MIMIC-IV Admissions table - Patients who died are later admitted to the hospital with the new hadm_id

Prerequisites

• Put an X between the brackets on this line if you have done all of the following:
  • Checked the online documentation: https://mimic.mit.edu/
  • Checked that your issue isn't already addressed: https://github.com/MIT-LCP/mimic-code/issues?utf8=%E2%9C%93&q=

Description

I am doing an exploratory data analysis of patients readmission to the hospital. I am looking at the readmission rates depending on the discharge location. I now came across quite a strange scenario: there are patients who died in the hospital (discharge_location = 'DIED', deathtime is equal to admittime in the ADMISSIONS.csv table. But these died patients were readmitted to the hospital 1 minute later to the PACU admission location with the new HADM_ID. I have found 110 patients with the same issue (e.g. subject_id = 10304050, 10068026). In their subsequent admission they do not have deathtime and hospital_expire_flag, giving the impression that they are alive.
I could not find the description to this problem in the available documentation. Maybe this is how hospital system works. I would be grateful if you could provide any clarifications on this issue.

Thanks,
Nodira

Start with the MIMIC online documentation and the ADMISSIONS.csv records for subject_id 10304050 and 10068026. Compare the DIED admission with the later PACU admission, including deathtime and hospital_expire_flag, and determine whether the documented data model explains the sequence. Done when the admission sequence has a clear documented interpretation or a reproducible data defect is identified.