Library Resources

Owned by Erik Nielsen
Last updated: 30-07-2024 by malthe lange
11 min read

Library Resources for Automated Processing-type Rules

The Library resources described in this section are either production-ready or slated for production (requiring approval for production use).

The Fallback Library and Rule

The Fallback Library contains a rule that ensures creation of a Task that draws attention to a measurement needing evaluation. It is intended for implicit use in case no other automated processing type Library has been set up for a given ServiceRequest through its ActivityDefinition.

The Library has:

  • .type set to automated-processing

  • .parameter

    • .use set to in for input

    • .type set to Observation

The Library rule sets up instructions that causes production of a:

  • a Task with:

    • .category set to MeasurementForAssessment (a Coding signifying a measurement should be assessed/evaluated)

    • .priority set to routine

    • .description set to “Måling til vurdering” which is Danish for measurement to be assessed

    • further elements set as described in Output from automated processing

The Null Library and Rule

The Null Library contains a rule that causes no creation of Task, ClinicalImpression or Communication. It is intended for explicit use in cases where no Task, ClinicalImpression nor Communication is desired as a result of performing automated processing. It will not override other automated-processing type Library associated with an ActivityDefinition and is therefore expected to be used as the sole Library associated.

Example use: The Null Library can be associated with an ActivityDefinition specifying answering of Questionnaire intended for assessing some quality of a measurement (specified in a different ActivityDefinition). When answered, the QuestionnaireResponse is intended to be subjected to a logic-library type Library which determines a quality code based on the answers. Even when/if submitted, Practitioners might have little interest in the QuestionnaireResponse and in order not to get Task, ClinicalImpression nor Communication about it, this can be planned ahead and avoided by associating the Null Library.

The Library has:

  • .type set to automated-processing

  • .parameter

    • .use set to in for input

    • .type set to Resource

Assessment of Absolute Reference Ranges Library and Rule

This Library contains a generic rule that performs assessment of an Observation based on absolute reference range(s) automatically copied from a ServiceRequest referencing an ActivityDefinition that references the Library. The rule in this Library is automatically performed as a reaction to submitting of the Observation. This Library can be referenced from an ActivityDefinition along with other Library, even the Relative Reference Ranges Library described below.

The Library has:

  • .type set to automated-processing

  • .parameter

    • .use set to in for input

    • .type set to Observation

The Library rule sets up instructions that causes production of:

 

The rule processes absolute reference ranges only, that is, Observation.referenceRange.type where the Coding has:

  • system: urn:oid:1.2.208.184.100.1

  • code: RAL or GAL, for red alarm absolute reference range and yellow alarm absolute reference range, respectively.

If no absolute reference range is found, the rule causes creation of a single Task asking the clinician to assess the observation.

If an observation contains components, it is assumed that all values are contained in components.

The rule accepts observation values and reference ranges without units. However if both have a unit specified, they must match.

Error handling for different situations which result in the rule setting up instructions for creation of one or two Task:

Situation

Generic Task

Possible Specific Task (Stating what was wrong)

Situation

Generic Task

Possible Specific Task (Stating what was wrong)

No absolute reference range found

Task (Stating that measurement needs manual assessment):

  • .category with .code = MeasurementForAssessment, .system = http://ehealth.sundhed.dk/cs/task-category, .display = Need assessment of measurement

  • .description = Måling til vurdering

  • .restrictionCategory with .code = measurement-monitoring, system = http://ehealth.sundhed.dk/cs/restriction-category, .display = Monitoring of measurement(s)

  • .priority = routine

  • .focus = reference to measurement being processed, for instance Observation, QuestionnaireResponse or Media

No additional task.

