Skip to main content
main content, press tab to continue
Report

Digital Health Big Data Risks

By Kirsten Beasley | May 25, 2021

Digital health companies need to explore and consider the emerging areas of concerns and potential liability, related to Big Data within their organizations.
Risque de pandémie

Undeniably, all digital health innovations rely on Big Data – they collect, store, exchange, synthesize and/or produce substantial amounts of health data.

The promise of Big Data to transform healthcare is identical to the promise of digital health innovations: reduced costs, increased efficiency and improved outcomes and access.

Therefore, it is important for digital health companies to explore and consider the emerging areas of concerns and potential liability related to Big Data within their organizations.

What is Big Data?

There is no universal agreement on the definition of ‘Big Data’; the term is generally used to describe the growth and availability of large datasets.

Big Data is frequently characterized in terms of the 7Vs: volume, variety, velocity, validity, value, volatility and veracity.”

Kirsten Beasley
Head of Healthcare Broking, North America, Willis Towers Watson

Big Data is frequently characterized in terms of the 7Vs: volume, variety, velocity, validity, value, volatility and veracity.1

Big Data are data whose scale, and complexity require new architecture, techniques, algorithms, and analytics to
manage it and extract value and hidden knowledge from it.2

In a healthcare context the term often has a multidimensional meaning that incorporates:

  1. the volume and diversity of data available from disparate sources with
  2. the efficient real-time linking and analysis of those data in order to
  3. provide actionable insights and enable informed decision making.

Consequently, ‘Big Data’ in healthcare is not wholly focused on the flood of data; rather, the emphasis is on the analysis, parsing and synthesis of the data into knowledge and understanding.

Why does it matter?

Many healthcare systems, though often data rich, do not properly utilise existing datasets to generate a better understanding of how to improve access to better quality care and to reduce waste.

Such missed opportunities result in unnecessary patient harm and serve to increase the gap between the cost of healthcare and the outcomes achieved.

These system limitations could be overcome by the development of a continuous learning healthcare system that harnesses Big Data to ‘fuel’ a virtuous cycle, in which research informs and influences clinical practice and clinical practice informs and influences research.

Big Data is like a virtuous cycle, in which research informs and influences clinical practice and clinical practice informs and influences research.
Virtuous cycle of continuous learning

In such a system, Big Data can help facilitate a more empirically driven healthcare system, ideally, free from bias, to drive lowered costs, improved quality of care and patient safety and ultimately better outcomes. This is the promise of ‘Big Data’.

Yet, the pathway to these potentially transformational changes is littered with challenges.

Key challenges in the use of data in healthcare

Some of the key challenges in the use of data in healthcare include:

  • Interoperability challenges
  • Data governance
  • Data storage
  • Data accessibility/mobile competency
  • Data ownership
  • Data mining liability
  • Management challenges/need for healthcare analytics talent
  • Cybersecurity

These challenges are covered in more detail in our Big Data in Healthcare whitepaper, download to find out more.

These challenges are covered in more detail in our Big Data in Healthcare whitepaper, download to find out more.”

Doris Fischer-Sanchez | Leader, Clinical and Enterprise Risk Management,
Willis Towers Watson

Data sharing initiatives

Global data sharing initiative examples
Secondary Use Project Nature of data share Objections
A data program A program aimed at extracting data from GPs for a central database Lack of 1) patient awareness of the program and 2) clarity around opt-out options.
Social network company The social network’s personal data was harvested for political advertising Lack of consent and transparency.
Consulting company and a government agency 180,000 lung cancer patients’ anonymized data were shared with the consulting firm for a study on lung cancer trends. Consent of the patients was not obtained.
Artificial intelligence (AI) company 1.6M patients’ data were transferred to AI company to test an acute kidney injury altering system. Inadequate public engagement, awareness and lack of transparency.
A healthcare organization and a technology company Identifiable patient data shared with a technology company to pilot an electronic health record (EHR) search tool. Lack of notification to patients; Federal inquiry to ensure compliance with The Health Insurance Portability and Accountability Act of 1996 (HIPAA).
A healthcare company and a technology corporation Agreement to share identifiable patient data to develop cancer algorithms. Identifiable data utilized; how has consent been obtained?
Online pharmacy Names and addresses of >20,000 customers were sold to a marketing company. Breach of data protection rules by not seeking customers’ consent.

Key takeaways

Clearly, health-related longitudinal data sets hold great promise in the eyes of multiple stakeholders seeking to harness their power in varying ways and to differing purposes.

However, organizations must recognize that when tapping into that potential there is a responsibility to ensure that the public’s health-related data is not abused.

Furthermore, there should be a concerted effort to be proactively transparent about health data reuse in order to increase public awareness and understanding.

The individual is at the core of the health data ecosystem, thus prioritizing their views enables the construction of data reuse policies that will better reconcile the inherent frictions that exist within data sharing.

Improved public awareness about data sharing and linking in conjunction with more transparency about commercial involvement will lay a path towards equitable, sustainable and community-centric secondary use strategies.

Download
Title File Type File Size
Digital Health Big Data Risks PDF 1.8 MB

Sources

1 Khan, M., Uddin, M., and Gupta, N. (2014) Seven V’s of Big Data Understanding Big Data to Extract Value. Accessed at: http://www.asee.org/documents/zones/zone1/2014/Professional/PDFs/113.pdf

2 Bellazzi R. Big Data and biomedical informatics: a challenging opportunity. Yearb Med Inform. 2014;9:8–13. doi: 10.15265/IY-2014-0024. Accessed at: https://pubmed.ncbi.nlm.nih.gov/24853034/

Authors

Head of Bermuda Office,
WTW

Contact us