Skip to main content Skip to secondary navigation

STARR vision is to support a number of data visualization, access and analysis tools. Today, we support a few different cohort and data access tools.


Main content start

STARR Tools (formerly known as STRIDE)

STRIDE was launched in 2008 as a fully self-service tool and is composed of a Cohort Discovery and a Chart Review tool. STRIDE has since been since re-branded to STARR Tools. The cohort building tool, Anonymous Patient Cohort Discovery Tool, returns patient counts matching a given query and is used to design studies prior to Institutional Review Board (IRB) submission. The tool is designed to answer the question “How many patients in the Clinical Data Warehouse (CDW) contain these attributes?” No individual patient data is exposed during the cohort searches. The chart review tool, accessible after IRB approval,  allows a rapid view of the retrieved patient cohort for manual review and refinement and supports exporting data for analysis using researchers’ preferred tools. In 2021, these tools were migrated from on-premise Oracle-based infrastructure to Google Cloud to leverage the power of cloud technologies. These are intuitive web based interactive tools based on Stanford in-house data model.

The most significant value add of these tools is that they are easy to use, seamlessly integrated with research compliance processes, and offer low barrier to entry.  The STARR Tools features are driven by researchers.  Research IT works closely with the tool users to add features that users request, allowing the tools to be customized for the needs of our customers.  

Learn more about accessing STARR Tools

OHDSI ATLAS Cohort Analysis Tool

ATLAS is an open-source software tool for researchers to conduct scientific analyses on standardized observational data converted to the OMOP Common Data Model. Researchers can create cohorts by defining groups of people based on exposure to a drug or diagnosis of a particular condition using healthcare claims data. ATLAS has vocabulary searching of medical concepts to identify people with specific conditions, drug exposures etc. Patient profiles can be viewed within a specific cohort allowing visualization of a particular subject's health care records. Population effect level estimation analyses allow for comparison of two different cohorts and leverages R packages. Stanford ATLAS was launched in Q1 2020 on top of STARR-OMOP, specifically, OMOP-deid-lite. This tool is particularly suitable for researchers who want to use community defined study protocols in a GUI environment.

The most significant value add of ATLAS is the researcher's ability to participate in a network study at the touch of a button or use community defined phenotypes for their own studies to accelerate research. Stanford ATLAS has the execution engine turned on, thus making is easier to run complex analysis.

Learn more about accessing Stanford ATLAS

Advanced Cohort Engine

Launched in Q2 2022, the Advanced Cohort Engine (ACE) is a web tool that combines a unique temporal query language (TQL) to search and extract patient data. It is particularly suitable for time-aware search of longitudinal patient records with a patient object datastore for rapid electronic phenotyping, cohort extraction, and exploratory data analyses. ACE operates over data in the OMOP CDM, and Stanford ACE is configured to run on top of OMOP-deid-lite. ACE TQL supports complex time-based searches and automatic query expansion using clinical knowledge graphs. 

The most significant value add of ACE is the reduction in time it takes a data scientist/analyst to develop a complex phenotype using the TQL language. For example, for the Diabetes II cohort defined by: "Male patients over 65 years old who have type II diabetes (defined by at least 2 occurrences of type II diabetes ICD9 codes or 2 elevated A1C lab results) with no history of stroke and who went on to have a stroke within 3 months after the administration of glipizide",  time to construct a SQL query by an expert is 2.5 hours and time to construct a TQL query is 5 minutes.

ACE was developed at Stanford in Dr Nigam Shah’s lab. More details are available in the publication. The R and Python APIs are available in a public GitHub repository. Learn more about how to use ACE.

Learn more about accessing ACE