Skip to content

CLIF Tables Overview

CLIFpy implements all 9 tables defined in the CLIF 2.0.0 specification. Each table represents a specific aspect of ICU patient data, with standardized columns and validation rules. Detailed CLIF Data Dictionary is available here

Data Standards

Each table follows CLIF standards for:

  • 🏷️ Standardized Categories

    Consistent values across institutions, validated against permissible value lists, and mapped from institution-specific values.

  • 🏥 Source Preservation

    Original EHR data elements maintained alongside standardized mappings for institutional transparency.

  • 🕒 Timezone Handling

    All datetime columns are timezone-aware with consistent timezone across all tables and automatic conversion during loading.

  • 🔑 Composite Keys

    Unique record identification with duplicate detection and data integrity validation.

Available Tables

graph TD
    Patient[Patient] --> |patient_id| Hospitalization[Hospitalization]
    Hospitalization --> |hospitalization_id| ADT[ADT]
    Hospitalization --> |hospitalization_id| Labs[Labs]
    Hospitalization --> |hospitalization_id| Vitals[Vitals]
    Hospitalization --> |hospitalization_id| Meds[Medications]
    Hospitalization --> |hospitalization_id| Assess[Assessments]
    Hospitalization --> |hospitalization_id| Resp[Respiratory Support]
    Hospitalization --> |hospitalization_id| Pos[Position]

Patient

Core demographic information including birth date, sex, race, ethnicity, and language. This is the primary table that links patient_id field to the hospitalization_id field in the Hospitalization table. For detailed API documentation, see Patient API

ADT

Admission, Discharge, and Transfer events tracking patient movement through different hospital locations (ICU, ward, ED, etc.).

Hospitalization

Hospital admission and discharge information, including admission source, discharge disposition, and length of stay.

Labs

Laboratory test results with standardized categories (chemistry, hematology, etc.) and reference ranges.

Vitals

Vital signs measurements including temperature, heart rate, blood pressure, respiratory rate, and oxygen saturation.

Position

Patient positioning data, particularly important for prone positioning in ARDS management.

Respiratory Support

Ventilation and oxygen therapy data, including device types, settings, and observed values.

Medications

Continuous medication infusions with standardized drug categories and dosing information.

Patient Assessments

Clinical assessment scores including GCS, RASS, CAM-ICU, pain scores, and other standardized assessments.

Common Table Features

All tables inherit from BaseTable and share these features:

Data Loading

table = TableClass.from_file(
    data_directory='/path/to/data',
    filetype='parquet',
    timezone='US/Central'
)

Validation

table.validate()
if table.isvalid():
    print("Validation passed")

Summary Statistics

summary = table.get_summary()
print(f"Rows: {summary['num_rows']}")
print(f"Memory: {summary['memory_usage_mb']} MB")

Choosing Tables for Your Analysis

For Patient Cohort Building

  • Start with Patient for demographics
  • Add Hospitalization for admission criteria
  • Include ADT for ICU/location-specific cohorts

For Clinical Outcomes

  • Use Labs for laboratory markers
  • Add Vitals for physiological parameters
  • Include PatientAssessments for severity scores

For Treatment Analysis

  • Use RespiratorySupport for ventilation data
  • Add MedicationAdminContinuous for drug therapy
  • Include Position for positioning interventions

Next Steps