[15/2/4 上午11:23:12] wanghaisheng: http://hl7-fhir.github.io/datatypes.html in section 1.22.0.4 there is “A code system version may also be supplied. If the meaning of codes within the code system is consistent across releases, this is not required. The version SHOULD be exchanged when this the system does not maintain consistent definitions across versions. If a value set is referenced, and the value set defines the code system, and the code system definition specifies a version, the version SHALL be the same as in the value set. Note that the following systems SHOULD always have a version specified: ”
If a value set is referenced, and the value set defines the code system, is this value set define cs a typo one or ?
[15/2/4 上午11:25:10] Grahame Grieve: "is this value set define cs a typo one or " - ?
[15/2/4 上午11:26:11] wanghaisheng: the value set defines the code system, does value set reference code from one code system, how could value set define a code system？
[15/2/4 上午11:34:55] Grahame Grieve: Valueset.define
[15/2/4 上午11:39:36] wanghaisheng: if here value set define code system means value set define its own codes instead choose from external code system, what does the following description “ the code system definition specifies a version, the version SHALL be the same as in the value set”
[15/2/4 上午11:42:52] Grahame Grieve: reference to ValueSet.define.version - if the value set has one of these, then that version must match the stated version in the Coding that refers to it
[15/2/4 上午11:49:53] wanghaisheng: i know ,when value set defines its own codes, you simply call it code system,then I could not get the description about the scope and usage in 6.22.1 ,is value set more broader than code system or?
[15/2/4 上午11:52:06] Grahame Grieve: value sets can cross code systems, and they can be tied to no particular version.
[15/2/4 上午11:54:38] Russell Hamm: Can someone point me to some documentatio outlinign the use case for a value set being able to define its own codes?
[15/2/4 上午11:55:17] Russell Hamm: I want to make sure I understand the use case thoroughly
[15/2/4 上午11:55:22] wanghaisheng: 6.22.1 “Value sets that define their own codes are effectlvely providing inline code system definitions. These are intended for small, simple code systems that are found throughout the implementation context (e.g. lists of wards, status codes, enumerations). The inline code system definition is not intended to represent large publically defined terminologies such as LOINC etc - these have their own distributions formats . ”
[15/2/4 上午11:55:45] Russell Hamm: that works. thanks!
[15/2/4 下午12:00:32] wanghaisheng: another one in the 1.22.0.4 , in the Constraints. “ If a valueSet is provided, a system URI SHALL also be provided. ” is this system the coding,system one or ?
[15/2/4 下午12:01:36] Lloyd McKenzie: y
[15/2/4 下午12:03:55] wanghaisheng: if so, means there shall be a ValueSet.define.version or ValueSet.compose.include.version and this two version shall be the same
[15/2/4 下午12:11:52] Lloyd McKenzie: No. It says that if you specify a value set for a coding, you have to declare the system. You don't have to declare a version and the value set doesn't have to declare a version either.
[15/2/4 下午12:17:16] wanghaisheng: if you specify a value set for a coding, if value set define its own code,we shall populate the ValueSet.define.system not coding.system, if value set chooses codes externally ,we shall populate the coding.system ,and perhaps the ValueSet.compose.include.system
[15/2/4 下午12:18:18] Lloyd McKenzie: No, you have to populate the coding .system if you specify coding.valueset
[15/2/4 下午12:18:32] Lloyd McKenzie: coding.valueset must always be ignorable
[15/2/4 下午12:19:05] Lloyd McKenzie: As well, the referenced value set might easily reference multiple code systems, so having coding.valueset isn't sufficient to be able to infer coding.system
[15/2/4 下午12:20:09] wanghaisheng: it is said “The system URI SHALL NOT contain a reference to a value set. (If the value set defines its own codes, then the correct value for the system is ValueSet.define.system, and the value set contains a direct reference to the value set resource. ”
[15/2/4 下午12:20:56] Lloyd McKenzie: Right. the system URI and the value set URI have to be distinct
[15/2/4 下午12:22:28] wanghaisheng: how could this happen when value set define its own code, this value set is a code system now……what is the value for coding.system.take http://hl7-fhir.github.io/RS-link.xml.html as an example
[15/2/4 下午12:23:50] Lloyd McKenzie: If a value set defines its own code system, it still assigns a distinct URI to the system. Essentially the URI implied by ValueSet.id != ValueSet.define.system
[15/2/4 下午12:24:42] Lloyd McKenzie: The content of ValueSet.define.system doesn't have anything to do with the URL of the value set itself - so long as it's a unique URL for the code system being defined within the value set.
