Meaningful Matches
Meaningful Matches matches patients to Primary Care Providers (PCPs) buy creating ideal member-PCP matches based on provider performance measures and member demographics / health needs.
If you think it's a challenge to create an algorithm to match a doctor and a patient, it's a whole different ballgame to ensure that the patient goes to that doctor. This project has a little of service design and a lot of talking to craigslist strangers on the other side of the country.
Default Matching & Compliance Research Project
Timeframe: August 2018- Jan 2019 (~4 months)
My role: I led this research project alongside 1 PM and a team of 3 data scientists tackling the problems from a quantitative perspective.
My role: I led this research project alongside 1 PM and a team of 3 data scientists tackling the problems from a quantitative perspective.
Research Goals
Understand how Medicaid Beneficiaries and their Caregivers find, choose and change their primary care provider. Following this research, we will be able to ideate ways to improve our algorithms and member touch-points in order to make our matches more meaningful, such that members use them to get care. Participants
Members of Medicaid Managed Care plans and their Caregivers in Ohio that were default assigned and went to a different healthcare provider. |
Output
|
Research Methods
- Screener Survey via Craigslist, Reddit, Facebook, Nunas, etc
- Remote, open-ended interviews conducted with Validately.
Insights
Impact
- Worked with Data Scientists to change algorithms to measure compliance at the practice level rather than the provider level
- Wrote language to show patients that are matched, to nudge them to go to the assigned physicians