‹ Home
Exported

CSTE National ELR Workgroup

Welcome to the CSTE National ELR Workgroup Basecamp! We hope you will utilize the space to share resources with colleagues, start discussion threads, and identify topics for the workgroup to tackle on future calls. All workgroup members will have the opportunity to join. Please note that you may edit your notification preferences to suit your needs (and your email inbox). If you are not already on the ELR distribution list and would like to receive call communications, please enroll using this subscription form (copy and paste in browser): https://app.smartsheet.com/b/form/5a91a4579952496682084796b82112e6
Becky Lampkins, Manager, Surveillance/Informatics Program at CSTEMegan Toe, CSTEScott Troppy, Surveillance Epidemiologist at MA DPHMeredith Lichtenstein, CSTEAmy Drake, NMJohn RobertsNancy Barrett, Epi 4/PH Informatics SpecialistErin Holt, TDHRita Altamore, EpidemiologistDavid DiCesareJason HallWalter Kemper, ELR Coordinator, NC DHHS Div of Public HealthJulie LuepkeCarmen Pugh, Labcorp Manager State Reporting at LabCorpTom Safranek +399 others
Posted by Michael Nuss ·

Respiratory Panel Question

Good Morning/Afternoon,

I have a quick question regarding how you handle your panels.

When you have multiple results in a panel there are times when a certain results should be assigned to a specific program area.  A good example is if an OBX is FLU A, another OBX is FLU B, and the third OBX is COVID-19.  In Nebraska, Flu should be assigned to Syndromic Surveillance, but COVID-19 should be assigned to Gen/Epi Program area.  How do you handle results like this?  Do you assign the entire panel to one program area or do you have some sort of splitter in place where each result is assigned to it's own program area?

Thank You,
Mike
<Nebraska>

Comments & Events

Garrett Jenkins, CalREDIE Lab Coordinator at California Department of Public Health Garrett Jenkins, CalREDIE Lab Coordinator
California has our Rhapsody intake engine split the messages along the observation results prior to delivering to our DSS and route any that are HIV to another program area.
Notified 306 people
Mark Dittman Mark Dittman
In PA:

* We parse each OBR/OBX pair or set into it's own message, creating a unique report in our NEDSS system (all program areas share this system).
* In your example, we would create three test/result sets based on the OBR-4 LOINC. This process takes place in our "ELR Workflow Processor" logic.
* You get three PA-NEDSS reports, one per condition, A, B, and COVID.
* All reports are linked to the same Accession number
* All reports are linked to the same patient

If we had two program areas, as you do, one being Flu and one COVID, we would simply route those to our 'in-box' system which filters based on conditions which have been mapped in the back-end to a 'program area'. The Flu tests would go to "Syndromic" the COVID to "Epi", in your example.

Hope that helps,
Mark
~WRD0000.jpg 823 Bytes • Download
Notified 306 people
Danny Power Danny Power
Mark,

Do you keep all tests for the panel in your system? Say it is positive for SARS-COV-2 and then negative for Flu A and Flu B, would the Negative flu results still go in?

If not do you store them somewhere else or just disregard them?

Thanks,

Danny Power, Montana
Notified 306 people
Michael Nuss, Informatician at Nebraska Department of Health and Human Services Michael Nuss, Informatician
Yep. At one time we kept the entire panel in one message and moved it to a specific program area.  Now we have a splitter installed that splits the message along the observation results, and each result is assigned to it's own program area. 

I was curious how other states are doing it to see if we are handling the labs properly or if we need to do something else :)

Thanks,
Mike
<Nebraska>
Notified 306 people
Mike Firkser Mike Firkser
In New Jersey, we create two cases and they are each assigned to the applicable program area.


Thank you,

Mike Firkser
Technical Lead – Electronic Lab Reporting
NJ Dept. of Health/Health Information Technology
michael.firkser@doh.nj.gov
Notified 306 people
Mark Dittman Mark Dittman
Generally we tell labs to not send us negatives from a panel but most cannot filter those reports. Our internal auto-create rules for reports can create and close reports so those never get in front of staff.
~WRD0001.jpg 823 Bytes • Download
Notified 306 people
Swathi Ramasahayam Swathi Ramasahayam
In Iowa, we parse each OBX as a seperate record and assign a condition/disease to each one. From our backend database, we route to different program areas based on the condition.
Notified 306 people
Nancy Barrett, Epi 4/PH Informatics Specialist Nancy Barrett, Epi 4/PH Informatics Specialist
We parse results in our transformation/validation platform based on the OBX3/OBX5 information which maps to a disease/condition. Results are parsed into files for upload into our Maven system or posted to folders if a program is not in Maven (e.g. HIV). In Maven for the flu example, if the two positive flu results were sent under the same OBR, both would go into the same flu event (we would count the flu types as separate by external data processes). Any other positive results from that panel would create their own disease event. Program staff have permissions to access the diseases they are allowed to see in the system. If a lab cannot parse out sending us negative results that we don't want, we can program our transformation/validation platform to do that (we disregard them). 
Notified 306 people
Walter Kemper, ELR Coordinator, NC DHHS Div of Public Health Walter Kemper, ELR Coordinator, NC DHHS Div of Public Health
NC uses Rhapsody filters that split a message into separate messages:

* A message that contains only tests for blood lead or environmental tests.
* A message that contains only COVID-19 tests.
* A message that contains only HIV and syphilis tests.
* A message that contains only tests other than the above..
If a resultant message still contains tests for more than one disease it creates/updates a separate disease event (disease case) for each disease for which the message contains a qualifying result for that disease. All the tests from that message are imported into each disease event but the disease event is for only one disease. Each disease event gets assigned to the appropriate program area for that particular disease,.

Walter Kemper
Interfaces Consultant for NC DHHS Division of Public Health
Email: Walter.Kemper@dhhs.nc.gov<mailto:Walter.Kemper@dhhs.nc.gov>
(919) 426-3468 (mobile)r
Notified 306 people
Deb Loniewski Deb Loniewski
In Michigan we split all panels up using the OBX-3 loinc to label them by condition then send the individual results to our surveillance system, MDSS. There's also logic to split large respiratory (BioFire) panels to not send Negative, Non-reportable observations to MDSS, using OBX-3 and OBX-5.
Notified 306 people
Jonathan Johnson, Informatics and Data Science Lead (HI - DOH) Jonathan Johnson, Informatics and Data Science Lead (HI - DOH)
Aloha, in Hawaii since our Maven is a centralized resource across most of our disease groups and teams, our Rhapsody does not de-batch based on OBX values. However, Maven does de-batch based on OBX-3 values. So a Covid/Flu panel would be de-batched into multiple disease events.

For programs that are not in Maven, such as HIV, they have a dedicated view from our data warehouse, the relational database also splits along OBX-3. We do not currently parse out negative sends.
Notified 306 people