Principles for Fostering Health Data Integrity

Almost every generation is confronted with the effects of its past and must adapt. In his 1962 “We choose to go to the Moon” speech, President Kennedy juxtaposed the challenges of his postwar era—intelligence vs. ignorance, good vs. evil, leadership vs. fear-fueled passivity—and harnessed the national animus to achieve a lunar landing.

Today, our challenge categories are similar. We are confronted with the effects and portents of concurrent changes in medicine, science, and technology, which in turn change how we educate scientists, manage the implementation of new technology, and respond to the effects, both planned and unforeseen, of the application of our discoveries.

Computational and data science technologies, some rooted in JFK’s ’60s, have entered all facets of life at breakneck speed. Our understanding of the societal effects of emerging technologies is lagging. When data, data transfer, and artificial intelligence meet medicine, game-changing implementation effects—positive or negative—are imminent.

In “How Health Data Integrity Can Earn Trust and Advance Health” (Issues, Winter 2024), Jochen Lennerz, Nick Schneider, and Karl Lauterbach tackle this complex landscape and identify pivotal decisions needed to create a system that equitably benefits all stakeholders. They highlight a requisite culture shift: an international ethos of probity for everyone involved with health data at any level. They propose, in effect, a modern-day Hippocratic Oath for health data creation, utilization, and sharing—a framework that would simultaneously allow advancement in science and population health while adhering to moral and ethical standards that respect individuals, their privacy, and their medical needs.

Without this health data integrity framework, the promise of medical discovery through big data will be truncated.

When data, data transfer, and artificial intelligence meet medicine, game-changing implementation effects—positive or negative—are imminent.

Within this framework, we open new horizons for medical advancement, and we augment the safety of data and of tools such as artificial intelligence. AI is an oxymoron: it is neither artificial nor intelligence. AI determinations derive from real data scrutinized algorithmically and, at least currently, they appear intelligent only as the data evaluation is iterative and cumulative—temporally updated evaluations of compounding data sets—a heretofore quasi-definition of intelligence. These data serve us—patients, health care providers, researchers, epidemiologists, industry, developers, or regulators. With greater harmonization and data integrity, data utilization becomes globalized. Wider use of data sets can lead to more discoveries and reduce testing redundancies. Global data sharing can limit the biases of small numbers and identify populations of low prevalence (e.g., rare diseases), allowing the creation of larger, global cohorts.

Pathologists, like the article’s coauthor Jochen Lennerz, are physician specialists trained to understand data; we are responsible for the generation of roughly 70% of all medical data. Pathologists, along with ethicists, data scientists, data security specialists, and various other professionals, must be at the table when a health data integrity framework is being created.

Within this framework, we will benefit from a system of trust that recognizes and respects the rights of patients; understands, and supports, medical research; and ensures the safe, ethical, transfer and sharing of interoperable, harmonized medical data.

We must ensure the steps we take with health data are not just for a few “men,” to borrow again from the lunar-landing lexicon. Rather, we must create a health data ecosystem of integrity—a giant step for humankind.

Vice President for Medical Affairs, Sysmex America

Governor, College of American Pathologists (CAP)

Chair, CAP Council on Informatics and Pathology Innovation

Jochen Lennerz, Nick Schneider, and Karl Lauterbach report how efforts to share health data across national borders snag on legal and regulatory barriers and suggest that data integrity will help advance health.

In today’s digital transformation age, with our genomes fully sequenced and widely deployed electronic health record systems, addressing collaborative digital health data use presents a variety of challenges. There is, of course, the need to ensure data integrity, which will demand addressing such issues as the relative lack of well-defined data standards, poor implementation, and adherence, as well as the asymmetry of digital knowledge and innovation adoption in our society. But a more complex challenge arises from the propensity of humans to push major inventions beyond their benefits—and into the abyss. Therefore, we must engage together for human integrity in collaborative health data use.

Yet another challenge—one that the authors cite and I agree with—arises from deep-rooted conflicts of interest among all stakeholders (patients, health care professionals, the health management industry, payors, and governments) in health care. There also are generational differences between tech-savvy younger health care professionals, who are generally more open to structured data collection and documentation, and more senior ones, who struggle with technology and contribute health data that is more difficult to process.

There is, though, overall agreement among health care professionals that their foremost task is to serve as their patients’ advocate and go above and beyond to help them overcome or manage their medical problems using every available resource, which today would clearly include taking full advantage of digital health innovations, health data, and associated technologies such as artificial intelligence.

A more complex challenge arises from the propensity of humans to push major inventions beyond their benefits—and into the abyss. Therefore, we must engage together for human integrity in collaborative health data use.

However, since medicine has become such a complex profession, health professionals often practice in large care facilities embedded in organizations operated by corporations that seek profits, and where payors strictly regulate access to and extent of utilization of care on behalf of governments that struggle with expenditures. Unsurprisingly, the goals of nonpatients, administrators, and others outside of health care might not be what health professionals would view as ethical and responsible in terms of health data collection and use.

Among still other obstacles to the protection of health care data, cybercrime risks with hackers who either for personal gain or on behalf of third parties attack our increasingly digital world are a major threat. And then there is the important matter of people’s individual freedom, which at least in most Western democracies includes the right to informational self-determination and privacy. Ensuring these rights needs to be balanced with the societal goal of fostering increasingly data driven medical and scientific progress and health care delivery.

Once all stakeholders in medicine, health care, and biomedical research realize that our traditional approach to diagnosis, prognosis, and treatment can no longer process and transform the enormous volume of information into therapeutic success, innovative discovery, and health economic performance, we can join forces to unite for precision health. For details, I’ve laid out a vision for collaborative health data use and artificial intelligence development in the Nature Portfolio journal Digital Medicine.

Put briefly, precision health is the right treatment, for the right person, at the right time, in the right place. It is enabled through a learning health system in which medicine and multidisciplinary science, economic viability, diverse culture, and empowered patient’s preferences are digitally integrated and conceptually aligned for continuous improvement and maintenance of health, well-being, and equity.

Professor of Medicine and Adjunct Professor of Computing Science

University of Alberta

Director, Collaborative Research and Training Experience “From Data to Decision

Natural Sciences and Engineering Research Council of Canada