Unit mismatch between reference range and observation value

Task (Stating that measurement needs manual assessment as auto processing failed):

  • .category with .code = MeasurementForAssessmentFailureInAutoProcessing, .system = http://ehealth.sundhed.dk/cs/task-category, .display = Measurement Assessment failure during automated processing

  • .description = Vurdering af måling er fejlet under automatisk behandling

  • .restrictionCategory with .code = measurement-monitoring, system = http://ehealth.sundhed.dk/cs/restriction-category, .display = Monitoring of measurement(s)

  • .priority = routine

  • .focus = reference to measurement being processed, for instance Observation, QuestionnaireResponse or Media

Task:

  • .category with .code = RefRangeFixingNeeded, .system = http://ehealth.sundhed.dk/cs/task-category, .display = Missing in force reference base required for automatic processing

  • .description = Automatiseret behandling af måledata fejlede. Grænseværdi krævet af automatiseret behandling skal udbedres

  • .restrictionCategory with .code = measurement-monitoring, system = http://ehealth.sundhed.dk/cs/restriction-category, .display = Monitoring of measurement(s)

  • .priority = routine

  • .focus = reference to ServiceRequest referenced by the measurement being processed

Missing observation value

Task (Stating that measurement needs manual assessment as auto processing failed):

  • .category with .code = MeasurementForAssessmentFailureInAutoProcessing, .system = http://ehealth.sundhed.dk/cs/task-category, .display = Measurement Assessment failure during automated processing

  • .description = Vurdering af måling er fejlet under automatisk behandling

  • .restrictionCategory with .code = measurement-monitoring, system = http://ehealth.sundhed.dk/cs/restriction-category, .display = Monitoring of measurement(s)

  • .priority = routine

  • .focus = reference to measurement being processed, for instance Observation, QuestionnaireResponse or Media

Task:

  • .category with .code = LibraryUseMismatchFixingNeeded, .system = http://ehealth.sundhed.dk/cs/task-category, .display = Input and rule did not match

  • .description = Automatiseret behandling af måledata fejlede. Aktivitet i plan og tilknyttet regel passer ikke sammen. Det bør overvejes om borger skal have ny plan.

  • .restrictionCategory with .code = measurement-monitoring, system = http://ehealth.sundhed.dk/cs/restriction-category, .display = Monitoring of measurement(s)

  • .priority = routine

  • .focus = reference to ServiceRequest referenced by the measurement being processed

Assessment of Relative Reference Ranges Library and Rule

This Library is similar to the Absolute Reference Range library (see above) in some aspects: Input and output are identical. The difference is the the type of reference ranges that the rule handles. Absolute and relative reference range libraries can co-exist and can both process the same Observation. They will each handle their corresponding type of reference ranges associated with the Observation.

The rule processes relative reference ranges only, that is, Observation.referenceRange.type where the Coding has:

  • system: http://ehealth.sundhed.dk/cs/reference-range-type

  • code: RELRAL or RELGAL, for red alarm relative reference range and yellow alarm relative reference range, respectively.

A vital difference from the rule in the Absolute Reference Range Library is that Relative Reference Range Library rule must have a reference base to compare against.

A relative reference range can be stated:

  • in same unit as the Observation

  • as a percentage or, in case the Observation is already expressed with unit percentage (as is the case for oxygen saturation, for instance), as a percentage point.

Example relative reference range as -5 percentage point with reference base 90% for oxygen saturation measurement

The oxygen saturation measurement unit is expressed as percent (system=http://unitsofmeasure.org, code=%), the reference base likewise and in this case with a value of 90.

A red alarm relative reference range given as -5 percentage point would be given by a SimpleQuantity for the reference range’s low element with:

  • value=-5

  • unit=percentage point

  • system=http://ehealth.sundhed.dk/cs/ehealth-unit-code

  • code=percentagepoint

Thus, the rule would treat an Observation with value 85% as a red alarm case for this relative reference range and current choice of reference base with value 90%.

The rule handles different situations which result in the rule setting up instructions for creation of one or two Task:

Situation

Generic Task

Possible Specific Task (Stating what was wrong)

Situation

Generic Task

Possible Specific Task (Stating what was wrong)

Neither reference base nor relative reference range(s) found

Task:

  • .category with .code = MeasurementForAssessmentNotTriaged, .system = http://ehealth.sundhed.dk/cs/task-category, .display = Need assessment of measurement where triage was not performed during automated processing

  • .description = Måling til vurdering, hvor triagering ikke er udført af automatiseret behandling

  • .restrictionCategory with .code = measurement-monitoring, system = http://ehealth.sundhed.dk/cs/restriction-category, .display = Monitoring of measurement(s)

  • .priority = routine

  • .focus = reference to measurement being processed, for instance Observation, QuestionnaireResponse or Media

No additional task.

