Data, Data Everywhere, But Access Remains a Big Issue for Researchers

A Review of Access Policies for Publicly Funded Patient-Level Health Care Data in the United States


This study captures the policy inconsistencies and hurdles that can hinder use of publicly funded federal and state datasets for researchers. These limitations can make it harder to conduct high-quality research that seeks to inform clinical decision-making, identify cost-effective, evidence-based care and evaluate innovative systems and payment designs. 

The researchers identified 116 federal and state datasets, and provided an in-depth analysis of nine federal datasets and 10 publicly funded, state all-payer claims databases (APCDs). The nine federal datasets examined represent five federal agencies and include individuals covered under Medicare, Medicaid, the Veterans Health Administration and commercial insurance plans. The 10 state APCDs represented all publicly funded APCDs, which had access policies finalized at the time the review was conducted (November 2014). The authors outlined how the databases differ in accessibility and compared access restrictions to the type or level of health information available in the data.

The study found that there is significant variation—sometimes within the same federal agency—in access restrictions based on the data request’s purpose and the requestor’s affiliation and funding source. In other cases, there were numerous indirect hurdles to using the data, including high user fees, prolonged wait times for data request approval and data delivery, and recency of data. One innovative approach to data access is the decision by the Centers for Medicare and Medicaid Services (CMS) to allow access to its claims data to “innovators and entrepreneurs,” regardless of funding or affiliation.  However, cost and timeliness issues persist.

The authors similarly found variation in data access policies across the 10 APCDs analyzed. Four states—Kansas, Minnesota, Tennessee and Vermont— provided access to state government and contractors only. The other six states—Colorado, Maine, Maryland, Massachusetts, New Hampshire and Oregon—produce types of files that are available to external requestors; however, policies varied considerably.

To reduce barriers and restrictions of data accessibility, the authors presented four key recommendations:

  • Emphasize research quality and intent—Rather than simply focusing on a requestor’s affiliation and/or funding source, data access policies placing emphasis on quality and intent of the research request create greater opportunities while maintaining protection of patient privacy.
  • Increase availability of granular data that can be linked—The ability to link across multiple datasets, including claims and electronic health records, is increasingly important to accelerate understanding and evaluation of the impact that initiatives, such as precision medicine, have on health care quality, effectiveness and value.
  • Utilize technology to enhance data security—Privacy concerns may be allayed by providing remote access to data and ensuring a secure environment by limiting researchers’ ability to download individual-level results.
  • Alleviate indirect access barriers—Use of approaches such as tiered pricing for data user fees, whereby commercial entities with greater resources are charged higher fees that help offset lower fees collected from academic and nonprofits, could enable more investigators to explore important research questions.