The Data-Driven Future of Healthcare

DZone 's Guide to

The Data-Driven Future of Healthcare

Data promises to transform the relationship between us and healthcare providers.

· AI Zone ·
Free Resource

Over the last few years, I’ve written repeatedly about both the tremendous value society can derive from better use of health data and the various challenges involved in doing so. It’s well-known that most advances in AI today have been fueled by remarkable increases in the amount of data available to train algorithms on, and indeed, the various breakthroughs in healthcare have all followed this pattern.

Most of these rely on a single form of data, such as a medical scan rather than a series of data about the individual, but there are nonetheless some fascinating projects along those lines. Perhaps the most extensive is the Project Baseline that is being run by Google’s life science division Verily.

The project, which is run with partners like Stanford Medicine and Duke University School of Medicine, aims to harvest huge quantities of health data from the 10,000 or so participants in the project. This will include not only traditional medical data such as x-rays and heart scans but also lifestyle data captured via Google’s Study Watch. They’ll even have their genomes mapped to provide a holistic overview of each participant.

“No one has done this kind of deep dive on so many individuals. This depth has never been attempted,” the team said upon the launch of the project. “It’s to enable generations to come to mine it, to ask questions, without presupposing what the questions are.”

Cultural Change

Of course, achieving this on a larger scale is very hard, and a recent report from Health Data Research UK argues that a radical culture change is needed for the National Health Service (NHS) to start benefiting from the huge potential of health data.

The paper was the result of a number of workshops conducted with the public, patients, and healthcare professionals, as well as policy workshops with a number of stakeholders in health data, including from the pharmaceutical industry and digital health companies. The evidence from these sessions was then aggregated with desk research to inform a number of principles developed by a steering group.

It began by identifying the purpose of using data, and especially in striking a balance between realizing the value of data to improve healthcare, whilst also ensuring that these benefits are distributed across the population so that people can manage their own health.

Interestingly, the issue of data ownership remains pretty muddy. For instance, the paper suggests that most participants were happy for their data to be used to improve their care, providing the data was used transparently. This perhaps makes sense with data held in electronic medical records, but less so when the panapole of data generated by wearable devices, mobile apps, or genetic tests are added to the mix. It’s surely much less clear what should happen with any data of this nature that are shared, and who it should be shared with. There is a little real impression that the individual would retain control and ownership of their own data, which is something of a worry.

Co-Creation of Better Health

Perhaps most interesting, however, is how data promises to transform the relationship between us and healthcare providers. Traditionally, the relationship has been very prescriptive, as doctors had the knowledge and the tools that patients lacked. Data is empowering people to take more control of their own health, however, and in many instances of maintaining health and managing long-term conditions, our knowledge will surpass that of our health professionals.

This is likely to result in a shift towards a more equitable partnership-based approach, both in the way care is provided but also in the kind of tools used in doing so.

“The need for active and meaningful partnership with patients and the public as a foundation for decision-making in health and research has also been repeatedly highlighted,” the authors say.

Suffice to say, this represents a considerable challenge as the NHS is not set up for such work at the moment, and in many cases, lacks the skills or the focus to do so. Historically, it has proven incredibly difficult to change the system, such as its size and complexity, and the slow pace of data-driven transformation is testament to these difficulties.

In many instances, the NHS is struggling to adopt technologies that are incredibly mature, much less those of more recent vintage. What’s more, the fragmented nature of the system has resulted in technological leadership across the NHS being sorely lacking. There is a strong sense that the NHS could play a crucial role in both setting the standards for data-sharing in healthcare in the UK and also governing that framework, but there is little evidence that it’s a leadership role they are willing or able to take on.

“Health technologies that use patient data have huge potential to improve our health and wellbeing. We are already seeing the development of wearable monitors linked to automated treatment that are revolutionizing the lives of patients with long-term conditions such as diabetes. Our workshops with the public emphasized that they want to see the NHS deliver on the potential of data-driven technologies, giving better and safer health care for all,” the authors say.

Data-driven transformation has promised to disrupt a great many industries, but you sense that nowhere is this promise greater than in healthcare. If past evidence is anything to go by, it’s a nettle that the NHS will fail to adequately grasp.

artificial intelligence, data science, machine learning

Published at DZone with permission of Adi Gaskell , DZone MVB. See the original article here.

Opinions expressed by DZone contributors are their own.

{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}