No reference base found

Task:

  • .category with .code = RefBaseNeeded, .system = http://ehealth.sundhed.dk/cs/task-category, .display = Missing in force reference base required for automatic processing

  • .description = Mangler gældende referenceværdi påkrævet for automatisk behandling

  • .restrictionCategory with .code = measurement-monitoring, system = http://ehealth.sundhed.dk/cs/restriction-category, .display = Monitoring of measurement(s)

  • .priority = routine

  • .focus = reference to ServiceRequest referenced by the measurement being processed

No relative reference range found

Task:

  • .category with .code = RefRangeNeeded, .system = http://ehealth.sundhed.dk/cs/task-category, .display = Missing reference range required for automated processing

  • .description = Mangler grænseværdi påkrævet for automatisk behandling

  • .restrictionCategory with .code = measurement-monitoring, system = http://ehealth.sundhed.dk/cs/restriction-category, .display = Monitoring of measurement(s)

  • .priority = routine

  • .focus = reference to ServiceRequest referenced by the measurement being processed

Library for Assessment of Questionnaire Response

This Library contains a generic rule that performs assessment of a QuestionnaireResponse in relation to any answer significances defined in the corresponding Questionnaire.

During processing of the rule, each answer significance defined in the Questionnaire (see https://ehealth.sundhed.dk/fhir/StructureDefinition-ehealth-questionnaire.html#answer-significance ) on questions and their linked answers in the QuestionnaireResponse are traversed and checked whether the condition(s) of the answer significance is met by the linked answer. When this is the case, a finding basis is added to internal bookkeeping. Ultimately, the list of finding basis determines the created ClinicalImpression finding and the priority of the created task:

  • In case the list of finding basis contains at least one red question/answer pair, the priority is set to asap

  • In case the list of finding basis contains at least one yellow question/answer pair and no red question/answer pair , the priority is set to urgent

  • Otherwise, the priority is set to routine

 

The Library has:

  • .type set to automated-processing

  • .parameter

    • .use set to in for input

    • .type set to QuestionnaireResponse

  • .parameter

    • .use set to in for input

    • .type set to Questionnaire

  • .parameter

    • .use set to in for input

    • .type set to ServiceRequest

The Library rule sets up instructions that causes production of:

  • A ClinicalImpresssion:

    • .finding as follows:

      • If the triggered red set is not empty, a finding.item with a Coding:

        • system = http://ehealth.sundhed.dk/cs/clinicalimpression-finding-codes

        • code = red

        • display = Red overall assessment

      • If the triggered yellow set is not empty, a finding.item with a Coding:

        • system = http://ehealth.sundhed.dk/cs/clinicalimpression-finding-codes

        • code = yellow

        • display = yellow overall assessment

      • If both triggered sets are empty, a finding.item with a Coding:

        • system = http://ehealth.sundhed.dk/cs/clinicalimpression-finding-codes

        • code = green

        • display = Green overall assessment

      • If the Questionnaire does contain answer significance definitions but none have been triggered, a finding.item with a Coding:

        • system = http://ehealth.sundhed.dk/cs/clinicalimpression-finding-codes

        • code = no-effective-answer-significance

        • display = Ingen triageringsindikator for spørgsmål/svar-kombination har fundet anvendelse

      • If the Questionnaire does not contain answer significance definitions, a finding.item with a Coding:

        • system = http://ehealth.sundhed.dk/cs/clinicalimpression-finding-codes

        • code = no-answer-significance-defined

        • display = Ingen triageringsindikator defineret i spørgeskemaet

    • list of .ehealth-questionnaireresponse-finding-basis for each basis for the overall ClinicalImpression as follows:

      • linkId set to linkId for the Questionnaire item/QuestionnaireResponse item pair which triggered the finding basis

      • value set to value of the answer which triggered the finding basis

      • finding set to green-question-answer, yellow-question-answer or red-question-answer from http://ehealth.sundhed.dk/vs/clinicalimpression-finding-codes based on signifiance on triggered ehealth-questionnaire-answerSignificance

      • ehealth-questionnaire-answerSignificance is the triggered answerSignificance which is gathered from Questionnaire item

  • A Task related to the ClinicalImpression with:

    • priority set to the priority value determined above.

