Experts Decry Data Privacy's Costs for Health Care
While well-intentioned, privacy regulations in health care result in fragmented patient data across systems and institutions, which impedes optimal care and scientific advancements, said a Tuesday blog post from consultants at Access Partnership, a public policy firm.
Sign up for a free preview to unlock the rest of this article
Not surprisingly, data shows patients find the health care system a bear to navigate, but the system is fixable, wrote Nada Ihab, deputy lead of data governance, and Trey Flowers, senior manager of tech-enabled verticals.
Privacy isn't solely responsible for the fragmented health care data ecosystem, the consultants said. For example, cultural and institutional attitudes treat data as a competitive asset and guard information jealously. Moreover, concerns about data breaches motivate health care systems to share information reluctantly, they said. Differing interpretations of the Health Insurance Portability and Accountability Act (HIPAA) also contribute, as they result in a conservative, restrictive approach to data sharing, they said.
The resulting data silos slow researchers, who can't collect "sufficient data for meaningful studies," said the consultants: Silos also block clinicians who sometimes lack “comprehensive patient information." The results are often "repeated tests, delayed treatments, and increased frustration, ultimately compromising the quality of care,” not to mention inefficient spending, said Ihab and Flowers: Privacy restrictions also "limit analytical tools that could identify patterns across large populations."
Workarounds exist, they said. For instance, Estonia's national health information system issues citizens a digital ID that allows them to access their records and control who can view them. The blockchain-based network integrates data at all levels of care.
In the U.S., the Mayo Clinic uses federated learning techniques, a clinic platform and de-identification practices, which ensure patient data safety while still permitting access to a blend of health care data, the bloggers said. Other methods include distributed analytics, privacy-preserving computation methods and standardized interfaces and frameworks, they said.