2

There is a table in MIMICII named "MIMIC2V26.ICUSTAY_DETAIL" and there is a field named "SAPSI_FIRST" inside the table. It seems that this field contains SAPS score for the first hospitalization of patients in MIMICII.

Based on the fact that there were some patients in MIMICII who did not have any SAPS score in their first hospitalization. So:

How could I find recorded SAPS score for the first hospitalization of patients in MIMICIII v1.3?

As I understood, severity scores (I mean SAPS) is not a recorded distinct data in MIMICIII and it can be calculated based on other parameters found in MIMICIII(Vital signs, etc,). So, there is a SAPS score available for all patients ICU stay in MIMICIII. However, it seems that in MIMICII v2.6, there were some patients that did not have any recorded or calculated SAPS score at least on their first admission. Am I right? If yes, how could we recognise patients that did not have SAPS score on their first admission? I am asking cause that I am going to reproduce a paper result but with new database version. In the paper, authors filtered out the patients that had recorded SAPS score on their first admission. So that I am looking for the same records but in the MIMICIII database. I appreciate it if someone guides me to find such records but in the new database version.

Thanks a lot,

|improve this question
4

Code for severity scores and other related clinical concepts can be found in the mimic-code repository: https://github.com/MIT-LCP/mimic-code

|improve this answer
  • Thank you Alistair. As I understood, severity scores (I mean SAPS) is not a recorded distinct data in MIMICIII and it can be calculated based on other parameters found in MIMICIII(Vital signs, etc,). So, there is a SAPS score available for all patients ICU stay in MIMICIII. However, it seems that in MIMICII v2.6, there were some patients that did not have any recorded or calculated SAPS score at least on their first admission. Am I right? If yes, how could we recognise patients that did not have SAPS score on their first admission? – Soheil Mar 1 '16 at 21:00
  • You are correct - SAPS was null for some of the patients in MIMIC-II v2.6. In MIMIC-III, all patients have a SAPS calculated. However, that does not mean you should use it on all patients (e.g., using it on neonates is probably inappropriate). The best approach is to look at what components of the score are missing, and set the final score to missing if you feel that too many components are missing. By default, missing values are scored as 0 (which makes sense, as missing lab values are likely normal, but maybe missing heart rates make less sense). – alistairewj Mar 2 '16 at 22:03

Not the answer you're looking for? Browse other questions tagged or ask your own question.