NHSX to create policy guide for use of AI in healthcare
New tech organisation to produce guide for how new technology can be ‘responsibly applied’
NHSX is to create a “policy toolkit” designed to ensure that artificial intelligence is used responsibly in the delivery of healthcare.
The newly created tech organisation has been asked to work with patient groups and other parts of civil society to come up with guidelines for the use of AI, according to Jackie Doyle-Price, minister for mental health, inequalities and suicide prevention.
“NHSX is driven by user needs, so any technology we introduce, including that related to artificial intelligence, must be done with, and not for patients,” she said. “We are developing a policy toolkit for responsibly applied AI in health and care, and we are working closely with public-voice organisations to ensure that patients are involved and engaged.”
- Why the NHS needs to ‘take the lead’ on sharing data with the private sector
- AI will save lives of 22,000 cancer patients a year, prime minister announces
- NHS explores options for training staff on AI and robotics
Doyle-Price, who was responding to a written parliamentary question from Conservative MP for Crawley Henry Smith, added that NHSX will support ongoing work to develop “tools to help technology vendors comply” with the Code of Conduct for Data-Driven Health and Care Technology published in September 2018 by the Department for Health and Social Care.
Last month, tools to aid compliance with principle 7 of the code – which covers transparency around how algorithms are created, and how they work in practice – “were tested with patients and we are currently refining based on the feedback we received”, according to Doyle-Price.
She added: “We will continue to engage with patients as we work to maintain and improve public trust and patient safety.”
In answer to a further question from Smith, the mental health minister revealed that NHSX will also head up “a cross-cutting programme of work across the health and care system” dedicated to AI.
A particular remit of this programme will be delivering the recommendations made in a recent report – dubbed Putting patients at the heart of artificial intelligence – published by the all-party parliamentary group on heart and circulatory diseases.
The group’s report made six key recommendations:
- NHSX should work charities and citizens to explore patients’ views and concerns about the use of AI in healthcare
- Non-profit organisation Understanding Patient Data should work with with the public and charities to create tools to better engage citizens on issues related to AI
- The 15 Academic Health Science Networks across England should support the flow of information between patients’ groups and commercial providers
- NHS England and NHS Digital should examine how both AI and wearable technology will affect healthcare inequality
- NHSX and Understanding Patient Data should engage with the public, the third sector, and patients’ groups to ensure policy is designed that protects “public values”
- NHS England and The National Institute for Health and Care Excellence should create “standards for publication for AI research”
Doyle-Price said: “The programme will also make it easier for suppliers to develop technologies that tackle some of the biggest issues in healthcare. Officials from the Department [of Health and Social Care] were part of the advisory board for the report and supported the development of the recommendations.”
The creation of NHSX – which will assume overall nationwide responsibility for technology strategy across the health service – was announced in February. Last month, DCMS digital and media director Matthew Gould was announced as the chief executive of the new entity. He will lead an organisation that will, in effect, sit above NHS Digital, and Gould will report into health secretary Matt Hancock and NHS England head Simon Stevens.
The body dedicated to upholding ethical standards across the public sector has published a major report examining how to ensure those standards are not threatened by AI and automation
With many government-developed services seeing poor uptake, the answer may lie in allowing citizens to ‘bring your own identity’, according to Arthur Mickoleit of Gartner
Exercise launched following conclusion of Paterson Review
Two national services have been migrated to cloud environment