Skip to content
This repository has been archived by the owner on Jan 5, 2021. It is now read-only.

Source Code for Calculating QPP/MIPS Quality Measures from Medicare Claims Data

License

Notifications You must be signed in to change notification settings

CMSgov/qpp-claims-to-quality-public

Repository files navigation

MIPS - Claims-to-Quality Analyzer - Open-Source

This is the open-source version of the C2Q Analyzer repository.

To learn more about the architecture of the project, please have a look here. To learn how we keep them in sync, follow the instructions here.

About

claims-to-quality-analyzer is intended to calculate claims-based quality measures for providers. The analyzer works as follows:

- Read provider data and filter on quality codes
- For each of 74 claims-based quality measures and for each provider, filter to relevant claims.
- Calculate measures and aggregate for each provider.
- Resulting measurements are submitted to the QPP Submissions API.

Project Structure

  • audit: contains scripts relating to verification of measure processing.
  • bin: contains scripts employed during setup of the C2Q system.
  • claims_to_quality: contains the bulk of the source code. Broken down into the following three components:
    • analyzer: calculating the quality measures
    • lib: shared resources
  • data: contains data sources used for analysis and exploration.
  • tests: contains automated unit tests.

Instructions

We use Docker for easier collaboration and deployment. If you don't already have Docker installed, install it here.

Local Development

Tests and Linter

Several entries exist in the makefile for running tests. Use make local-test to run the full test suite and linters locally.

You can simply start the app following the instructions below.

Analyzer

make local-analyzer

Note

To detach from a running container which you are attached to: ctrl + p, ctrl + q - hold ctrl in between.

Poirot Secrets Testing

A secrets pattern file hubzone-poirot-patterns.txt is included with the app to assist with running Poirot to scan commit history for secrets.

Install

It is important to run Poirot locally and not through docker-compose, to make sure that you are looking at all the files in the repo.

Run

It is recommended to run this on the current branch only:

  poirot --patterns c2q-poirot-patterns.txt --revlist="develop^..HEAD"

Poirot will return an error status if it finds any secrets in the commit history between HEAD and develop. You can correct these by: removing the secrets and squashing commits or by using something like BFG.

Note

Note that Poirot is hardcoded to run in case-insensitive mode and uses two different regex engines (git log --grep and a 3rd-party Python regex library https://pypi.python.org/pypi/regex/ ). Refer to Lines 121 and 195 in <python_path>/site-packages/poirot/poirot.py. The result is that the 'ssn' matcher will flag on: 'ssn', 'SSN', or 'ssN', etc., which also 'className', producing false positive errors in the full rev history. Initially we included the (?c) flag in the SSN matchers: .*(ssn)(?c).*[:=]\s*[0-9-]{9,11} however this is not compatible with all regex engines and causes an error in some cases. During the --revlist="all" full history Poirot runs, this pattern failed silently with the git --grep engine and therefore did not actually run. During the --staged Poirot runs, this pattern fails with a stack trace with the pypi/regex engine. The (?c) pattern has been removed entirely and so the ssn patterns can still flag on false positives like 'className'.

Contributing

This repository is open for contributions, although it is not its primary intent. Please read CONTRIBUTING to learn more on how to contribute.

We strive for a welcoming and inclusive environment for the Claims-to-Quality project.

Please follow this guidelines in all interactions:

  1. Be Respectful: use welcoming and inclusive language.
  2. Assume best intentions: seek to understand other's opinions.

Security Issues

Please do not submit an issue on GitHub for a security vulnerability. Please contact the development team through the Certify Help Desk at [email protected].

Be sure to include all the pertinent information.

The agency reserves the right to change this policy at any time.

About

Source Code for Calculating QPP/MIPS Quality Measures from Medicare Claims Data

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages