Get FHIR Record
Description: Retrieves patient medical records, including vaccination history and other health data, for a specific patient. Particularly useful for checking immunization records, COVID-19 vaccinations, and other medical events. Requires patient ID and FHIR resourceType.
get_fhir_record Parameters
Required Parameters
- resource (string)
- Available FHIR resource types:
- "MedicationRequest"
- "MedicationDispense"
- "MedicationStatement"
- "Procedure"
- "CarePlan"
- "DiagnosticReport"
- "Lab"
- "Vitals"
- "Condition"
- "AllergyIntolerance"
- "Encounter"
- "Immunization"
- Description: Name of FHIR resource to retrieve.
- Available FHIR resource types:
Optional Parameters
-
active_only (boolean)
- Default:
false - Description: Whether to filter for active records only. Default is False.
- Default:
-
count (integer or null)
- Default:
300 - Description: Number of raw FHIR records to fetch BEFORE filtering. MINIMUM: 300 (automatically enforced). This parameter controls the raw fetch size to ensure sufficient data exists for filtering. Always use high values (≥300) for initial queries. Increase to 1000-5000 for comprehensive historical search scenarios.
- Default:
-
end_date (date string or null)
-
- Default:
null - Format:
YYYY-MM-DD - Description: The latest date (inclusive) to include FHIR records. If not provided, no upper date bound is applied.
- Default:
-
maximum_tokens (integer or null)
- Default:
null - Description: The maximum number of tokens to return in the result. Default is None.
- Default:
-
start_date (date string or null)
-
- Default:
null - Format:
YYYY-MM-DD - Description: The earliest date (inclusive) to include FHIR records. If not provided, no lower date bound is applied.
- Default:
-
text_filters (array of strings or null)
- Default:
null - Description: List of text strings to prioritize matching records in results. Matching records appear first, improving answer accuracy. Use when user asks about specific medical data items, especially:
- Specific test results ('A1c', 'cholesterol', 'glucose')
- Particular vital signs ('blood pressure', 'temperature', 'weight')
- Time-series queries ('last 5 readings', 'most recent values')
- Counting/trending ('how many times', 'changes over time')
Generate multiple variations for each term:
- Full name + abbreviations (e.g., 'hemoglobin' + 'hgb')
- Medical synonyms (e.g., 'glycohemoglobin' for A1c)
- Related components (e.g., for 'blood pressure' add 'systolic', 'diastolic')
- Common alternatives (e.g., for 'cholesterol' add 'ldl', 'hdl', 'triglyceride')
Examples:
- 'last 5 A1c results' → ['hemoglobin', 'hgb', 'a1c', 'hba1c', 'glycohemoglobin']
- 'cholesterol levels' → ['cholesterol', 'ldl', 'hdl', 'triglyceride', 'lipid']
- 'recent blood pressure' → ['blood pressure', 'bp', 'systolic', 'diastolic']
Don't use for:
- General summaries ('tell me about my health', 'what's in my chart')
- Broad categories ('all lab results', 'all medications')
- Vague queries ('anything abnormal', 'what should I know')
- Default:
When to Use get_fhir_record
This tool retrieves patient medical records, including vaccination history and other health data, for a specific patient. It's particularly useful for:
- Checking immunization records, COVID-19 vaccinations, and other medical events
- Retrieving specific FHIR resource types
- Getting detailed medical data with filtering capabilities
TIP: For better results when the user mentions medical terms (labs, conditions, medications, vitals, etc.), use the 'text_filters' parameter with variations of those terms. This helps prioritize relevant data.
Updated 3 days ago
