Observations

APHL Comments on ISA 2022

APHL suggests that USCDI contain some clarification on how this class is expected to be used compared to these other classes, that all contain observations. Clinical Tests Clinical Test = Observation Code Clinical Test Result/Report = Observation Value   Diagnostic Imaging    Diagnostic Imaging Test = Observation Code Diagnostic Imaging Report = Observation Value   Laboratory   Tests = Observation Code Values/Results = Observation Value Laboratory Test Performed Date = Observation Timing   Vital Signs This class is slightly different, as it created a single element for the combination of Observation Code Observation Value; so for example: Systolic blood pressure  references as vocabulary Logical Observation Identifiers Names and Codes (LOINC®) version 2.67 to identify its respective Observation Code and The Unified Code of Units for Measure (UCUM), Revision 2.1 to define the units for the Observation Value, since in this case it will be a numeric value

Vizient comments

Vizient recommends adding to USCDI v3 Observation Value; Observation Code; and Observation Performer. These 3 elements combined could be used to detail what kinds of observations took place, what the observations resulted in, and who ended up performing those observations during the visit.

Level 2 Data Class: Observations

Level 2 Data Element: Observation Value Level 2 Data Element: Observation Code Level 2 Data Element: Observation Timing Level 2 Data Element: Observation Subject Level 2 Data Element: Observation Performer IMO supports the inclusion of the proposed Level 2 Data Class: Observations and the specified data elements described above. CDA R2 C-CDA Templates for Clinical Notes R2.1 incorporates observations in mature and commonly implemented document types including Procedures and Progress notes. Numerous VSAC value sets specified in SNOMED CT, LOINC, ICD 10, HCPCS and CPT represent the data elements in this proposed Level 2 data class.  ONC certified HIT can support these data elements.

CMS-CCSQ Support for Observation Data Class for USCDI v3

CMS-CCSQ recommends adding to the USCDI v3 an observation data class with associated codes, values, and the performer, as a data capture structure that allows for exchange of standard clinical assessments and observations that routinely occur and are captured in discrete, structured fields.

1. Data Element: Observation codes; CMS recommends adding observation codes related to clinician-administered assessments/observations to the USCDI. These include observations for screenings (i.e., depression screenings), clinical assessments, and risk assessments (i.e., pain intensity assessment, fall risk assessment) via standard assessment instruments.
2. Data Element: Observation values; defined as the discrete values (results) of the observations

Rationale: In addition to lab and vital signs, many clinical observations assessed for patients shape quality patient care. Clinical observations are an essential structure for recording many kinds of health information, with results that inform clinical care and condition management decisions and are used extensively throughout CMS quality measurement. USCDI v2 added a data class for Clinical Tests, which includes some, but not all, of these observations/results. CMS specifically uses the following types of clinical assessments/observations in measurement and requests the observation code and value structure be added to the USCDI to support exchange: clinically-administered assessments, clinical screenings. These types of observations are administered and captured in discrete fields with specific associated codes and values, are exchanged via the FHIR Observation profile, and are specified by category codes in FHIR to distinguish between different types of observations. Maturity:

• Current standards:
  • Mature and standardized terminology exists via LOINC and SNOMED to represent clinical observations and assessments.
  • The Observation FHIR resource, included in the QI Core IG, is used to exchange this information.
    • Category is used, within the profile, to specify the type of observation.
• Current uses, exchange, and use cases: These data are already extensively captured in EHRs by providers in discrete fields, and routinely exchanged for quality measurement and care coordination. For example, LOINC codes are used to define depression screening assessment tools used, and the results (or values) can take the form of quantitative results, ordinal scale values, or categorial values.

 

3. Data Element: Performer; CMS recommends inclusion of patient-reported data, or structured data that comes directly from the patient related to the status of a patient’s health condition, in the USCDI. This data is typically captured via questionnaires and transformed into observations for storage and exchange. The observation performer data element (observation.performer in FHIR) is crucial for understanding context for observations derived directly from the patient.

Rationale: CMS is committed to an expanded use of patient generated data and recommends the inclusion of data in the USCDI that comes directly from the patient – in this specific case, patient reported outcome survey data with data captured in a structured way and can be exchanged as observations, with the performer specified as the patient. This is important not only to quality measurement but to advancing patient-centered clinical care, increasing patient access to data, and improving patient engagement. This data element has been identified by the USCDI Task Force as a priority area, and the standards around this concept have continued to mature and have continued to be tested in FHIR Connectathons (most recently in September 2021). Maturity:

• Current standards:
  • Argonaut Questionnaire Implementation Guide
  • FHIR Structured Data Capture (SDC) Implementation Guide
  • LOINC
• Current uses, exchange, and use cases: Testing of FHIR resources continues in Connectathons, and CMS continues to capture patient reported data in quality measurement programs, as this is a priority area of focus across the healthcare ecosystem. This includes survey data captured from an instrument, such as PROMIS or HOOS and KOOS about mental or physical status. Use of PRO data is expanding nationally. Rapid technology advancements are simplifying mobile data collection and increasing integration of PRO data collection into clinical workflows and electronic medical records.

NCPDP Comment

NCPDP recommends the NCPDP SCRIPT Standard Version 2017071 be added to the list of standards.

Store the observation code with the results in the EHRs

The observation code is often stored in one of many mapping tables that researchers cannot find.  Suggest storing it as part of the result as defined in FHIR, CDA and HL7 V2.  Apple health does so, making it easy to find and to verify its correctness. 

Include Observation interpretation

An especially useful field in the Observation resource.  Almost universally included in laboratory results and other measurements. Especially useful in research studies for equivalencing tests about the same analyte across labs AND across numeric and ordinal valued tests. From FHIR, see: https://www.hl7.org/implement/standards/fhir/observation-definitions.html#Observation.interpretation