observation for each individual (a pseudo-"studyid") in order to avoid using a weight.

While this is only one of several CNV's I am studying, I know, from a prior meta-analysis that the unadjusted odds ratio should be around 54 (mine is 55) and the CMH odds ratio should be around 41 (mine is 284!). What could be causing my adjusted OR to be SO high? Is this normal? The previous meta-analysis used R instead of SAS. But, this shouldn't cause such a discrepancy, should it?

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

Posted in reply to bmacdonald627

10-18-2016 10:23 AM

Without data it is hard to say specifically why any result is obtained. Comparison with another product or procedure and differences of result, assuming same data is used, often comes down to either different algorithms used or options that may be optional in one but defaut in the other. Again without seeing both code, and knowing both languages (I don't speak R for instance) it would be hard to say.

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

Posted in reply to ballardw

10-19-2016 05:37 AM - edited 10-19-2016 05:41 AM

Here is the data:

The first column is just the observation numbers assigned by SAS. The second column is the study. The fourth column identifies it as a deletion or duplication (I am just looking at deletions at the moment). The fourth column identifies where it was a case (case=1) or control (case=0). "X", or the fifth column, identifies how many cases or controls possessed this deletion, and the final column has the total number of cases or controls in the study.

Obs study location deldup case x n

1 | 1 | 3q29 | duplicat | 1 | 3 | 420 |

2 | 1 | 3q29 | duplicat | 0 | 2 | 416 |

3 | 4 | 3q29 | deletion | 1 | 2 | 3391 |

4 | 4 | 3q29 | deletion | 0 | 0 | 3181 |

5 | 4 | 3q29 | duplicat | 1 | 0 | 3391 |

6 | 4 | 3q29 | duplicat | 0 | 1 | 3181 |

7 | 9 | 3q29 | deletion | 1 | 1 | 172 |

8 | 9 | 3q29 | deletion | 0 | 0 | 160 |

9 | 12 | 3q29 | deletion | 1 | 1 | 245 |

10 | 12 | 3q29 | deletion | 0 | 0 | 490 |

11 | 15 | 3q29 | deletion | 1 | 4 | 6882 |

12 | 15 | 3q29 | deletion | 0 | 0 | 6316 |

13 | 17 | 3q29 | deletion | 1 | 0 | 1438 |

14 | 17 | 3q29 | deletion | 0 | 1 | 33246 |

15 | 19 | 3q29 | deletion | 1 | 6 | 4719 |

16 | 19 | 3q29 | deletion | 0 | 0 | 5917 |

17 | 19 | 3q29 | duplicat | 1 | 1 | 4719 |

18 | 19 | 3q29 | duplicat | 0 | 0 | 5917 |

19 | 20 | 3q29 | duplicat | 1 | 10 | 8290 |

20 | 20 | 3q29 | duplicat | 0 | 0 | 7431 |

21 | 21 | 3q29 | deletion | 1 | 1 | 150 |

22 | 21 | 3q29 | deletion | 0 | 0 | 268 |

23 | 23 | 3q29 | deletion | 1 | 5 | 3945 |

24 | 23 | 3q29 | deletion | 0 | 0 | 3611 |

25 | 26 | 3q29 | deletion | 1 | 1 | 834 |

26 | 26 | 3q29 | deletion | 0 | 0 | 672 |

27 | 27 | 3q29 | deletion | 1 | 2 | 1259 |

28 | 27 | 3q29 | deletion | 0 | 0 | 1173 |

Hopefully this can help!

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

Posted in reply to bmacdonald627

10-19-2016 06:59 AM

This is the outcome for the raw, unadjusted odds ratio:

This is the outcome for the mantel-haenszel odds ratio: