This project uses QHP data to create FHIR resources based on the DaVinci US Drug Formulary Implementation Guide Draft STU2. Currently, the following are created:

• PayerInsurancePlan: The PayerInsurancePlan profile of the FHIR R4 InsurancePlan resource defines the top level package of health insurance coverage benefits that a payer offers.

• Formulary: The Formulary profile of the FHIR R4 InsurancePlan describes a prescription drug insurance offering comprised of drug benefits including a definition of drug tiers and their associated cost-sharing models and additional information about the plan, such as networks, a coverage area, contact information, etc.

• FormularyItem: The FormularyItem profile of the FHIR R4 Basic describes a drug's relationship to a drug plan, including drug tier, prior authorization requirements, and more. The set of FormuaryItem resrouces associated with a particular drug plan represents the drug plans formulary.

• FormularyDrug: The FormularyDrug profile of the FHIR R4 MedicationKnowledge resource provides information about a prescribable drug which may be part of a formulary including its RxNorm code, synonyms, and optionally drug classification and alternatives.

The application has four main functions:

• Generate formulary data samples: this is handled by the generate.rb script, which generates conformant data samples for each profile listed above.

• Populate a FHIR server with the data: this is handled by the upload.rb script.

• Generate the ndjson files: this is handled by the convertNDJSON.rb script, which generates the ndjson files for each profile to be used for bulk export requests of all drug formulary data (e.g. [base]/InsurancePlan/$export).

• Generate ndjon files grouped by InsurancePlan: this is handled by the plan_specific_ndjson.rb script, which generates formularies ndjson files classified by individual InsurancePlan. This is useful for plan specific bulk export requests (e.g. [base]/InsurancePlan/[id]/$export).

Note: You can switch to STU1.1.0 branch of this repo to generate STU1 Formulary sample data.

This is a Ruby application and currently, it can only be run on development mode.

You should have the following programs installed to run this application:

• Ruby
• Bundler
• YARD

1. Clone this repository

git clone git@github.com:HL7-DaVinci/pdex-formulary-sample-data.git

2. Change directory into the project directory and pull in the dependencies

cd pdex-formulary-sample-data
bundle install

A couple Rake tasks were created to easily run the scripts at the command line. You can run rake -T to see the list of tasks available.

1. Edit the properties file (config.yml) as needed. By default,the app will generate a max of 1000 drugs per formulary.
2. Run the generate.rb script

rake run_script:generate_data

Or

bundle exec ruby scripts/generate.rb

The generated files are placed in the output/ folder.

Note: the output directory is not tracked by git.

Run the upload.rb script to upload the sample data to a server.

rake run_script:upload_to_server

Or

bundle exec ruby scripts/upload.rb

By default, the FHIR server base URL to upload the data is set to http://localhost:8080/fhir. The script also upload the conformance definition resources (code systems, search parameters, structure definitions, and value sets ) to the server. The URL to download those conformance resources is also set to a default value (the download URL of the current version of the implementation guide).

Command-line arguments may be provided to specify the server base URL and definitions URL to be used:

rake run_script:upload_to_server[http://exampleserver.com,http://download-definitions-url.json.zip]

Or

bundle exec ruby scripts/upload.rb -f http://exampleserver.com -c http://download-definitions-url.json.zip

Full usage info can be printed by passing the -h (help) option.

bundle exec ruby scripts/upload.rb -h

Run the convertNDJSON.rb script to generate ndjson files for each profile using the sample data generated above.

rake run_script:generate_ndjson

Or

bundle exec ruby scripts/convertNDJSON.rb

This will read all the files in the output directory and generate the new files in the export directory.

Note: The export directory is not tracked by git.

To generate ndjson files grouping each insurance plan with their associated formulary data, run the plan_specific_ndjson.rb script:

rake run_script:plan_specific_ndjson

Or

bundle exec ruby scripts/plan_specific_ndjson.rb

This will read all the files in the output directory and generate the new files in the bulk_export directory.

Note: The bulk_export directory is not tracked by git.

The generated ndjson files can be copied to the appropriate directory/location on the server to be used as a response to bulk export requests.

The application code is documented with YARD. You can check the code documentation to quickly get familiar with the code base.

P.S. Contributors, we will be super grateful if you document your new code. This will greatly improve the documentation of this app.

Run the following command to (re)generate the documentation files:

rake yard:build

The code documentation is published to github pages and the doc directory of the master branch is used pull the latest generated YARD files.

There's a github workflow configured to automatically deploy the documentation to github page with any changes on the master branch.

Contributions are welcome. Feel free to open an issue on this repo issue tracker to report a bug, ask specific questions about this project, or propose a new feature/feature enhancement. Make sure to label your issue with the appropriate tag.

Check Github issue tracker for any outstanding tasks.

Please submit a pull request with a clear description (can be a paragraph for bigger changes) of the changes you have made. It will be great if you include RSpec tests for your new code. We can always use more test coverage.