Washington, D.C.– The Bipartisan Policy Center in partnership with the Allegheny County Department of Human Services in Pennsylvania has launched a project to test an emerging technology that could improve data analysis for government programs while also protecting individual privacy.
It is among the first pilot projects using the emerging technology called secure multi-party computation outside of the defense and intelligence industries. The pilot project seeks to demonstrate that connecting data from across multiple programs is possible while preserving the privacy of individuals, with potential applications for health, human services, and other programs.
“We are hoping the pilot will demonstrate privacy improvements while allowing for the use of valuable data, and lead to better information for policymakers in decisions that affect the lives of residents in the community,” Nick Hart, director of BPC’s Evidence-Based Policymaking Initiative, said. “The deployment of this technology should provide a vastly stronger capability for privacy-preserving data analytics that may have applications across federal, state, and local governments.”
The project will use county data on services to the homeless, behavioral health services, causes and incidence of mortality, family interventions, and incarceration. Data used in the pilot will not include personally identifiable information. Final results are expected in January 2019.
“Allegheny County’s Department of Human Services is very pleased to partner with BPC in this groundbreaking work that brings a new method of privacy-preserving technology to human services while exploring innovative uses of data to forward our effectiveness and efficiency,” Marc Cherna, director of the Allegheny County Department of Human Services, said.
The U.S. Commission on Evidence-Based Policymaking unanimously recommended that applications of multi-party computation be further explored. The pilot’s success would demonstrate that this sophisticated technology can be used in human services systems and that the approach has the potential to unlock analyses that may not otherwise be possible today for improving the administration of government services and programs.
The project is funded by Schmidt Futures.