Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.

# st: Question about ir

 From "Reg Jordan" To Subject st: Question about ir Date Wed, 7 Nov 2012 10:03:31 -0500

```Stata 12.1 updated, Win7 64-bit system

I have a dataset from which I would like to determine the incidence rate of
coronary heart disease (chd) by BMI class.
ir var(cases)   var(exposed)   risk time, by(bmiclass) , should work, but
I'm having problems with var(exposed). It should be the number of subjects
in each bmiclass, but how do I express this as a variable?

bmiclass         |         N
---------------+----------
1. underweight |        57
2. normal        |      1936
3. overweight   |      1845
4. obese        |       596
---------------+----------
Total           |      4434
--------------------------

I suppose I could use another variable, like sex1:

Summary for variables: sex1
by categories of: bmiclass (BMI index)

bmiclass         |         N
---------------+----------
1. underweight |        57
2. normal        |      1936
3. overweight   |      1845
4. obese        |       596
---------------+----------
Total            |      4434
--------------------------

But

. ir chd sex1 time, by(bmiclass)   yields

BMI index       |      IRR       [95% Conf. Interval]   M-H Weight
-----------------+-------------------------------------------------
1. underweight |          .             .          .            0
2. normal        |          .             .          .            0
3. overweight   |          .             .          .            0
4. obese        |          .             .          .            0
-----------------+-------------------------------------------------
Crude          |          .             .          .
(exact)
M-H combined |          .             .          .
-------------------------------------------------------------------

A brute force approach yields, for example

. ir chd sex1 time if bmiclass == 1

| Sex, exam 1                        |
|   Exposed   Unexposed  |      Total
-----------------+------------------------+------------
Incident Hosp MI |         6                                 0  |          6
Time [years] to     |   1119.57                          0  |    1119.57
-----------------+------------------------+------------
|
|
Incidence rate    |  .0053592           .                |   .0053592
|
|
|      Point estimate                              |
[95% Conf. Interval]
|------------------------+------------------------
Inc. rate diff. |                .                                     |
.           .
Inc. rate ratio |                .                                    |
.           .
Attr. frac. ex. |                1                                   |
.           1
Attr. frac. pop |                1                                 |
+-------------------------------------------------
(midp)   Pr(k>=6) =                     0.5000 (exact)
(midp) 2*Pr(k>=6) =                     1.0000 (exact)

Thanks for your assistance.

Reg Jordan

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/
```