You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I am studying the article P. J. Thoral et al., “Explainable Machine Learning on AmsterdamUMCdb for ICU Discharge Decision Support: Uniting Intensivists and Data Scientists”
and I would like to know how the authors could have identified
"Palliative care patients and patients with do-notresuscitate or do-not-intubate orders"
Unfortunately, I'm not a dutch speaker.
Thanks,
Simone
The text was updated successfully, but these errors were encountered:
While it's currently not possible to completely replicate the exclusion of all patients from the public data set, since not all patients had a discrete DNR/DNI order, these are te terms you are looking for:
itemid
item
value
valueid
10673
Beleid
I
1
10673
Beleid
II
2
10673
Beleid
III
3
The 'Code' (Beleid) orders are
I: Full Code
II: Do not Resuscitate/Do not Intubate in addition to order 'Do Not' orders
III: Palliative Care
Thank you for your accurate answer!
In my first attempt to replicate your analysis I excluded patients who died on the same day or the day after discharge, obtaining higher performance.
This is not the right place to delve into this issue, but I believe that a similar exclusion strategy applied to readmission within 48 hours might not be considered cheating [M.J. Al-Jaghbeer et al.].
Hello,
I am studying the article
P. J. Thoral et al., “Explainable Machine Learning on AmsterdamUMCdb for ICU Discharge Decision Support: Uniting Intensivists and Data Scientists”
and I would like to know how the authors could have identified
"Palliative care patients and patients with do-notresuscitate or do-not-intubate orders"
Unfortunately, I'm not a dutch speaker.
Thanks,
Simone
The text was updated successfully, but these errors were encountered: