Difference between revisions of "VRDR Testing"
Jump to navigation
Jump to search
Line 101: | Line 101: | ||
==VRDR Canary Synthetic Data== | ==VRDR Canary Synthetic Data== | ||
+ | * Patient extensions, field names | ||
+ | ** For the extensions (race, ethnicity, birthsex), the FHIR IG states the XPath for the coded value is "valueCode". | ||
+ | ** In the synthetic data, the field names are (valueCoding, valueCoding, valueCodeableConcept) | ||
+ | ** Besides the field name difference, the structure of the encoding of birthsex is different from race and ethnicity | ||
* Ethnicity | * Ethnicity | ||
** Generated document uses wrong value for system for the coded value | ** Generated document uses wrong value for system for the coded value | ||
Line 149: | Line 153: | ||
** See: https://phinvads.cdc.gov/vads/ViewValueSet.action?id=FF7F17AE-3D20-473D-9068-E77A08491242 | ** See: https://phinvads.cdc.gov/vads/ViewValueSet.action?id=FF7F17AE-3D20-473D-9068-E77A08491242 | ||
** Proper code system is SNOMED-CT or NullFlavor | ** Proper code system is SNOMED-CT or NullFlavor | ||
− | |||
==Reference== | ==Reference== |
Revision as of 10:44, 28 December 2019
Notes/items for VRDR testing
Fourquet Test Data / Spreadsheet
- Alias
- Spreadsheet uses the value Alias in column B
- FHIR IG: Table 30 has an entry for "nickname". When you state Alias, do you mean to use "nickname"?
- Race of Cancer Patient
- Spreadsheet calls for Asian Indian, but there is no corresponding code defined
- https://www.hl7.org/fhir/us/core/ValueSet-omb-race-category.html
- I did find it in PHIN VADS using the same OID for the system. See https://phinvads.cdc.gov/vads/ViewCodeSystem.action?id=2.16.840.1.113883.6.238#
- Note to self, the HL7 FHIR pages might not have all coded values. Why do they not just point to a definitive reference?
- Ethnicity
- Three patients have values Puerto Rican, Salvadoran, Cuban
- These do not match the values on this page:
- https://phinvads.cdc.gov/vads/ViewCodeSystem.action?id=2.16.840.1.113883.12.189
- Your spreadsheet says sex
- Do you mean patient.gender?
- Do you want to specify a value for birthsex?
- Decedent/Mother's Maiden Name:
- VRDR FHIR IG: 14.3 VRDR Decedent Mother describes mother, but does not specify this is a maiden name
- Is there any place else in the document that describes mother's maiden name?
- Education of Cancer patient (and others)
- You have english values that make sense, but are not part of the value set.
- Are you expecting documents to use the proper encoding? I assume so.
- Industry of Cancer patient
- Spreadsheet says: Health Insurance
- I am using the list of industries here: https://phinvads.cdc.gov/vads/ViewValueSet.action?oid=2.16.840.1.114222.4.11.7187
- There is not a direct mapping to Health Insurance. You could be more specific or let people decide which code to use.
- Language
- Spreadsheet has a row for Language, but I do not see that in the FHIR IG
- Place of death
- For Cancer patient, spreadsheet says "nursing home"
- How does this map to Table 10. Is this ?
- description
- type
- physicalType
- Date signed, date pronounced dead, time pronounced dead
- I am not sure where to pull these values from the death report.
- I did find what I believe to be the date and time of death, but these values are not the same thing.
- Table 2 in the FHIR IG has separate fields for date and time, but there is no description that does with these to make it clear to me what these are. The Canary Tool uses the term "Certified Time" and I assume uses the data from Table 2. Since I am not in this business, the different naming without cross reference leads me to guessing (not what I want to do).
- Cause of Death
- Is this to be sent as a string or as a coded value?
VRDR FHIR IG: VRDR Death Certificate Document
- Table 10 describes the Death Location
- description: The only constraint is that this is a string. Is there anything else to say about this? Might be enough
- type: Constraint says this is a codeable concept, but there is no information on the code/coding system. Just use constraints from a FHIR Location?
