Analytics on FHIR®
2-day online conference
.jpg)
Sessions
Introducing FHIR Analytics with SQL on FHIR
A high-level introduction to the FHIR Analytics Collaborative and the SQL on FHIR specification. Arjun will discuss the state of the FHIR Analytics ecosystem, why we created SQL on FHIR v2, how it can unlock key barriers to today's and next-generation use cases, and where the FHIR Analytics Collaborative is headed next.
Sanyal
SQL on FHIR v2.1: A Technical Tour & What's New for 2025
Nikolai, a lead author and implementer of SQL on FHIR, will give a deep dive into the technical details of the specification. This will include a close look at the heart of the spec, the ViewDefinition and present the enhancements for 2025: the FHIR APIs and Library-based Queries.
Validation of SQL on FHIR via Clinical Study Replication & Introduction to Pathling
John will present findings from the recent SQL on FHIR paper published in npj Digital Medicine.
The paper documents the development and validation of a standard method for transforming FHIR data into tabular structures. John will examine the design principles behind the SQL on FHIR ViewDefinition, including the deliberate constraints that balance expressivity with implementation feasibility.
John will demonstrate how ViewDefintions were used to replicate a clinical study on racial disparities in oxygen therapy (based on the MIMIC-IV dataset) across multiple technology platforms, illustrating the portability of the approach.
He will also provide an update on recent developments in Pathling, including enhancements to its view runner implementation and integration with the broader SQL on FHIR ecosystem.
Grimes
Helios Software: High-performance Clinical Analytics with SQL on FHIR using Rust
Helios Software is an open source Rust implementation of SQL on FHIR that includes a simple CLI for batch transformations, a HTTP server ideal for microservices, and Python bindings for using SQL on FHIR directly in your data science and analytics projects.
In this session, Steve will demonstrate a real-world laboratory analytics challenge across different workflows, illustrating concrete patterns for integrating SQL on FHIR into pipelines for batch, microservice, and data science workloads.
Munini
FlatQuack: FHIR Resources to SQL Tables with DuckDB
FlatQuack is an open source, SQL-on-FHIR View Definition to SQL compiler that targets DuckDB, a lightweight, high performance database engine.
DuckDB can scale from running as a single executable that queries a directory of FHIR resources on a local machine to querying across billions of resources in the cloud. By compiling View Definitions to plain SQL, FlatQuack can be easily integrated into existing data pipelines that use off-the-shelf orchestration tools like DBT and Apache Airflow.
Dan will show how FlatQuack can deliver low-cost, high-performance FHIR Analytics at scale.
Gottlieb
Semantic Intelligence with Kodjin: Democratizing Real-time FHIR Analytics
This presentation demonstrates how semantic intelligence makes advanced FHIR analytics accessible to all users in real-time.
Kodjin Analytics enables organizations to:
• Analyze treatment pathways and patient care journeys
• Execute temporal reasoning for clinical event sequences
• Generate dynamic cohorts with complex selection criteria
• Interact with data through conversational natural language queries
• Access insights within seconds through event-driven processing
By layering semantic abstraction over real-time data streams, we enable healthcare teams to operationalize advanced FHIR analytics without technical expertise or batch processing delays. The session includes a live demonstration of pathway analysis and explores how this architecture transforms sophisticated analytics from a specialist capability into a universal healthcare tool.
Nikolaienko
How We Used SQL on FHIR to Shrink LLM Context by 92%
How we leveraged SQL on FHIR to dramatically reduce token usage in Flexpal, our smart health agent, while maintaining rich clinical context and simplifying our tooling architecture.
Kelly
Bridging Forms and Analytics: AI-Assisted FHIR Questionnaire and ViewDefinition Authoring
Olim will demonstrate how AI can bridge the gap between data capture and analytics in FHIR. This session presents a real-world workflow for authoring a coded Vital Signs Questionnaire and its corresponding ViewDefinition — all within a single, seamless pipeline. He will show how AI-assisted tooling accelerates FHIR Questionnaire design, generates realistic test data, flattens resources for analysis, and visualizes insights in real time. The demo highlights how these innovations streamline the path from structured clinical data entry to actionable analytics.
Saidov
CQL to SQL Conversion exploration
This presentation explores a practical bridge between clinical quality logic and scalable analytics—demonstrating how Clinical Quality Language (CQL) can be automatically converted into SQL to unlock the power of the existing SQL ecosystem for FHIR-based analytics.
We’ll show how using CQL as a coded technical requirements framework allows organizations to:
Define population health and quality measures in a consistent, standards-based way.
Automatically translate those definitions into executable SQL queries.
Run those queries directly within existing data warehouse or analytics platforms—without building or maintaining a separate technology stack.
By transforming CQL into SQL, we enable teams to operationalize FHIR-based measure logic in familiar analytic environments, bridging the gap between clinical standards and enterprise data systems. The session includes an overview of the CQL-to-SQL converter architecture, examples of end-to-end query generation, and a discussion of how this approach supports transparency, scalability, and alignment across clinical, analytic, and quality reporting use cases.
Vestel
Natural language queries to SQL-on-FHIR
The goal is to answer population level questions over a FHIR server using natural language queries (NLQ). Building on top of the FHIR Analytics component of Open Health Stack, we show how large language models can be utilised to transform NLQ into SQL over flat tables derived from FHIR resources using SQL-on-FHIR standard. Different components are discussed and in particular we focus on how terminology challenges can be addressed.
We explore several examples including some from the EHRSQL benchmark over the FHIR version of MIMIC-IV.
This presentation is mostly an extended version of my FHIR DevDays presentation on the same topic.
Sadjad
Using SQL on FHIR in the HL7 FHIR IG Publisher
This presentation will describe how the HL7 FHIR IG publisher uses SQL on FHIR ViewDefinitions to make it easier to include content out of the resources in pages in the implementation guide, and show same examples of the kinds of things that can be done with this technique.
Closing Roundtable
Industry experts will recap key takeaways, share future perspectives on FHIR analytics, and answer audience questions. Join the open discussion to shape the next steps in analytics for healthcare!
Sanyal
Organizers