Please login to the form below

Not currently logged in
Email:
Password:

AI UK – one year on

Are patients still cautiously embracing new healthcare tools?

Duncan Arbour

By Duncan Arbour, SVP Innovation,Syneos Health Communications Europe

In autumn last year, we started researching our research paper Artificial Intelligence for Authentic Engagement, informed by a survey of over 1,000 patients and caregivers across the EU and United States.1

Published in January 2018, our work found that despite great excitement for healthcare AI in tech and venture capital communities, patients themselves were at best only cautiously embracing the new, artificially intelligent healthcare tools and services.

Our survey data suggested that there were two main barriers that developers and providers of future tools should address: the perception that AI might come to replace human interactions with physicians; and the concern that personal health – and personal healthcare data – could be put at risk.

To overcome these barriers, our research suggested that the public needed to see strong leadership around future ethics and regulation of AI in medicine and healthcare. Specific to the UK, this was a role that patients wanted to see taken on by the NHS and the physicians it employs.

One year on from that research, a review of progress against those barriers highlights an increasingly positive picture of AI tools that will be well received by both patients and physicians, and be properly supported by Government and the NHS.

A changing understanding of AI vs. physician roles

The dominant media narrative around healthcare AI during the period of our research process was driven by Silicon Valley and investors. It was a story that suggested AI could outperform and would inevitably replace physicians, with a particular focus on imaging-led disciplines. Now this narrative is changing for the better.

Of particular interest is a recent study from Stanford University, presented at the 2018 SIIM Conference on Machine Intelligence in Medical Imaging this September, which looked explicitly at the role of radiologists – the specialists long- considered to be the canary in the AI coal mine.2

This study focused on the use of chest x-rays to diagnose pneumonia and compared three different approaches: using individual physicians alone; using an algorithm previously proven to have outperformed physicians in the same task; and using a combination of the two.

This final approach won, showing that a small group of doctors moderated by AI algorithms could make a more accurate diagnosis than physicians (by 33%) versus AI alone (by 22%).

Strong moves to address concerns around trust and privacy

Our 2017 survey data made it clear that while most of us are comfortable with using the services of Big Tech (eg Apple, Amazon, Google and Facebook) in our daily lives, healthcare is clearly different. Only 14% of our EU respondents reported, for example, that they would be likely to use an AI-powered virtual healthcare assistant if provided by one of these companies.

Their declared preference was for such future tools to be developed instead under the guidance of healthcare professionals, and with ethics and regulation supervised top-down by national healthcare systems.

Now this viewpoint is further borne out in How the UK can win the AI race3 published by KPMG this September, which asked 2,000 respondents which organisations they would be willing to share their personal data with in exchange for improved services or capabilities. At the bottom of the league table were tech companies, media companies and political organisations, and at the top – trusted by 56% – is the UK’s NHS. Then in the same week as the report’s publication, UK Health Minister Lord O’Shaughnessy announced an initial Code of Conduct for Data-Driven Health and Care Technology4 to guide use of third-party technologies in the health service going forward, including strong provisions for shaping how data should be both protected and used for real benefit to patients and NHS staff.

More than ever, then, it seems that the best path for bringing properly patient- centric applications of healthcare AI into practice remains the same as we suggested in our 2018 report. Doing it alone won’t cut it, nor will partnership with tech companies alone. Success requires active participation in a far more open conversation, and in far more open partnerships – with government, with patient groups and physicians.

What have been largely missing up to this point, however, are the frameworks, guidance and structure to make this happen. But now all the signs point to an engaged and supportive NHS becoming the driving force behind
this conversation. It’s a conversation that there’s never been a better time to join, and maybe even, as the KPMG report suggests, to help the UK become a leader in this field.

1. https://syneoshealthcommunications.com/perspectives/artificial-intelligence-for-authentic-engagement/
2. https://cdn.ymaws.com/siim.org/resource/resmgr/mimi18/abstracts/18paper4-Halabi.pdf
3. https://assets.kpmg.com/content/dam/kpmg/uk/pdf/2018/09/how-the-uk-can-win-the-artificial-intelligence-ai-race.pdf 
4. https://www.gov.uk/government/publications/code-of-conduct-for-data-driven-health-and-care-technology

In association with

Syneos Health

19th October 2018

From: Marketing, Healthcare

Share

Tags

Featured jobs

Subscribe to our email news alerts

PMHub

Add my company
90TEN Healthcare

90TEN is an award-winning healthcare communications consultancy that puts people at the heart of everything we do. Our Life.Changing. campaigns...

Latest intelligence

World Pancreatic Cancer Day 2018: Combating misconceptions in pancreatic cancer
Patients diagnosed with pancreatic cancer face a dismal prognosis, with the disease having the lowest survival rate of all major cancers. In spite of this, pancreatic cancer research is chronically...
World Diabetes Day: Interaction and impact of diabetes on mental health
For World Diabetes Day on the 14th November 2018, Nisha Shahrukh - Medical Writer at Mednet Group has written an article depicting the impact diabetes has on mental health. Including...
EU
Innovation in merger control and the impact on the pharmaceutical sector
Is focusing on pipeline products enough to assess regulatory risks?...

Infographics