 

A difference between the finding in finding.item.code and the finding basis in ehealth-questionnaire-finding-basis is that the former is a conclusion or summary while the latter provides the details that led to the conclusion. In ehealth-questionnaire-finding-basis there may be multiple entries for the same question-answer pair and those entries may or may not have the same significance (color). That depends on the how the answer (or lack of answer) fits with answer significance definitions in the Questionnaire.

Assessment of TeleCare Nord COPD Questionnaire

This Library contains a rule that performs assessment of a QuestionnaireResponse which answers the Questionnaire TeleCare Nord COPD Questionnaire.

The rule contains two sets of question/answer value pairs:

  • A red trigger set of question/answer pairs

  • A yellow trigger set of question/answer pairs

During processing of the rule, all questions and answers in the QuestionnaireResponse are traversed and checked whether each particular question/answer pair is defined as a red and yellow trigger, respectively. When this is the case, the question/answer pair is added to internal bookkeeping in the form of a triggered red set and triggered yellow set, respectively. Ultimately, the triggered sets determine a priority as follows:

  • In case the triggered red set contains at least one question/answer pair, the priority is set to asap

  • In case the triggered red set is empty and the triggered yellow set contains at least one question/answer pair, the priority is set to urgent

  • Otherwise, the priority is set to routine

 

The Library has:

  • .type set to automated-processing

  • .parameter

    • .use set to in for input

    • .type set to QuestionnaireResponse

  • .parameter

    • .use set to in for input

    • .type set to Questionnaire

  • .parameter

    • .use set to in for input

    • .type set to ServiceRequest

The Library rule sets up instructions that causes production of:

  • A ClinicalImpresssion:

    • .finding as follows:

      • If the triggered red set is not empty, a finding.item with a Coding:

        • system = http://ehealth.sundhed.dk/cs/clinicalimpression-finding-codes

        • code = red-question-answer

        • display = Rød spørgsmål/svar-kombination fundet i spørgeskemabesvarelse

      • If the triggered yellow set is not empty, a finding.item with a Coding:

        • system = http://ehealth.sundhed.dk/cs/clinicalimpression-finding-codes

        • code = yellow-question-answer

        • display = Gul spørgsmål/svar-kombination fundet i spørgeskemabesvarelse

      • If both triggered sets are empty, a finding.item with a Coding:

        • system = http://ehealth.sundhed.dk/cs/clinicalimpression-finding-codes

        • code = green-question-answer

        • display = Grøn spørgsmål/svar-kombination fundet i spørgeskemabesvarelse. Når anført som opsummering for hel spørgeskemabesvarelse er der ikke fundet røde eller gule spørgsmål/svar-kombinationer

  • A Task related to the ClinicalImpression with:

    • priority set to the priority value determined above.

Library Resources for Calculation-type Rules

The Library resources described in this section are either production-ready or slated for production (requiring approval for production use).

Library for Calculating Quality of Height Observation for Use in BMI Calculation

The Library for calculating the quality of a height observation can be used to evaluate a height observation for intended use in a Body Mass Index (BMI) calculation. The older a height observation is, the less likely that it is still applicable and reflecting the current height. When evaluated to be too old (with code Low, see below), a client such as a citizen solution may request a fresh height measurement from the patient and use that in the BMI calculation.

The Library has:

  • .type set to logic-library

  • .parameter[0]

    • .use set to out for output

    • .type set to Coding

  • .parameter[1]

    • .use set to in for input

    • .type set to Observation

  • .parameter[2]

    • .use set to in for input

    • .type set to ServiceRequest

The input Observation must have

It is not necessary to provide the ServiceRequest as input. The Libary$evaluate operation will retrieve it by resolving the basedOn reference in the Observation.

