Model-conformant data out of CHCS and VistA
Genshi Python M HTML JavaScript CSS
Latest commit 0baab98 Oct 17, 2016 @conorcaregraf conorcaregraf setup not up to date
Failed to load latest commit information.
Datasets/CGVistA patient 9 data from cg vista Apr 5, 2014
MUMPS added comments on IDX code Apr 18, 2014
Releases 1.3 release Jan 21, 2016
Test README Apr 4, 2014
fmaf setup not up to date Oct 18, 2016
imgs reshaped imgs Jul 23, 2016
webclients moved webclients for 1.4 Aug 20, 2016
webservice fix for strict Feb 4, 2016
.gitignore ignore pyc Nov 16, 2012
LICENSE licnese Nov 24, 2015 img Jul 23, 2016

FileMan Query Language (FMQL)

Together the Department of Defense's CHCS and VistA from the Veterans Administration represent the most widely deployed medical record system on the planet.

Both are built around a MUMPS-based data repository called FileMan - VistA or CHCS data is FileMan data. Though efficient and flexible, FileMan has no remote query language or formal schema definition mechanism - there is no "SQL for FileMan".

Instead data is transferred to and from FileMan using three kinds of mechanism: thousands of remote procedure calls (RPCs), flavors of HL7 v2 and "SQL-projection", where select data types (files) are mapped to SQL tables. But no one mechanism works for all data and each demands careful handling.

FMQL, the FileMan Query Language, changes this. Patient data, Institution data, Conceptual information ("Know-how"), System records - to FMQL, this is all just data to extract. By delivering one uniform mechanism for extracting any data, FMQL makes VistA and CHCS data access easy and let's you focus on analyzing and presenting data.

All Data and Schema too

FMQL from FileMan (c) Caregraf 2012

FMQL provides access to both FileMan data - a patient's demographics -

and the schema of that data - the type "Patient" -

Identity, Formats and Query Nuance

It addresses:

  1. Identity: every entity and entity type in FileMan gets a unique identifier - a URI
  2. Data Formats: a consistent form of JSON, the web-friendly response format, for every type of data in the system as well as HTML for human readers and RDF for web-data practitioners
  3. Query Nuance: from the precise - SELECT - to the broader - DESCRIBE - or just COUNT, covers data hierarchies and graphical layouts, paging and filtering

Add FMQL to your VistA or CHCS

Download, unzip and follow the install instructions. Any questions or problems, just contact Caregraf support.

FMQL has been tested on both fully populated and test FileMan-based systems:

  • OSEHRA VistA, VxVistA, and WorldVistA – the open source FOIA based VistAs
  • CHCS
  • GT/M on Linux, Cache on Windows and Linux

For more details, consult the FMQL Wiki.