Data Integrity Check under review

Cross checks that need review

Current number of Data Integrity Checks: 8

	DICSummary	DiscussQuestion
Function long LOS()	LOS/Length of Stay should not be unlikely long based on historical LOS for a given ward (Service/Location field).
Query Check_VADT_too_close_to_first_boarding_loc	no idea hope JM or LK remember	We discussed this in our after-Task meeting 2023-07-23 and I just stumbled across it. I do not remember what this was about, nor what a reasonable delay would be. Could one of you fill in the details for the query if we still want it? No hurry. In fact I'd prefer if you not hurried ;-). Ttenbergen 21:38, 13 August 2025 (CDT) this came about when the Pre-admit Inpatient Institution location is own site and previous location is own ER and the VADT is too close to first boarding loc dttm. If I remember correctly you mentioned 6 hours (0.25d) gap as maybe entry error on the preadmit inpt. Is 12 hours (0.50d) reasonable? --JMojica 09:23, 14 August 2025 (CDT) So this is also about Patients who appear to transfer from in-patient situation to Emergency Room, which would be checked by Query Check_from_ED_and_prev_location_same_site. If the patient is from anywhere else there would be no such thing as vadt being too close to first boarding loc, right? So what is this cut-off, then? Is it a hard check? We discussed this in our after-Task meeting 2023-07-23 and I just stumbled across it. I do not remember what this was about, nor what a reasonable delay would be. Could one of you fill in the details for the query if we still want it? No hurry. In fact I'd prefer if you not hurried ;-). Ttenbergen 21:38, 13 August 2025 (CDT) this came about when the Pre-admit Inpatient Institution location is own site and previous location is own ER and the VADT is too close to first boarding loc dttm. If I remember correctly you mentioned 6 hours (0.25d) gap as maybe entry error on the preadmit inpt. Is 12 hours (0.50d) reasonable? --JMojica 09:23, 14 August 2025 (CDT) So this is also about Patients who appear to transfer from in-patient situation to Emergency Room, which would be checked by Query Check_from_ED_and_prev_location_same_site. If the patient is from anywhere else there would be no such thing as vadt being too close to first boarding loc, right? So what is this cut-off, then? Is it a hard check? SMW Cargo Categories
Query Import request matcher	Records in for which we have patients in L_Log but no lab records from DSM
Query check CCI each count vs LOS	Confirm that a Category:Labs Imaging count is not unreasonably high
Query check long transfer delay	Is the Transfer Delay (Critical Care) or Transfer Delay (Medicine) unreasonably long?
Query check_tmp_AHC	For patients with Previous Location ER at same site there should be a Boarding Loc ER.	I implemented 'item must not be "not entered" and 40 records in the data I had at the time a "not entered" in complete data. Did I misunderstand the instructions? Or are these correct instructions and should be implemented as that? JALT if there is referral sent there must be a referral received entry and a consult dealt with entry Lisa Kaita 11:31, 7 August 2025 (CDT) pt could die in between? consult could go missing? In a way those would be really the ones we would want to know about, no? I suppose we could make it a soft check... Ttenbergen 16:26, 19 August 2025 (CDT) this almost sounds like the opposite of how I would have understood the current instructions. I would have thought those to mean to only enter "consult received" if there was no good data for consult sent. How do we actually want to use this? late answer: how did Julie analyze this? at the time all fields were mandatory, unless there was no consult, current status, collect consult sent and if no data found for this then use consult received. Lisa Kaita 12:59, 13 January 2026 (CST) I don't know, flagging for Julie and putting this on the JALT agenda; collection is still going, so we may still want to implement this. Ttenbergen 14:58, 13 January 2026 (CST)
Query check_tmp_Overstay2	Not yet filled in	others?
Query s_tmp_MR	Checks for Project MR