Output will be a code from the http://hl7.org/fhir/v3/ObservationInterpretation CodeSystem:

Output Coding

Evaluation

Output Coding

Evaluation

code=IND, system=http://hl7.org/fhir/v3/ObservationInterpretation, display=Indeterminate

The referenced ServiceRequest does not contain ehealth-reuseCriteria specifying a max age of the height Observation. This max age, although stated for reuse, is used for the height Observation use in an BMI calculation also.

code=N, system=http://hl7.org/fhir/v3/ObservationInterpretation, display=Normal

The height Observation has not exceeded the max age specified in ehealth-reuseCriteria in the referenced ServiceRequest.

code=L, system=http://hl7.org/fhir/v3/ObservationInterpretation, display=Low

The height Observation has exceeded the max age specified in ehealth-reuseCriteria in the referenced ServiceRequest.

Library for Calculating Body Mass Index

The Library for calculating the Body Mass Index (BMI) is intended for providing a value to be submitted as a BMI Observation as specified in an ActivityDefinition and possibly associated with reference ranges. Given an Observation of height and an Observation of weight as parameters, the evaluate operation returns a Quantity which can be set as the BMI Observation’s Observation.valueQuantity.

The Library has:

  • .type set to logic-library

  • .parameter[0]

    • .use set to out for output

    • .type set to Quantity

  • .parameter[1]

    • .use set to in for input

    • .type set to Observation

  • .parameter[2]

    • .use set to in for input

    • .type set to ServiceRequest

The library expects two observations as input:

It is not necessary to provide the ServiceRequest as input. The Libary$evaluate operation will retrieve it by resolving the basedOn reference in the weight Observation.

Output is a Quantity containing the calculated BMI:

Library for Calculating CAT-score (COPD Assessment Test)

The Library for calculating the CAT-score is intended for providing a value to be submitted as a CAT-score Observation as specified in an ActivityDefinition and possibly associated with reference ranges. Given a QuestionnaireResponse for the COPD Assessment test as input parameter, the evaluate operation returns a Quantity which can be set as the CAT-score Observation’s Observation.valueQuantity.

The Library has:

  • .type set to logic-library

  • .parameter[0]

    • .use set to out for output

    • .type set to Quantity

  • .parameter[1]

    • .use set to in for input

    • .type set to QuestionnaireResponse

  • .parameter[2]

    • .use set to in for input

    • .type set to Questionnaire

The library expects a QuestionnaireRespone as input:

  • The QuestionnaireResponse must reference a Questionnaire containing all 8 questions that are part of the COPD Assessment Test

  • The QuestionnaireResponse must contain valid answers for all 8 questions that are part of the COPD Assessment Test

It is not necessary to provide the Questionnaire as input. The Libary$evaluate operation will retrieve it by resolving the questionnairereference in the QuestionnaireResponse.

Output is a Quantity containing the calculated CAT score:

Library for Calculating MRC-score (Medical Research Council score)

The Library for calculating the MRC-score is intended for providing a value to be submitted as a MRC-score Observation as specified in an ActivityDefinition and possibly associated with reference ranges. Given a QuestionnaireResponse for the MRC test as input parameter, the evaluate operation returns a Quantity which can be set as the MRC-score Observation’s Observation.valueQuantity.

The Library has:

  • .type set to logic-library

  • .parameter[0]

    • .use set to out for output

    • .type set to Quantity

  • .parameter[1]

    • .use set to in for input

    • .type set to QuestionnaireResponse

  • .parameter[2]

    • .use set to in for input

    • .type set to Questionnaire

The library expects a QuestionnaireRespone as input:

  • The QuestionnaireResponse must reference a Questionnaire containing the question that constitutes the MRC Test

  • The QuestionnaireResponse must contain a valid answer for the question that constitutes the MRC Test

It is not necessary to provide the Questionnaire as input. The Libary$evaluate operation will retrieve it by resolving the questionnairereference in the QuestionnaireResponse.

Output is a Quantity containing the calculated MRC-score: