Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Question about ir


From   "Reg Jordan" <[email protected]>
To   <[email protected]>
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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index