Unlocking the Power of Open Health Data: A Checklist to Improve Value and Promote Use

    loading  Checking for direct PDF access through Ovid

Excerpt

Within the last decade, a “quiet open data revolution” has diffused, whereby government agencies are proactively publishing their data online in searchable directories and “open formats” for citizens, journalists, researchers, developers, and others to use.1 The Obama Administration's 2009 Open Data Directive catapulted this movement by requiring that federal agencies publish downloadable data in open formats that are “platform independent, machine readable, and made available to the public without restrictions that would impede the reuse of that information.”2(p2) Many states and local governments have subsequently developed their own open data policies. It was previously possible to access some government data online, but data were frequently difficult to locate or complicated approval processes were required to access data, as well as required the use of specialized software. The innovation is making these data readily discoverable often through interactive Web portals; posting them in multiple data formats; ensuring data are free and with unlimited use and distribution rights; posting data that were not previously available online; systematically preparing detailed metadata (“data about the data,” such as codebooks and data collection procedures) to help users understand the data; and actively engaging data “consumers.”
The US Department of Health and Human Services (DHHS) and the New York State Department of Health (NYSDOH) were early leaders, with robust open data platforms featuring 3000 and 500 health-related data offerings as of January 2017, respectively.3,4 Early benefits include improved data quality, timeliness, and usefulness as agencies prepare data and metadata for publication and more users examine the data; improving external users' data access to generate scientific discoveries and applications; breaking down data silos, thereby improving access within agencies; promoting data-driven improvements in the quality of care and built environment; improved health literacy and medical decision making as patients can access information such as professional medical conduct actions5; reaching new data consumers including journalists and developers; and promoting government transparency.6 Data use cases include employing New York's nursing home bed census data to identify open beds to transfer patients during the Hurricane Irene evacuation; data-based media reports about epidemic rates of childhood obesity in New York and disparities across school districts, the Chicago Tribune's crime statistics visualization; improving public awareness of pedestrian injuries in San Francisco and guiding policy development; predictive modeling to help Chicago public health workers target food establishments for inspections; allowing “medical hot spotters” to use de-identified hospital discharge data to highlight variation in cost, quality, and outcomes; using open data resources to examine local variation in the relationship between childhood obesity and the built environment; the development of the iChoose600 application that the public can use to select restaurant meals under 600 calories; and U.S. News & World Report partnering with RowdMap, Inc, to integrate quantitative data from the federal HealthData.gov portal into its annual reports on best hospitals and doctors to provide patients with information on local physicians who provide unnecessary or inappropriate “low-value” care.1,7–9 As open data are relatively new, these early use cases are only the tip of the iceberg; future benefits could possibly include pilot studies for academic research, using open data in the classroom to train the future public health workforce, and synthesizing data across sectors for new discoveries.6,10
These use cases highlight open data's potential to reduce health care costs, improve health care quality, and promote health, but data are only valuable if used. This requires fostering an open data ecosystem that allows diverse stakeholders with varying interests and technical skills to access usable data. We propose 10 strategies for agencies and private corporations to consider when opening their health data.
    loading  Loading Related Articles