- physicalType: Constraint says this is a codeable concept, but there is no information on the code/coding system. Just use constraints from a FHIR Location?
- Table 14: Manner of Death
- See CONF:4393-345 which states: 7. VALUECODEABLECONCEPT, which SHALL be selected from ValueSet DispositionTypeVS http://hl7.org/fhir/ValueSet/DispositionTypeVS DYNAMIC (CONF:4393-345)
- Selecting from a Disposition Value Set does not make sense to me
- Table 18 has a row for effectiveDateTime but no text/requirement to go with it. There is a line below the table, but it does the same. It just says effectiveDateTime.
Table 20: Disposition Method
- See CONF 4393-166 which states ......SHALL be selected from ValueSet abc http://www.google.co/abc
- I don't think that is the right value set (see Table 14; is that the right value set?)
- Table 30 says there shall be a Name entry with use value nickname (cardinality 1..1)
- Do you really mean the cardinality is 1..1?
- That means that someone without a nickname is going to have an entry for nickname that is empty
- Table 30: Address
- address has cardinality of 0..1 and verb SHALL
- Is this combination correct (I thought SHALL implied cardinality >= 1, but I am not certain).
- Table 31 likely has a typo
- http://hl7.org/fhir/us/vrdr/StructureDefinition/-VRDR-Certifier
- Note the - before VRDR-Certifier
- Sections 3.1 and 3.2 are confusing
- Both have the title: VRDR Cause Of Death Condition
- Section 3.1 says code is a codeable concept, but does not break it down further. The example from canary has code/text
- Section 3.1 has this condition: http://hl7.org/fhir/us/vrdr/StructureDefinition/-VRDR-Cause-Of-Death-Condition
- Is the - before VRDR-Cause-Of-Death-Condition a typo or on purpose?
- Section 3.2 states: SHALL contain exactly one [1..1] onset[x], where onset[x]
- and then it stops. Section 3.1 actually fills this in, but that is in 3.1
- Requirements are slightly different
- Section 3.3 has typo in the title of the section.
- Sections 11.1 and 11.2, are these duplicates?
- Both sections discuss VRDR Certifier
- They have slightly different requirements
- LOINC as a coding system
- Why does the document refer to urn:oid:2.16.840.1.113883.6.1 as the value for "system" when using LOINC rather than http://loinc.org ?
- Synthetic data from canary uses http://loinc.org
VRDR Supplement
- Table 6.3.1.D1.4.1-1, Method of Disposition coding system
- Supplement says: Observation.valueCodableConcept which SHALL be selected from (CodeSystem: 2.16.840.1.114222.4.11.7379 Methods of Disposition (NCHS))
- 2.16.840.1.114222.4.11.7379 is the identifier for a value set and not for a code system
- See https://phinvads.cdc.gov/vads/ViewValueSet.action?id=F4C8B6A9-0DCB-E711-ACE2-0017A477041A
- FHIR system would be http://snomed.info/sct
- Table 6.3.1.D1.4.1-1, Pregnancy
- Supplement says: Where value data type is CD and uses values from value set: (CodeSystem: 2.16.840.1.114222.4.11.6003 Pregnancy Status (NCHS))
- 2.16.840.1.114222.4.11.6003 is the identifier for a value set and not for a code system
- See https://phinvads.cdc.gov/vads/ViewValueSet.action?id=98603205-77AC-448A-ABBD-A43D5B34E627
- PHIN VADS says coding system is "PHIN VS (CDC Local Coding System)" but does not give a value that one would use for "system", at least on that page. Google tells me the right value for system would be: urn:oid:2.16.840.1.114222.4.5.274. But, that is based on a random page I found. Would like to find a definite reference.
- Table 6.3.1.D1.4.1-1, Tobacco Use
- Supplement says: Contributory Tobacco Use (NCHS) [Observation: templateId 2.16.840.1.114222.4.11.6004]
- Not sure that the OID listed qualifies as a template ID. It is the OID of a value set (see next item).
