29.06.2014 Views

Coding ICD-9 and Charlson Index Scores in R for SAS Users ...

Coding ICD-9 and Charlson Index Scores in R for SAS Users ...

Coding ICD-9 and Charlson Index Scores in R for SAS Users ...

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

• Use the cast function <strong>in</strong> the Reshape package to transpose the data by Patient ID.<br />

• Replace any disease category count over 1 with a 1 to <strong>in</strong>dicate presence. Note, the<br />

number of times a disease is recorded is not important <strong>in</strong> scor<strong>in</strong>g the <strong>Charlson</strong> <strong>Index</strong>.<br />

• Assign appropriate weights per disease category.<br />

• Compute a comorbidity count <strong>for</strong> each patient.<br />

• Compute the <strong>Charlson</strong> <strong>Index</strong> <strong>for</strong> each patient.<br />

• Return the f<strong>in</strong>al dataset.<br />

The beg<strong>in</strong>n<strong>in</strong>g of the program classifies each <strong>ICD</strong>-9 code to the correspond<strong>in</strong>g disease category<br />

<strong>for</strong> the <strong>Charlson</strong>. The data are then transposed by patient so that across each patient row we can<br />

see which of the disease categories are present. Then a count of comorbidities <strong>and</strong> the <strong>Charlson</strong><br />

<strong>Index</strong> is calculated.<br />

Portion of sample <strong>in</strong>put data:

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!