Part I: Data Transformation in Healthcare and Government Starts With Shared Data

machine learning
Data Society           
machine learning
  November 2022              
machine learning         Blog
healthcare data science

Enterprises across industries and sectors have adopted data science technologies to remarkable effect. However, much of the revolutionary power of data remains untapped as organizations grapple with data-sharing challenges. This article is the first of a two-part series examining data sharing in two critical domains with profound public impact, healthcare and government, exploring the opportunities and obstacles accompanying large-scale data-sharing initiatives. 

Better Outcomes Through Data Sharing

healthcare data science

For healthcare, life sciences, and government organizations, the pandemic highlighted the need for—and promise of—enhanced data-sharing facilities. Responding to a global crisis allowed them to demonstrate their capacity to fill this need promptly and effectively. This achievement provides a glimpse of possibilities for future advances through coordinated efforts to pool data, and some recent policy developments promise that this momentum can and will continue.

Data sharing can improve government functions ranging from routine civic services to critical crisis responses by facilitating the communication of insights between agencies and departments. According to a McKinsey article, collective access to the vast supply of government data can help reduce administrative redundancy in data gathering, inform insight-driven policymaking, enhance security capabilities, and aid in fraud and waste detection.

Secure and efficient data sharing has long been a major hurdle. As a result, the U.S. Department of Justice, Department of Homeland Security, and Department of Health and Human Services have formed a partnership, National Information Exchange Model (NEIM), that supports information exchange processes and standards to facilitate data flow across agencies. NEIM has proven effective as a source of data-sharing resources serving as the foundation for programs such as the National Electronic Interstate Compact Enterprise (NEICE), which streamlines what was once a paper-based information exchange between federal, state, and civilian agencies to help states place children in families across state lines. The result has been a significant decrease in processing time from 6-12 months to 1-2 days. 

Similar issues confront healthcare organizations. The healthcare and life sciences industries are beneficiaries of a booming data influx from public and private healthcare sources, researchers, and digital devices. With this bounty of data comes an increased impetus to unite data that can, when combined, inform strategies for better patient care, the development of new therapies, and operational efficiency. This trend toward increased data sharing is reflected in a statistic from the US National Library of Medicine’s PubMed Central, which notes that the number of PubMed articles containing the keywords “data sharing” rose from 46 in 1,980 to 5,960 in 2019.

For organizations in these fields, data sharing prepares a solid foundation for AI- and ML-driven progress in personalized medicine, public health, clinical trials, drug discovery, and operational efficiency. For example, improved data-sharing capabilities can help accelerate and rescue the cost of research validation. According to a Nature article estimate, irreproducible biomedical studies cost between $10 and $50 billion in the US each year, leading to delays in potentially groundbreaking discoveries. The National Institutes of Health (NIH) introduced its Data Management and Sharing Policy to address this issue. Effective in 2023, these procedural guidelines will require scientific research teams to provide data sharing and management plans along with NIH research applications, encouraging the exchange of research data that other groups would need to validate and replicate their research. 

healthcare data science

Hurdles Between Data Sharing’s Present and Future

Despite the strides government and healthcare organizations have made toward robust data sharing, challenges lie ahead on the road to optimal data sharing. Obstacles yet to be cleared include:  

Data Ownership and Public Trust Concerns - For both government agencies and healthcare providers, gathering and sharing personal data must be accompanied by responsibility for security, accuracy, and a level of transparency that doesn’t compromise record de-identification. Public reluctance to share personal data with government agencies can be especially challenging, given common concerns about how the data will be used and how anonymous it is. 

The healthcare industry faces similar doubts about its use of patient data. According to a Q-Centrix survey, 35 percent of respondents said they didn’t believe their electronic medical records contained accurate information. In addition, when asked if they believed that organizations were doing everything they could to protect their health information, 35 percent of respondents said they were unsure, 37 percent did not think so, and 71 percent reported feeling comfortable sharing their health data only with providers, hospitals, and pharmacies that have treated them personally. 

Measures to help educate the public about using and owning their personal data could help ease some of these concerns. For example, federal regulations that took effect on October 6 might help ease public concerns about the use of electronic health records. The new rules governing information blocking expand the definition of Electronic Health Information (EHI) and, therefore, the breadth of personal health data that patients can now access and control. In addition, by granting patients fuller ownership of their health data, these regulations enable them to understand how their records are used and decide how they are shared. 

healthcare data science

Data Disconnect - Many data resources remain siloed, hindering access and flow between disciplines, departments, organizations, and agencies. These persistent data-sharing limitations have various causes, including entrenched institutional tendencies to hoard or protect data, concerns about exposure to security risks, and technical challenges associated with standardizing conflicting data types and reconciling mismatched formats from diverse platforms. This fragmentation can slow or prevent the discovery of potentially significant links to population trends or individual health. In addition, it can bog down efforts to glean revealing insights that can drive more efficient, equitable, and effective public services. 

What the Future of Data Sharing Demands

Given data sharing’s potential to unlock tremendous community value, healthcare and government organizations must continue to evolve toward seamless, secure, and reliable data-sharing technologies and procedures. However, achieving this future demands long-term strategies to overcome the cultural and technical roadblocks described above. Part II of this series will explore solutions necessary to ensure these strategies succeed, such as skilled data management and interoperability practices.


Subscribe to our newsletter

cross linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram