Please make CONCEPT_ID 32220 a Standard Concept

[clairblacketer]

There is no standard Visit concept for the idea "still in hospital" which is a concept we use to identify patients that are still in the hospital at the time of the data extract. There is a UB04 code for this, represented by the non-standard concept 32220. Can you please make this a standard concept?

[cgreich]

@clairblacketer:

Is this trying to solve the problem of visit_end_date at the end of the database?

[Alexdavv]

@clairblacketer It looks like we've found a kind of solution here. Has something changed?

[clairblacketer]

So the solution suggested in the the thread has us put the OBSERVATION_PERIOD_END_DATE as the VISIT_END_DATE for patients still in the hospital. I agree with this assessment but there is still not a good way to determine which patients are still in the hospital and which patients were actually discharged on that day. As suggested in the forum post we would like to use this concept for that purpose. Regardless, this source code does not have a mapping to a standard concept so it makes sense to me to just make this one standard.

[Alexdavv]

Option #2 should satisfy all use cases. It’s just against the rules right now. And I would like the rule to be changed to allow NULL Visit end_dates ONLY when a Visit is ongoing. If the Visit is not ongoing, then an end_date needs to be populated. Populating the date of ETL as the Visit end_date for an ongoing Visit is not necessary and will affect analysis.

@clairblacketer I thought it was agreed that the NULL end date would be an indicator for continuing visit. But now I see it’s not.

@cgreich @chrisknoll @MelaniePhilofsky @ericaVoss Do you feel we agreed on something?

[chrisknoll]

For me, a null end date is fine provided the duration of visit can be found by using the observation period end date. For cohort definitions, I can make queries substitute observation_period_end_date for when visit_end_date is null.

I was arguing for non-null end dates in visits due to that I had no way of determining the duration of a visit as of a certain 'data-freeze' in time. Substituting observation_end_date for a null visit_end_date seemed like a reasonable compromise.

[clairblacketer]

Bringing this back up - @Alexdavv we still need this concept 32220 to be made standard

[chrisknoll]

Can we do a quick thought experiment using a visit_occurrence with concept_id 32220?

You have a cohort definition, entry events are visits....you want to characterize time in hospital. Is the record with 32220 have a visit end date? If we have existing phenotypes that look for IP visits, does a record not show up until the person is finished with their visit? Those are just a couple of questions to get the juices flowing, but I would like to understand how this type of visit_occurrence record is used in practice.

[clairblacketer]

@chrisknoll this concept is more for hospitals that convert their data on a daily or weekly basis. The basis of this question is for how to identify patients that are still in the hospital at the time of conversion. The idea is that the visit_end_date is the date of conversion and the concept of 32220 would denote that they are still in the hospital. This would change to 9201 once the person is discharged and the visit_end_date becomes the date of discharge.

[clarkevans]

I do not think this is a vocabulary issue. It's a data model issue. I think a reasonable fix could be put into tools currently, and then this change made official in a later CDM.