NHS data issues could stall AI healthcare revolution, experts warn

Former Prime Minister Tony Blair recently advocated for the integration of AI “doctors and nurses” within the National Health Service (NHS), emphasizing the need to leverage health data for innovation. Speaking at SXSW London, Blair highlighted the potential of AI to transform healthcare, yet experts caution that without addressing fundamental data issues, such advancements could introduce more complexity than solutions.

The NHS faces significant challenges, notably in early cancer diagnosis. As of 2021, only 54.4% of cancers were diagnosed at stage 1 or 2, falling considerably short of the 75% target set for 2028. While Blair’s call for greater ambition in AI is acknowledged, the critical prerequisite remains: accessible, high-quality data. Without this, the ambitious 75% target may remain elusive.

Flann Horgan, Vice President, Healthcare at NTT DATA UK & Ireland, whose organization is partnering with The Royal Marsden on an AI-powered radiology platform for cancer research, echoes this sentiment. “The real barrier to Sir Tony Blair’s vision of AI doctors is secure access to high-quality data,” Horgan stated. He pointed to decades of “localised attempts at digitisation” that have struggled due to legacy systems not designed for integrated AI care.

Horgan emphasized that AI’s role is to support clinicians, not replace them. “I’ve seen first-hand how AI can support clinicians instead of replacing them. The Royal Marsden… is already developing and evaluating AI algorithms to improve the accuracy of cancer diagnosis. But that impact depends entirely on having fast, accurate data available at the point of care.” He warned that without interoperability and clear data standards, “AI risks becoming just another layer of complexity… rather than a tool that genuinely improves care.”

However, there are positive developments. Horgan noted progress with Regional Secure Data Environments (SDEs) providing structured access to population-level health data. He also highlighted the NHS Federated Data Platform as a step towards more intelligent, system-wide decisions and a potential “unified national backbone.” Furthermore, the proposed £600 million investment by Keir Starmer and the Wellcome Trust in a National Health Data Research Service signals growing momentum for robust clinical research infrastructure.

“We have a golden opportunity to reframe the patient experience,” Horgan concluded. “But data is not a static resource. Its value depends on ensuring clinicians can trust and act on AI insights. And the value of AI itself lies in its ability to enhance clinical judgement, not replace the doctors who drive patient care forward every day.”

The consensus is clear: while the potential for AI in the NHS is vast, its successful implementation and the achievement of critical health targets hinge on overcoming the long-standing challenge of data accessibility and quality.

Featured image credit