bayesimpact/bayeshack-hhs-marketplace

Bayes Hack 2016

Department of Health and Human Services prompt #1

How can data connect individuals with the health providers they need?

Prompt

Since the launch of health insurance marketplaces as part of the Patient Protection and Affodable Care Act (PPACA), millions of Americans have obtained new health care coverage. There have, however, been widespread consumer complaints; in particular, people tend to choose the wrong plan because relevant information isn't available or is inaccurate. Often, patients don't discover until after a purchase that their physician isn't in-network, or that an in-network specialist they need isn't taking patients.

In November 2015, the Centers for Medicare & Medicaid Services (CMS) enacted a new regulatory requirement for health insurers who list plans on insurace marketplaces. They must now publish a machine-readable version of their provider network directory, publish it to a specified JSON standard, and update it at least monthly. Finally, this data is becoming accessible.

But computer- and engineer-accessibility doesn't make it particularly accessible to the general market of health care consumers. The new challenge, then, is to transform this vast directory of provider data into insights that can guide individuals to the health care they're paying for, that they deserve, and that they often badly need.

David Portnoy of HHS has created this challenge page for Bayes Hack. It has all you need to get started.

In this Repo

• data/download.sh - A shell script to dowload CMS data:

  for Mac only:

  Run source data/download.sh year to download and unzip all datasets for year=year.

  For example run source data/download.sh 2016 to download 2016 data, if you want to run the sample analysis notebook. To download the entire dataset, run source data/download.sh 2014 2015 2016.

  for non-Mac:

  You can download the files manually from here.

  Also, look at this github repo, for an awesome Makefile that creates a SQLite database from the raw data.

• analysis/ - iPyton notebook files (which you can view right here on GitHub) loading the data and exploring a few things. Good to understand the datasets and get ideas for your project. If you want to run this notebook, run pip install -r requirements.txt inside a virtualenv first.

##Available data formats: Prior to 2016, the marketplace datasets are available in simple CSV formats. However, since November 2015, insurance issuers are required to submit and maintain their plans in a machine-readable (MR) format. Therefore, for 2016, the health insurance marketplace PUF data is available in two formats: CSV and JSON. To learn more about these JSON file structures look at this github repo. This spreadsheet contains URL to access all the JSON formated files for each insurance issuer.

##Other Resources

• The CMS Health Insurance Marketplace Public Use Files, this prompt's core dataset. The 2016 part of this dataset is available in /data folder.
• HHS Health Insurance Marketplace challenge on Kaggle.
• Directory data schemas from the HHS DDOD.
• Detailed analysis of the Provider Network Directories from David Portnoy, who contributed heavily to the development of data standards for the directory. #How can data connect individuals with the health providers they need?