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

st: bug in charlson.ado program

From   "Kieran McCaul" <>
To   <>
Subject   st: bug in charlson.ado program
Date   Thu, 12 Nov 2009 15:43:51 +0800


A few days ago a question came up regarding the user-written -charlson-
program and I indicated that there was a bug in this program.

The following code is at line 379 in charlson.ado:

  forvalues i=1/17 {
    gen weightch`i'=0
    quietly replace weightch`i'=1 if ynch`i'==1
*Change weights for more serious comorbidites (for calculation of
charlson index, based on sum of weights)

  quietly replace weightch12=2 if ynch12>0
  quietly replace weightch13=2 if ynch13>0
  quietly replace weightch14=2 if ynch14>0
  quietly replace weightch15=3 if ynch15>0
  quietly replace weightch16=6 if ynch16>0
  quietly replace weightch17=6 if ynch17>0

This block of code assigns weights to diagnoses flagged by the Charlson
program.  There are 17 comorbidities (ynch1 to ynch17) and if these are
flagged then the weights are set to 1 initially in the -forvalues- loop.

More serious comorbidities are weighted more heavily and this is done in
the lines that follow the -forvalues- loop.

So this code would assign a weight of 1 to any of the first 11
comorbidities (ynch1 to ynch11) and higher weights to the remaining 6
comorbidities (ynch12 to ynch17).

Now the problem is that ynch11 is "Diabetes with complications" and in
the Charlson weighting scheme this should receive a weight of 2, but in
the code above it receives a weight of 1.

So an extra line is needed:
  quietly replace weightch11=2 if ynch11>0

This was discovered by some of my colleagues (Janine Calver and Zoe
Hyde) some time ago and they did contact the author of charlson.ado, but
the correction has not appeared (yet).


Kieran McCaul MPH PhD
WA Centre for Health & Ageing (M573)
University of Western Australia
Level 6, Ainslie House
48 Murray St
Perth 6000
Phone: (08) 9224-2701
Fax: (08) 9224 8009
If you live to be one hundred, you've got it made.
Very few people die past that age - George Burns

*   For searches and help try:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index