We updated our production environment to build 60.3 on Wednesday and have noticed some inconsistencies between the “Chronic Homelessness Y/N” field from the CA module and what is reported under the new “Chronically Homeless” Yes/No field available on the client profile.
We use the CA module in order to run our local prioritization, so it’s important that we understand how it’s working and why the Y/N fields aren’t matching with what HIFIS is saying for the client on their profile.
This example client shows “Yes” for the new Chronically Homeless field on the client profile. Please see the relevant housing history for this client below as well.
From the CA Module, looking for this same client, he is reporting “N” under Chronic Homelessness Y/N. Here is a query demonstrating the relevant fields:
SELECT CA.ClientID ,CA.HomelessIn365Days ,CA.HomelessIn1095Days ,CA.ChronicallyHomelessYNFROM vw_CoordinatedAccess as CAWHERE ClientID = 125I am attempting to troubleshoot to trace the issue back to the source, but I thought I would start by reporting it and provide a sample query to the devs to demonstrate the issue.
SELECT CA.ClientID ,CA.HomelessIn365Days ,CA.HomelessIn1095Days ,CA.ChronicallyHomelessYN as CAChronicallyHomelessYN ,HIFIS_Clients.IsChronicallyHomelessYN as ClientProfileChronicallyHomelessYNFROM vw_CoordinatedAccess as CAINNER JOIN HIFIS_Clients on CA.ClientID = HIFIS_Clients.clientIDWHERE CA.ChronicallyHomelessYN <> HIFIS_Clients.IsChronicallyHomelessYNThe following query is currently giving us 126 results (with 1376 total results from vw_CoordinatedAccess). 69 of the 126 are showing Y for Chronic in the CA Module, and N for Chronic on the client profile. The remaining 57 are showing N for Chronic in the CA Module, and Y for Chronic on the client profile.
May have been fixed in 4.0.61.1
See also: Homeless Days in CA module
Another contributing factor to the mismatches between Client Profile and CA calculations for chronic homelessness Editing Client Vitals sets Chronically Homeless Status to 'No'
Christie: One more update for this ticket. We changed the vw_CoordinatedAccess view in our QA environment and adjusted some filters to include homeless days before public institution days, and adjusted the days in 1095 threshold to match what is used in sp_UpdateClientChronicityStatus (546)
After that, running the sp_UpdateClientChronicityStatus stored procedure with no parameters to update all clients seems to have eliminated the differences we found between the CA calculation for Chronic Homelessness and what is reported on the client’s profile for Chronic Homeless. After waiting a few days, we are now seeing more mismatches again, many of which are just over the 180 days in the past 365 threshold.
I tried running it again, and found that there are now no more mismatches.
I have asked for confirmation if this stored procedure runs overnight automatically in 60.3? I seem to recall something like that being the case in a previous version but I am not sure. Maybe there is there some other mechanism that is supposed to keep the Client Profile Chronic Homelessness Y/N field updated in v60?
Christie: I have some further information to add to the ticket below about the issues with Chronic Homelessness in v60. Our priority is making sure the Coordinated Access Module is calculating chronic homelessness correctly as that affects our prioritization process.
I have a more simplified example to provide that proves the issue in the CA module “Chronic Homelessness Y/N” is related to housing history records that map to the new Public Institution category in the Reaching Home continuum.
It appears that both the “HomelessIn365Days” and “HomelessIn1095Days” fields in the vw_CoordinatedAccess view are not calculating correctly when the client has a Public Institution housing history record.
I have created a sample client that has a Public Institution housing history record in between 2 homeless records. When the public institution record is in the middle, the first homeless record is not counting. When the Public Institution record is updated and changed to a housed record, the first homeless record begins to count correctly.