Government needs to adopt a share first mentality for departmental data. Its easy exchange should be the default, not the arduous and exceptional result of lengthy administrative and legal battles to pry data from legacy computer systems and siloed departments.
Departments within a single government entity should be able to freely use data that's been scrubbed of sensitive personal information, although clear policies concerning the process for anonymizing the information, as well as how long and under what conditions it will be archived, are necessary.
Unfortunately, concerns about exchanging sensitive personal information, particularly as it relates to human services or student data governed by such federal privacy laws as the Health Insurance Portability and Protection Act and the Family Educational Rights and Privacy Act, too often override the free exchange of information. My involvement in these areas, even as a mayor or deputy mayor over the last two decades, has always been somewhere between frustrating and excruciating. Information that would actually help a client often goes unused due to obstacles more rooted in folklore than in law.
A data-sharing program in Missouri, which includes the departments of Health and Senior Services, Mental Health, and Social Services, stands as one example of how effective data sharing allows for improved delivery of care and saving of taxpayer dollars. Hospital use by clients of the state's Medicaid program fell by 20 percent as of last year, and emergency room visits fell by 12 percent. The decline in emergency room visits alone saves the state $8 million annually. Data sharing accounts for much of the credit for these efficiencies.
The sharing of health data, in particular, often needs a legal framework that both allows access that meets individual departments' needs and ensures compliance with privacy laws and regulations. Minneapolis utilizes a streamlined process that makes legal resources available specifically for these kinds of discussions. The city clerk's office and representatives from individual departments work with the city's legal counsel to vet data as necessary and set any legally mandated boundaries. This city's open-data policy encourages all other types of data to be open automatically, limiting complex legal discussions to an as-needed basis.
But even in successful data sharing and analytics projects, legal battles can prove a difficult challenge and a significant roadblock. Indiana's highly regarded 2014 data sharing project, which led off with an effort to reduce infant mortality rates, was stalled for months as officials handling the technological infrastructure argued with lawyers from individual departments over what data would be released into the system. It took a lead legal advisor appointed by the governor to referee the negotiations, resolving issues across agencies and turning discrete data into an integrated data-driven solution.
The resulting project equipped officials with data analytics that allowed them to identify populations with the highest risk factors for infant mortality and target preventive strategies, including referring mothers who had missed prenatal appointments because they could not find rides to public transportation resources.
Opening up data across departments by default facilitates a comprehensive, ongoing set of collaborative discoveries that are not slowed by needless negotiations that create transaction friction. Whether it is done by running data through a dedicated team at the Minneapolis clerk's office on an as-needed basis or a governor calling in counsel to mediate, as in Indiana, there are strategies aplenty for governments to navigate confidentiality laws and open up departmental data.
In addition, federal privacy laws in many cases are less restrictive than imagined by state or local officials who often act in unrealistic fear of the consequences of a misstep. In March, the U.S. Department of Education released a manual with guidance about sharing student data while maintaining privacy. It breaks down some of the myths about the Family Educational Rights and Privacy Act, including the perception that it effectively prevents most data sharing. The manual also proposes several strategies, including depersonalization of data, so that it can be legally shared, and it highlights the ability of data sharing to help schools better serve students.
A comprehensive approach to data sharing must include policy transparency that enhances trust and reduces the chances of future embarrassment. The benefits of data-driven governance are well worth the time it takes to explain the process to the public. But these benefits can only be realized with an open first approach, effective and efficient anonymization practices, and a problem-solving legal adviser with the authority to make decisions and overcome misperceptions about legal restrictions.
In the end, the risks that come from not sharing data across a governmental enterprise greatly exceed the risks of finding ways to use the data to provide more efficient and effective services. The default setting should be the open one.
This column was originally published by Governing.
Stephen Goldsmith is the Daniel Paul Professor of the Practice of Government and the Director of the Innovations in American Government Program at Harvard's Kennedy School of Government. He previously served as Deputy Mayor of New York and Mayor of Indianapolis, where he earned a reputation as one of the country's leaders in public-private partnerships, competition and privatization. Stephen was also the chief domestic policy advisor to the George W. Bush campaign in 2000, the Chair of the Corporation for National and Community Service, and the district attorney for Marion County, Indiana from 1979 to 1990. He has written The Power of Social Innovation; Governing by Network: the New Shape of the Public Sector; Putting Faith in Neighborhoods: Making Cities Work through Grassroots Citizenship; The Twenty-First Century City: Resurrecting Urban America, and The Responsive City: Engaging Communities through Data-Smart Governance.