Challenge: Geography-specific, vertically-integrated collection of health and social determinant data

From Demand-Driven Open Data for HHS
Jump to: navigation, search

See Valley Vision use case

Dear DDOD community, feel free to edit, correct or improve any part of this article. Just click "Sign in with Google" and then "Edit"

The problem

Data from our healthcare system is quite fragmented. We have Medicare and provider quality at the Federal level, Medicaid and ACF services at the state level, any number of additional services at the city and neighborhood level, clinical EHR data at local healthcare providers, and additional social determinant data held by nonprofits.

What improvements in population health or acceleration of medical discoveries are possible with...

Federal + State + Local data?

The approach

The hypothesis is that having a more complete portfolio of datasets would enable a great deal of insights about population health and wellbeing. That in turns could be used for informing policy, designing needed services and identifying interventions. Trying to accomplish this across the country is a daunting task. We propose picking a narrow geography, such as a city or even a neighborhood, and focusing efforts for obtaining data for it. The goal is to prove out this full vertical approach as a model that can be applied to other location.

The ask

We are asking for a thought experiment... What could be accomplished with a complete portfolio of health and social determinant data on a city or neighborhood? How can we use such data to better assess the health and well-being of a population? How can it be leveraged to design better services or promote medical discoveries?

In order to get started on an implementation, we need:

  1. A defined geography
  2. A champion or stakeholder responsible for a geography
  3. One or more high value use case for integrating federal, state and local data for that geography

We recognize that the biggest challenges are typically not technical. We will have to overcome issues around:

  1. Privacy and risk of reidentification
  2. Legal and regulatory
  3. Budgetary and competing priorities

The general approach is to:

  • Identify data sources
  • Lay out a roadmap for implementation
  • Follow the roadmap, approach the data producers and owners, adjust as the work progresses
  • Document the progress and challenges encountered along the way

Ultimately, the learnings documented in the pursuit of this goal become part of the public body of knowledge. That knowledge can then be used to replicate similar projects in other geographic locations.

Next step

Go to this Google document and add your entry. We're looking for (1) Geography, (2) Relevant owner / stakeholder, and (3) use case and associated value. (Be sure you're signed in, so that entries can be attributed to you.) Feel free to comment on existing ones as well.

Click here to edit list

Implementation models

Once we have use cases, we can start to work on obtaining and linking the datasets for the highest value use cases. We can look to the best practices around:

  • VRDCs (Virtual Research Data Centers),
  • NCHS's Record Linking program, and
  • HIEs (Health Information Exchanges)

Potential participants

  • States, cities: Illinois, New York, California, City of Chicago
  • Non-profits and social service organizations: Valley Vision, Purple Binder