- Supplement says: Where value data type is CD and uses values from value set: (CodeSystem: 2.16.840.1.114222.4.11.6004 Contributory Tobacco Use (NCHS)),
- The OID is not a CodeSystem OID. It is the OID of a value set.
- See https://phinvads.cdc.gov/vads/ViewValueSet.action?id=FF7F17AE-3D20-473D-9068-E77A08491242
- Code system for this value set will be either SNOMED-CT or NullFlavor
- Supplement says: Contributory Tobacco Use (NCHS) [Observation: templateId 2.16.840.1.114222.4.11.6004]
VRDR Canary Synthetic Data
- Patient extensions, field names
- For the extensions (race, ethnicity, birthsex), the FHIR IG states the XPath for the coded value is "valueCode".
- In the synthetic data, the field names are (valueCoding, valueCoding, valueCodeableConcept)
- Besides the field name difference, the structure of the encoding of birthsex is different from race and ethnicity
- Ethnicity
- Generated document uses wrong value for system for the coded value
- Synthetic data uses: urn:oid:2.16.840.1.113883.6.23
- This is the OID for race
- Proper value is urn:oid:2.16.840.1.113883.12.189
- Marital Status
- Synthetic data uses this value for system: http://hl7.org/fhir/v3/MaritalStatus
- I think the proper value is: http://terminology.hl7.org/CodeSystem/v3-MaritalStatus
- See: https://www.hl7.org/fhir/valueset-marital-status.html
- us-core-birthsex
- Synthetic data uses the wrong value for system: http://hl7.org/fhir/us/core/ValueSet/us-core-birthsex
- Correct value is: urn:oid:2.16.840.1.113883.5.1
- Occupation
- FHIR IG requires a component with code=21847-9
- Sample document uses code 21843-8
- Occupation / System and for the certifier qualification
- Sample document uses this value for system: urn:oid:2.16.840.1.114222.4.11.7186
- This is the OID of the value set and not the coding system itself
- See this page: https://phinvads.cdc.gov/vads/ViewCodeSystem.action?id=2.16.840.1.114222.4.5.314
- I think this is the proper value for system: urn:oid:2.16.840.1.114222.4.5.314
- Occupation / Industry
- Sample document uses this value for system: urn:oid:2.16.840.1.114222.4.11.7187
- This is the OID of the value set and not the coding system itself
- See this page: https://phinvads.cdc.gov/vads/ViewCodeSystem.action?id=2.16.840.1.114222.4.5.315
- I think this is the proper value for system: urn:oid:2.16.840.1.114222.4.5.315
- Death Location
- Specification requires description (string) and codes for type and physicalType
- Synthetic data is missing all of these.
- Autopsy Performed
- IG says to use ValueSet v2-0532
- Reference to that Value Set: https://www.hl7.org/fhir/v2/0532/index.html
- Synthetic data uses "http://hl7.org/fhir/ValueSet/v2-0532" for system.
- Value for system should be: http://terminology.hl7.org/CodeSystem/v2-0136
- Method of Disposition
- FHIR IG gives a bogus value for the Value Set of the disposition method
- VRDR Supplement gives the Value Set but claims that this is the value for system (not correct)
- Synthetic data uses the Value Set identifier for system when it should use the SNOMED/CT value
- (Wrong): urn:oid:2.16.840.1.114222.4.11.7379
- (Correct): http://snomed.info/sct
- Pregnancy
- Synthetic data uses "urn:oid:2.16.840.1.114222.4.11.6003" for system, but this is the Value Set identifier.
- See: https://phinvads.cdc.gov/vads/ViewValueSet.action?id=98603205-77AC-448A-ABBD-A43D5B34E627
- Note the code system is PHIN VS (CDC Local Coding System)
- I think the right value for system is urn:oid:2.16.840.1.114222.4.5.274
- Tobacco Use
- Synthetic data uses "urn:oid:2.16.840.1.114222.4.11.6004" for system, but this is the Value Set identifier.
- See: https://phinvads.cdc.gov/vads/ViewValueSet.action?id=FF7F17AE-3D20-473D-9068-E77A08491242
- Proper code system is SNOMED-CT or NullFlavor