Healthcare and life sciences organizations using AWS HealthLake can now query clinical FHIR data live in Databricks through a novel OpenSharing architecture. This eliminates costly export pipelines, extends governance with Unity Catalog, and introduces natural language exploration via Genie One.

  • Live querying of HealthLake FHIR data directly from Databricks with no exports
  • Unity Catalog extends governance with attribute-based and role-based access controls
  • Natural language exploration enabled through Genie One for broader user accessibility

Infrastructure signal

The integration introduces a custom OpenSharing server implemented as a serverless Databricks App that mediates access between AWS HealthLake and Databricks. This server leverages Lake Formation generated temporary credentials and pre-signed S3 URLs for direct reads of Iceberg data files stored in HealthLake’s S3 buckets. This eliminates the previous necessity for creating and maintaining export pipelines or duplicated datasets, thereby reducing cloud storage and data processing costs.

From an infrastructure perspective, this architecture relies on AWS OIDC federation for identity trust between Databricks and AWS, ensuring secure, auditable access to sensitive clinical data. Since data files are read directly from S3 by Databricks compute nodes, the solution avoids bottlenecks or additional proxies, enhancing reliability and performance for analytics workloads.

Developer impact

Developers and data engineers benefit by accessing HealthLake clinical datasets in Databricks without building complex ETL pipelines or manual synchronization. The direct live access model enables more agile experimentation, reduces operational overhead, and accelerates time-to-insight with integrated analytics, machine learning, and AI features on normalized FHIR data.

The use of Unity Catalog to mount HealthLake shares as governed tables means that developers can apply fine-grained access control models (RBAC, FGAC, ABAC) uniformly across clinical data. This simplifies compliance with regulatory and internal governance policies. Additionally, developers can leverage Databricks notebooks, dashboards, and workflows directly on live HealthLake data without extra copies or delayed updates.

What teams should watch

Cloud architects and data platform teams should evaluate how this integration reduces the total cost of ownership by eliminating redundant data storage and management overhead in multi-tool clinical data workflows. They should also monitor governance implementation within Unity Catalog to ensure policies meet healthcare data compliance requirements.

Analytics and AI teams should focus on exploiting the unified data model and live connectivity for developing advanced healthcare insights, predictive models, and natural language interfaces powered by Genie One. Early adoption can offer competitive advantages in deploying conversational analytics over standardized clinical data formats.

Security and compliance teams need to oversee the correct setup of OIDC trust relationships and audit logs in Unity Catalog to validate that access to sensitive HealthLake data adheres strictly to organizational and industry regulations.

Source assisted: This briefing began from a discovered source item from Databricks Blog. Open the original source.
How SignalDesk reports: feeds and outside sources are used for discovery. Public briefings are edited to add context, buyer relevance and attribution before they are published. Read the standards

Related briefings