[15/2/4 下午12:27:59] wanghaisheng: could you explain more with example http://hl7-fhir.github.io/RS-link.xml.html
[15/2/4 下午12:29:34] Lloyd McKenzie: So the value set has the URL: http://hl7.org/fhir/ValueSet/RS-link. The system has the URL http://hl7.org/fhir/RS-link
[15/2/4 下午12:29:52] Lloyd McKenzie: Similar, but not the same.
[15/2/4 下午12:35:14] wanghaisheng: we always take http://hl7.org/fhir/ValueSet/RS-link. as Coding.system ,whether this code system define by a value set or somewhere else?
[15/2/4 下午12:36:44] Lloyd McKenzie: No. http://hl7.org/fhir/ValueSet/RS-linkcannot be used in Coding.system. The value for Coding.system is http://hl7.org/fhir/RS-link
[15/2/4 下午12:37:13] Lloyd McKenzie: http://hl7.org/fhir/ValueSet/RS-link can be used in Coding.valueset.reference
[15/2/4 下午12:37:37] Lloyd McKenzie: (And if you include Coding.valueset, you must also populate Coding.system)
[15/2/4 下午12:43:29] wanghaisheng: using ValueSet.define.system or ValueSet.compose.include.system as Coding.system ,
[15/2/4 下午12:44:41] wanghaisheng: earlier what i want to say is this one ,not the version :(
[15/2/4 下午12:51:02] Lloyd McKenzie: Correct
[15/2/4 下午12:52:34] Lloyd McKenzie: And Coding.version (if you bother to populate it at all) would be ValueSet.define.version or ValueSet.compose.include.version, provided that those are specified.
[15/2/4 下午12:54:30] wanghaisheng: in section 1.22.0.5 “so Coding is mostly used in extensions, which are usually intended to be defined for a well-controlled context of use. ” could you explain more?
[15/2/4 下午12:56:39] wanghaisheng: i have not seen any example about the valudeCoding used
[15/2/4 下午12:56:50] Lloyd McKenzie: Most places we need a code, we'll use either the "code" data type - because the code system is fixed by FHIR, or we'll use CodeableConcept because there could be multiple translations or possibly just original text. "Coding" as a stand-alone data type only makes sense in very limited situations where translation and original text don't make sense.
[15/2/4 下午12:58:03] Lloyd McKenzie: One place I can think of off the top of my head where we use Coding is DataElement.code. In that case, we're identifying where the meaning of the data element comes from. Translations don't make sense, nor does original text.
[15/2/4 下午12:59:04] Lloyd McKenzie: It gets used in the financial resources a bit too because translations aren't allowed - apparently the insurers all agree on a single set of codes.
[15/2/4 下午1:00:23] Grahame Grieve: y. typically in the financial area, there's actual incentive to align business practices. Amazing isn' it ;-)
[15/2/4 下午1:00:37] wanghaisheng: :D
[15/2/4 下午2:31:33] wanghaisheng: “Note that the units element will often contain text that is actually a valid UCUM unit, but it cannot be assumed that the units element actually contains a valid UCUM unit. ” does this mean usually the value of units element is a valid UCUM unit, but we can’t say it is a valid UCUM unit text when we get an units element
[15/2/4 下午2:32:45] Grahame Grieve: well, if you can validate it, then you can say that it is
[15/2/4 下午2:32:53] wanghaisheng: :D
[15/2/4 下午2:32:55] Grahame Grieve: but it doesn't have to be
[15/2/4 下午2:33:44] wanghaisheng: in China we just don`t use UCUM we have our national defined value set for that
[15/2/4 下午2:34:46] Grahame Grieve: well, you don't have to use UCUM. the units is just a text display
[15/2/4 下午2:34:55] Grahame Grieve: and if you want to code it, you can do that in system/code
[15/2/4 下午2:35:20] wanghaisheng: gottcha
[15/2/4 下午2:38:18] wanghaisheng: 1.22.0.6.0 there is a typo “Coding is used in the following places: Range, Ratio, SampledData, Supply, Group, OralHealthClaim, PharmacyClaim, CarePlan, Substance, QuestionnaireAnswers, ProfessionalClaim, Observation, InstitutionalClaim, VisionClaim, MedicationDispense, MedicationPrescription, MedicationStatement, VisionPrescription, Immunization, Medication, MedicationAdministration, NutritionOrder, Contract and Specimen” not Coding, Quantity
[15/2/4 下午2:40:43] Grahame Grieve: ta fixed
[15/2/4 下午2:40:51] Grahame Grieve: I have implemented the modifier :in
[15/2/4 下午2:40:53] Grahame Grieve: http://fhir-dev.healthintersections.com.au/open/Observation?name:in=http://hl7.org/fhir/vs/observation-CCDAVitalSignResult
[15/2/4 下午2:41:03] Grahame Grieve: you must refer to a value set that is on the fhir-dev server
[15/2/4 下午2:41:19] Grahame Grieve: I'm not sure what should happen if the value set reference is illegal. Error? Ignore?
