Skip to main content

Results and Review

When a performing laboratory or imaging facility completes a diagnostic study, the results are represented in FHIR as a DiagnosticReport linked to individual Observation resources. This guide covers how results connect back to orders, how to structure them for display, and how to handle abnormal findings.

How Results Connect to Orders​

Every DiagnosticReport should reference the original order via basedOn. This link is what allows the UI to display results in the context of the order that requested them.

A single ServiceRequest (e.g. a panel order) produces one DiagnosticReport containing multiple Observation results. Each Observation records a single measurement with its value, units, and reference range.

DiagnosticReport Structure​

A completed lab result looks like the following:

{
  "resourceType": "DiagnosticReport",
  "status": "final",
  "category": [
    {
      "coding": [{
        "system": "http://terminology.hl7.org/CodeSystem/v2-0074",
        "code": "LAB",
        "display": "Laboratory"
      }]
    }
  ],
  "code": {
    "coding": [{
      "system": "http://loinc.org",
      "code": "24323-8",
      "display": "Comprehensive metabolic 2000 panel - Serum or Plasma"
    }]
  },
  "subject": { "reference": "Patient/example" },
  "basedOn": [{ "reference": "ServiceRequest/order-123" }],
  "issued": "2024-07-15T14:30:00Z",
  "result": [
    { "reference": "Observation/glucose-result" },
    { "reference": "Observation/bun-result" }
  ]
}

Category Coding​

The category field determines where a report appears in the UI. The Medplum provider app uses the v2-0074 code system to classify reports:

CodeDisplayUse
LABLaboratoryGeneral lab results; required for the patient summary sidebar
HMHematologySpecialty subtype (always pair with LAB)
CHChemistrySpecialty subtype (always pair with LAB)
MBMicrobiologySpecialty subtype (always pair with LAB)
caution

Reports that only include a specialty code without LAB will not appear in the patient summary sidebar. Always include at least one LAB category entry alongside any specialty categories.

Report Status​

The DiagnosticReport.status field tracks the lifecycle of the result:

StatusMeaning
registeredReport has been created but results are not yet available
partialSome results are available but the report is incomplete
preliminaryResults are available but not yet verified or finalized
finalReport is complete and verified
amendedReport was modified after being finalized
correctedReport was corrected due to an error in a previously finalized result
cancelledReport was cancelled (e.g. specimen was rejected)

Observation Results​

Each line item in a lab report is an Observation:

{
  "resourceType": "Observation",
  "status": "final",
  "category": [{
    "coding": [{
      "system": "http://terminology.hl7.org/CodeSystem/observation-category",
      "code": "laboratory",
      "display": "Laboratory"
    }]
  }],
  "code": {
    "coding": [{
      "system": "http://loinc.org",
      "code": "2345-7",
      "display": "Glucose [Mass/volume] in Serum or Plasma"
    }]
  },
  "subject": { "reference": "Patient/example" },
  "valueQuantity": {
    "value": 95,
    "unit": "mg/dL",
    "system": "http://unitsofmeasure.org",
    "code": "mg/dL"
  },
  "referenceRange": [{
    "low": { "value": 70, "unit": "mg/dL" },
    "high": { "value": 100, "unit": "mg/dL" }
  }],
  "interpretation": [{
    "coding": [{
      "system": "http://terminology.hl7.org/CodeSystem/v3-ObservationInterpretation",
      "code": "N",
      "display": "Normal"
    }]
  }]
}

Reference Ranges and Interpretation​

Each Observation can carry a referenceRange that defines the expected normal range for the measured value. The interpretation field provides a coded flag (e.g. N for normal, H for high, L for low, HH / LL for critically abnormal).

Reference ranges may come from the performing lab as part of the result, or they can be configured by clinical administrators using ObservationDefinition resources. See Observation Reference Ranges for details on defining your own ranges.

Critical and Abnormal Results​

When a result falls outside the expected range, the interpretation code signals the severity:

CodeMeaning
NNormal
HHigh
LLow
HHCritically high
LLCritically low
AAbnormal

For integration-based workflows, the receiving system may also create a DetectedIssue resource to flag clinical workflow concerns such as unsolicited results or unmatched patients. See the Health Gorilla Receiving Results guide for examples of how detected issues are handled.

Imaging Results​

Imaging results follow the same DiagnosticReport pattern but typically differ in structure:

  • The category uses an imaging-specific code (e.g. RAD for radiology).
  • Reports often include a narrative interpretation in DiagnosticReport.conclusion rather than discrete Observation values.
  • The original study may be attached via DiagnosticReport.presentedForm (e.g. a PDF radiology report) or DiagnosticReport.imagingStudy for DICOM references.
{
  "resourceType": "DiagnosticReport",
  "status": "final",
  "category": [{
    "coding": [{
      "system": "http://terminology.hl7.org/CodeSystem/v2-0074",
      "code": "RAD",
      "display": "Radiology"
    }]
  }],
  "code": {
    "coding": [{
      "system": "http://loinc.org",
      "code": "36643-5",
      "display": "XR Chest 2 Views"
    }]
  },
  "subject": { "reference": "Patient/example" },
  "basedOn": [{ "reference": "ServiceRequest/imaging-order-456" }],
  "conclusion": "No acute cardiopulmonary abnormality."
}

Integrations​

Results can arrive through multiple paths depending on your integration:

  • Health Gorilla synchronizes structured lab results from Quest, Labcorp, and regional labs as FHIR resources via a Medplum Bot.
  • Direct HL7v2 feeds can be mapped to FHIR using the HL7 Interface and Bots.
  • Organizations with internal labs can create DiagnosticReport and Observation resources directly via the FHIR API.

Regardless of the ingestion path, the same FHIR resource patterns apply: DiagnosticReport references the ServiceRequest via basedOn, and individual measurements are stored as Observation resources linked through result.

See Also​