Multiple Death Records Due to Death_type or Cause - Version 5.4

[RogerJCarlson]

Issue Name: Multiple Death Records Due to Death_type_concept or Cause_concept Version 5.4

Issue Type: Duplicate records caused by multiple standard concept_ids for death cause or death type (provenance)

Description: http://ohdsi.github.io/CommonDataModel/cdm54.html#DEATH The table description for the Death table has the following: "A person can have up to one record if the source system contains evidence about the Death". However, an ICD diagnosis code may map to multiple SNOMED codes:

ICD10: J96.21 (Acute on chronic respiratory failure with hypoxia (HCC)) maps to SNOMED: 389086002 (Hypoxia) and 67905004 (Acute-on-chronic respiratory failure) This results in two records with the same death date.

Similarly, a person may have death records from multiple systems (provenance), i.e. EHR and death registry.

Suggested Solution: For OMOP version 5.4, allow multiple records in the Death table caused by either the death_type_concept_id and/or cause_concept_id. However the Death_Date(s) must match.

Possible Implications:

• CDM documentation will need to be modified.
• The rules of the DQD (Data Quality Dashboard) will have to be adjusted to match.
• Others?

Related forum post: https://forums.ohdsi.org/t/multiple-death-records-due-to-snomed-mapping/19517

Reported By: Roger Carlson

[askanter]

Tying summary data from Themis OHDSI/Themis#98: Summary of issues

Write out some of the questions that this convention is meant to address There are two reasons why there might be more than one cause of death for a patient.

There can be multiple different causes of death depending on the source. A single source cause of death might map to multiple standard codes. Summary of answer

Based on your findings, give a few bullets on what the community decided If multiple death records occur, the date and the person have to be the same, but the cause can be different. Can be reported by different sources as well. https://github.com/OHDSI/Themis/issues/5 For OMOP version 5.4, allow multiple records in the Death table caused by either the death_type_concept_id and/or cause_concept_id. However the Death_Date(s) must match. https://github.com/OHDSI/CommonDataModel/issues/758 (open) Related links

https://forums.ohdsi.org/t/themis-conventions-for-death-table-to-allow-multiple-records-version-5-4/19992 https://github.com/OHDSI/CommonDataModel/issues/758 (open)

https://github.com/OHDSI/Themis/issues/5 (CLOSED)

https://github.com/OHDSI/CommonDataModel/issues/209 (open)

[MelaniePhilofsky]

Based on discussion during the Themis call on June 6, 2024 to modify the proposal, this issue is being moved back to the "post to forum" step.

[RogerJCarlson]

Modified Proposal

Based on discussion during the Themis call on June 6, 2024 to modify the proposal, this is the modified proposal. Please discuss in Forum Post: https://forums.ohdsi.org/t/revised-multiple-death-records-due-to-death-type-or-cause-version-5-4/21796

Issue # and location

• [Themis issue OHDSI/CommonDataModel#758] (https://github.com/OHDSI/CommonDataModel/issues/758)

Issue summary

The table description for the Death table has the following: "A person can have up to one record if the source system contains evidence about the Death". However, an ICD diagnosis code may map to multiple SNOMED codes: ICD10: J96.21 (Acute on chronic respiratory failure with hypoxia (HCC)) maps to SNOMED: 389086002 (Hypoxia) and 67905004 (Acute-on-chronic respiratory failure) This results in two records with the same death date. Similarly, a person may have death records from multiple systems (provenance), i.e. EHR and death registry.

Convention type

Table

CDM table

Death Table

CDM field

CAUSE_CONCEPT_ID CAUSE_SOURCE_VALUE CAUSE_SOURCE_CONCEPT_ID

Links to issue discussion

1. • [Multiple Death records due to SNOMED mapping] ( https://forums.ohdsi.org/t/multiple-death-records-due-to-snomed-mapping/19517/8)
2. • [Death Table] ( https://ohdsi.github.io/CommonDataModel/cdm54.html#death)
3. • [Multiple Cause of Death] https://forums.ohdsi.org/t/multiple-cause-of-death/4551
4. • [Store death causes in condition_occurrence table] (https://github.com/OHDSI/CommonDataModel/issues/210)

Provenance of data

EHR death data Death Registries

Use case

Any research involving Cause of Death or outcomes involving death.

The proposed convention

1. Stop using CAUSE_CONCEPT_ID, CAUSE_SOURCE_VALUE, CAUSE_SOURCE_CONCEPT_ID to contain cause of death data. Remove these fields in future CDM versions.
2. Put cause records in the clinical event tables (I.e. Condition, Observation, etc.) 
3. Condition_status_concept_id – can be used to identify primary versus secondary; is contributing cause of death needed? 
4. Type_concept_id – can be used to identify the provenance of the record; the type concepts will need to be expanded (I.e. there isn't an "EHR death record")

How would we implement the change?

• THEMIS convention and CDM documentation.
• Forward to CDM Workgroup to remove CAUSE_CONCEPT_ID, CAUSE_SOURCE_VALUE, CAUSE_SOURCE_CONCEPT_ID from subsequent versions.

How much of a change is this for the tools? In particular, Atlas.

Unknown

How would be effectively communicate this change?

THEMIS Repository, note in CDM documentation, Forums

What studies would we break if we implement this solution?

Unknown

Alternative solutions

Allow multiple death records in the Death table IF they have the same date but different causes or sources. This was approved at one time Multiple deaths on the same day. This approach was considered too confusing for analysts.

Downstream implications

• Book of OHDSI
• CDM Convention
• Data Quality Dashboard (DQD): add checks
• Analysts: query conditions for death data instead of death table.

Link to DQD check

Unknown

Related conventions/further information

(Other helpful information, if needed. i.e. related conventions, queries to evaluate source or CDM data, or any additional information)

• [Store death causes in condition_occurrence table] (https://github.com/OHDSI/CommonDataModel/issues/210)
• [Multiple deaths on the same day] (https://github.com/OHDSI/Themis/issues/5)

#Tags

#Death #Cause_Of_Death #Multiple_Death_Records

[kkondratieff]

The following is a summary compiled by Katherine Simon, from discussion on how to handle multiple death records (causes, dates) on the OHDSI Network call between the VA OMOP group and its partners:

Idea: Add a new meta data table for death – allows 1 row per person while also recording multiple Causes of Death in the Death table. Opinion: Death is its own unique event and should retain its own table, not be split up among other OMOP tables. Building on the "Move Cause of Death Concepts out of Death table" option: If Cause of Death is moved to Observation/Condition etc., we need to be able to identify it as a cause of death vs any other condition/observation. Probably using a type concept or some other similar mechanism.

Idea: The hypothesis that there is only 1 Date of Death for each person might not be true. There could be different dates reported from different agencies. Whether we keep the Date of Death in the DEATH table, move it to the PERSON table, or create a Death Meta Data table; if we are only going to allow one row per PERSON we should: Review how Birth Date, Race, Ethnicity, and Gender are handled with regards to the representation of this information changing overtime as well as the ongoing conversations about how these should be handled when these data points change over time. Recommended to use the same/similar approach to Date of Death.

If we are going to allow multiple causes of death per person, should we also allow multiple Death Dates?

[MelaniePhilofsky]

Transferring to the CDM WG, @clairblacketer for CDM 5.5 discussions.