[15/2/4 下午2:44:10] Lloyd McKenzie: I think either would be a legitimate response.
[15/2/4 下午3:13:23] wanghaisheng: in 1.22.0.7 “ The range flag on the low or high elements cannot be present. Note that the Range type should not be used to represent out of range measurements: A quantity type with the comparator element should be used instead. ” there is no range flag element in Quantity data type
[15/2/4 下午3:57:07] wanghaisheng: are Period type what we learn in MATH the [a,b] ?
[15/2/4 下午4:29:50] Grahame Grieve: it should be "comparator" not "range" at the start of the sentence. I've fixed it
[15/2/4 下午4:29:58] Grahame Grieve: Period - like that, yes
[15/2/4 下午4:30:57] wanghaisheng: ok
[15/2/4 下午5:05:57] wanghaisheng: For given names, initials may be used in place of the full name if that is all that is recorded. what does this mean
[15/2/4 下午5:06:43] wanghaisheng: if we did not catch the full name ,we can use the initial part for short?
[15/2/4 下午5:13:18] Grahame Grieve: Y. Grieve g. So you put just g for the given
[15/2/4 下午5:14:43] wanghaisheng: “ Systems that do not support as many name parts as are provided in an instance may wish to append some of the parts together using spaces ”
[15/2/4 下午5:15:29] wanghaisheng: does this mean we can append different parts in text element with white space, but we can’t use white space in any one part value
[15/2/4 下午5:17:05] wanghaisheng: “The parts of a name SHOULD NOT contain whitespace”
[15/2/4 上午11:23:12] wanghaisheng: http://hl7-fhir.github.io/datatypes.html in section 1.22.0.4 there is “A code system version may also be supplied. If the meaning of codes within the code system is consistent across releases, this is not required. The version SHOULD be exchanged when this the system does not maintain consistent definitions across versions. If a value set is referenced, and the value set defines the code system, and the code system definition specifies a version, the version SHALL be the same as in the value set. Note that the following systems SHOULD always have a version specified: ” If a value set is referenced, and the value set defines the code system, is this value set define cs a typo one or ? [15/2/4 上午11:25:10] Grahame Grieve: "is this value set define cs a typo one or " - ? [15/2/4 上午11:26:11] wanghaisheng: the value set defines the code system, does value set reference code from one code system, how could value set define a code system？ [15/2/4 上午11:34:55] Grahame Grieve: Valueset.define [15/2/4 上午11:39:36] wanghaisheng: if here value set define code system means value set define its own codes instead choose from external code system, what does the following description “ the code system definition specifies a version, the version SHALL be the same as in the value set” [15/2/4 上午11:42:52] Grahame Grieve: reference to ValueSet.define.version - if the value set has one of these, then that version must match the stated version in the Coding that refers to it [15/2/4 上午11:49:53] wanghaisheng: i know ,when value set defines its own codes, you simply call it code system,then I could not get the description about the scope and usage in 6.22.1 ,is value set more broader than code system or? [15/2/4 上午11:52:06] Grahame Grieve: value sets can cross code systems, and they can be tied to no particular version. [15/2/4 上午11:54:38] Russell Hamm: Can someone point me to some documentatio outlinign the use case for a value set being able to define its own codes? [15/2/4 上午11:55:17] Russell Hamm: I want to make sure I understand the use case thoroughly [15/2/4 上午11:55:22] wanghaisheng: 6.22.1 “Value sets that define their own codes are effectlvely providing inline code system definitions. These are intended for small, simple code systems that are found throughout the implementation context (e.g. lists of wards, status codes, enumerations). The inline code system definition is not intended to represent large publically defined terminologies such as LOINC etc - these have their own distributions formats . ” [15/2/4 上午11:55:45] Russell Hamm: that works. thanks! [15/2/4 下午12:00:32] wanghaisheng: another one in the 