Skip to main content
CLIF Logo

Common Longitudinal ICU data Format (CLIF)

Advancing Critical Care Through Federated Research

CLIF streamlines longitudinal ICU data into a harmonized, federated framework — enabling rapid, privacy‑preserving multicenter research

Explore the Four Dimensions of CLIF

📄

High Impact Scientific Papers

Our consortium has published significant research in top-tier journals, demonstrating how CLIF enables rapid, multi-center studies that were previously impossible. From COVID-19 outcomes to sepsis prediction, CLIF supports discoveries that improve patient outcomes.

Read Our Latest Publication
🧾

Open Source Data Dictionary and ETL Tools

CLIF 2.0.0 represents years of refinement, incorporating feedback from 10+ billion data points. Our open-source data dictionary and ETL tools make it easy for any institution to transform their ICU data into the CLIF format, joining our federated research network.

View CLIF-2.0.0 Data Dictionary
🧬

Large Distributed Cohort in the Consortium

With 680,158 unique ICU patients and 862,400 encounters across 10 active sites, CLIF provides the statistical power needed for robust research. Our interactive dashboard lets you explore cohort characteristics and demographics in real-time, ensuring your studies have the data they need.

Explore Cohort Dashboard
🛠️

Open Source Products Built on Top of CLIF

From machine learning pipelines to quality improvement dashboards, the CLIF ecosystem includes open-source tools that accelerate research. Build on our foundation with confidence, knowing you're using proven infrastructure designed for critical care analytics.

See What You Can Build
0

Participating Institutions

0

Unique ICU Patients

0

ICU Encounters

0

Hospitals

Building a Foundation for Critical Care Research

We assembled a geographically diverse group of U.S.-based clinician-scientists and data scientists with deep experience in EHR-based clinical outcomes research and AI applications.

July 2023

Consortium Assembly Begins

Geographically diverse group of U.S.-based clinician-scientists and data scientists convene virtually

2023-2024

Challenge Identification & Solution Development

Identified practical challenges of using EHR data for critical illness research across institutions

2024
Institutions
18+

Standardized Framework Established

Developed standardized operating procedures, shared terminologies, and robust quality control methods

Patients
680K+
Data Points
862K encounters
2025

Publication & Implementation

Framework published in Intensive Care Medicine, enabling critical care research across all hospital settings

Our Guiding Principles

Efficient

Clinically meaningful data structures

Consistent

Harmonizable data elements

Scalable

Flexible for future innovations

Federated

Privacy-preserving collaborative research

Open Source

FAIR data principles & NIH compliance

Our Overarching Goal

To provide a standardized framework for representing critical illness data across all hospital care settings — from ICUs to emergency departments and hospital wards — enabling researchers to accomplish essential critical care informatics tasks wherever and whenever critical illness begins.

Advancing Federated Critical Care Research

CLIF represents a significant advancement in how we approach multi-center critical care research. By standardizing data at the source while preserving patient privacy through federated analytics, we're enabling important discoveries that improve patient care.

The Challenge We Solve

Fragmented Systems

Incompatible data formats and systems across institutions prevent effective collaboration

Time Barriers

Months needed for data use agreements, then harmonizing data at central sites delays critical research

Privacy Concerns

Centralized data repositories raise security and compliance challenges

The CLIF Solution

Rapid Analysis

Test hypotheses across millions of encounters in days, not years

10B+ data points analyzed

Privacy-First

Federated architecture ensures patient data security

Data never leaves your site

Standardized Format

CLIF 2.0.0 - Our most comprehensive data model

26 harmonized tables

Key Features

Accelerate Discovery

Go from hypothesis to multi-center analysis in days, not years. Our standardized format eliminates months of data wrangling.

Preserve Privacy

Federated architecture means patient data never leaves your institution. Achieve collaboration without compromising security.

Ensure Reproducibility

Standardized variables and harmonized vocabularies through mCIDE guarantee your research can be validated and extended.

Open Source Innovation

CLIF is proudly open source software, licensed under the Apache License 2.0. This ensures our tools and standards remain freely available to the global research community, fostering collaboration and supporting critical care research globally.

View Apache 2.0 License

Our Growing Collaborative Network

Leading academic medical centers across North America have joined CLIF to advance critical care through collaborative, data-driven research

Map of CLIF Consortium Sites
Sites with CLIF data already (10)
Sites planning to contribute (8)
17 institutions46 hospitals10 sites with data