1.22.0.4 , in the Constraints. “ If a valueSet is provided, a system URI SHALL also be provided. ” is this system the coding,system one or ? [15/2/4 下午12:01:36] Lloyd McKenzie: y [15/2/4 下午12:03:55] wanghaisheng: if so, means there shall be a ValueSet.define.version or ValueSet.compose.include.version and this two version shall be the same [15/2/4 下午12:11:52] Lloyd McKenzie: No. It says that if you specify a value set for a coding, you have to declare the system. You don't have to declare a version and the value set doesn't have to declare a version either. [15/2/4 下午12:17:16] wanghaisheng: if you specify a value set for a coding, if value set define its own code,we shall populate the ValueSet.define.system not coding.system, if value set chooses codes externally ,we shall populate the coding.system ,and perhaps the ValueSet.compose.include.system [15/2/4 下午12:18:18] Lloyd McKenzie: No, you have to populate the coding .system if you specify coding.valueset [15/2/4 下午12:18:32] Lloyd McKenzie: coding.valueset must always be ignorable [15/2/4 下午12:19:05] Lloyd McKenzie: As well, the referenced value set might easily reference multiple code systems, so having coding.valueset isn't sufficient to be able to infer coding.system [15/2/4 下午12:20:09] wanghaisheng: it is said “The system URI SHALL NOT contain a reference to a value set. (If the value set defines its own codes, then the correct value for the system is ValueSet.define.system, and the value set contains a direct reference to the value set resource. ” [15/2/4 下午12:20:56] Lloyd McKenzie: Right. the system URI and the value set URI have to be distinct [15/2/4 下午12:22:28] wanghaisheng: how could this happen when value set define its own code, this value set is a code system now……what is the value for coding.system.take http://hl7-fhir.github.io/RS-link.xml.html as an example [15/2/4 下午12:23:50] Lloyd McKenzie: If a value set defines its own code system, it still assigns a distinct URI to the system. Essentially the URI implied by ValueSet.id != ValueSet.define.system [15/2/4 下午12:24:42] Lloyd McKenzie: The content of ValueSet.define.system doesn't have anything to do with the URL of the value set itself - so long as it's a unique URL for the code system being defined within the value set. [15/2/4 下午12:27:59] wanghaisheng: could you explain more with example http://hl7-fhir.github.io/RS-link.xml.html [15/2/4 下午12:29:34] Lloyd McKenzie: So the value set has the URL: http://hl7.org/fhir/ValueSet/RS-link. The system has the URL http://hl7.org/fhir/RS-link [15/2/4 下午12:29:52] Lloyd McKenzie: Similar, but not the same. [15/2/4 下午12:35:14] wanghaisheng: we always take http://hl7.org/fhir/ValueSet/RS-link. as Coding.system ,whether this code system define by a value set or somewhere else? [15/2/4 下午12:36:44] Lloyd McKenzie: No. http://hl7.org/fhir/ValueSet/RS-link cannot be used in Coding.system. The value for Coding.system is http://hl7.org/fhir/RS-link [15/2/4 下午12:37:13] Lloyd McKenzie: http://hl7.org/fhir/ValueSet/RS-link can be used in Coding.valueset.reference [15/2/4 下午12:37:37] Lloyd McKenzie: (And if you include Coding.valueset, you must also populate Coding.system) [15/2/4 下午12:43:29] wanghaisheng: using ValueSet.define.system or ValueSet.compose.include.system as Coding.system , [15/2/4 下午12:44:41] wanghaisheng: earlier what i want to say is this one ,not the version :( [15/2/4 下午12:51:02] Lloyd McKenzie: Correct [15/2/4 下午12:52:34] Lloyd McKenzie: And Coding.version (if you bother to populate it at all) would be ValueSet.define.version or ValueSet.compose.include.version, provided that those are specified. [15/2/4 下午12:54:30] wanghaisheng: in section 1.22.0.5 “so Coding is mostly used in extensions, which are usually intended to be defined for a well-controlled context of use. ” could you explain more? [15/2/4 下午12:56:39] wanghaisheng: i have not seen any example about the valudeCoding used [15/2/4 下午12:56:50] Lloyd McKenzie: Most places we need a code, we'll use either the "code" data type - because the code system is fixed by FHIR, or we'll use CodeableConcept because there could be multiple translations or possibly just original text. "Coding" as a stand-alone data type only makes sense in very limited situations where translation and original text don't make sense. [15/2/4 下午12:58:03] Lloyd McKenzie: One place I can think of off the top of my head where we use Coding is DataElement.code. In that case, we're identifying where the meaning of the data element comes from. Translations don't make sense, nor does original text. [15/2/4 下午12:59:04] Lloyd McKenzie: It gets used in the financial resources a bit too because translations aren't allowed - apparently the insurers all agree on a single set of codes. [15/2/4 下午1:00:23] Grahame Grieve: y. typically in the financial area, there's actual incentive to align business practices. Amazing isn' it ;-) [15/2/4 下午1:00:37] wanghaisheng: :D [15/2/4 下午2:31:33] wanghaisheng: “Note that the units element will often contain text that is actually a valid UCUM unit, but it cannot be assumed that the units element actually contains a valid UCUM unit. ” does this mean usually the value of units element is a valid UCUM unit, but we can’t say it is a valid UCUM unit text when we get an units element [15/2/4 下午2:32:45] Grahame Grieve: well, if you can validate it, then you can say that it is [15/2/4 下午2:32:53] wanghaisheng: :D [15/2/4 下午2:32:55] Grahame Grieve: but it doesn't have to be [15/2/4 下午2:33:44] wanghaisheng: in China we just don`t use UCUM we have our national defined value set for that [15/2/4 下午2:34:46] Grahame Grieve: well, you don't have to use UCUM. the units is just a text display [15/2/4 下午2:34:55] Grahame Grieve: and if you want to code it, you can do that in system/code [15/2/4 下午2:35:20] wanghaisheng: gottcha [15/2/4 下午2:38:18] wanghaisheng: 1.22.0.6.0 there is a typo “Coding is used in the following places: Range, Ratio, SampledData, Supply, Group, OralHealthClaim, PharmacyClaim, CarePlan, Substance, QuestionnaireAnswers, ProfessionalClaim, Observation, InstitutionalClaim, VisionClaim, MedicationDispense, MedicationPrescription, MedicationStatement, VisionPrescription, Immunization, Medication, MedicationAdministration, NutritionOrder, Contract and Specimen” not Coding, Quantity [15/2/4 下午2:40:43] Grahame Grieve: ta fixed [15/2/4 下午2:40:51] Grahame Grieve: I have implemented the modifier :in [15/2/4 下午2:40:53] Grahame Grieve: http://fhir-dev.healthintersections.com.au/open/Observation?name:in=http://hl7.org/fhir/vs/observation-CCDAVitalSignResult [15/2/4 下午2:41:03] Grahame Grieve: you must refer to a value set that is on the fhir-dev server [15/2/4 下午2:41:19] Grahame Grieve: I'm not sure what should happen if the value set reference is illegal. Error? Ignore? [15/2/4 下午2:44:10] Lloyd McKenzie: I think either would be a legitimate response. [15/2/4 下午3:13:23] wanghaisheng: in 1.22.0.7 “ The range flag on the low or high elements cannot be present. Note that the Range type should not be used to represent out of range measurements: A quantity type with the comparator element should be used instead. ” there is no range flag element in Quantity data type [15/2/4 下午3:57:07] wanghaisheng: are Period type what we learn in MATH the [a,b] ? [15/2/4 下午4:29:50] Grahame Grieve: it should be "comparator" not "range" at the start of the sentence. I've fixed it [15/2/4 下午4:29:58] Grahame Grieve: Period - like that, yes [15/2/4 下午4:30:57] wanghaisheng: ok [15/2/4 下午5:05:57] wanghaisheng: For given names, initials may be used in place of the full name if that is all that is recorded. what does this mean [15/2/4 下午5:06:43] wanghaisheng: if we did not catch the full name ,we can use the initial part for short? [15/2/4 下午5:13:18] Grahame Grieve: Y. Grieve g. So you put just g for the given [15/2/4 下午5:14:43] wanghaisheng: “ Systems that do not support as many name parts as are provided in an instance may wish to append some of the parts together using spaces ” [15/2/4 下午5:15:29] wanghaisheng: does this mean we can append different parts in text element with white space, but we can’t use white space in any one part value [15/2/4 下午5:17:05] wanghaisheng: “The parts of a name SHOULD NOT contain whitespace”