Add us as a preferred source
Bob Davidson, managing director of Total Doctor, explains that greater access to information does not necessarily produce better-informed decisions.
Healthcare has never had more available information. Patients can access thousands of articles about a condition, compare treatments and clinicians, watch medical discussions and ask generative AI systems to explain symptoms, diagnoses and treatment options. Search and AI may also introduce specific hospitals, clinicians or technologies.
An extraordinary volume of instant information is within reach. However, greater access to information does not necessarily produce better-informed decisions. In some circumstances, it creates a new problem – which information deserves to be trusted? We refer to this as clinical authority.
Clinical authority asks a series of questions of healthcare information: Who produced it? What expertise do they have? What evidence underpins the information? Is that evidence current? Are there credible alternative interpretations? Are commercial interests transparent? And is the information being presented within the appropriate clinical context? These questions are important because healthcare information increasingly influences decisions before the patient even enters the consulting room. This is what makes clinical authority an issue of clinical risk as well as communication.
Clinical risk can begin upstream
Clinical-risk management understandably concentrates on what happens within healthcare delivery. The key stages include: diagnosis, treatment, medicines, procedures, systems, consent and outcomes. However, the risk begins earlier.
Poorly evidenced, outdated, commercially influenced or decontextualised information can create inappropriate expectations or misunderstandings about benefits, risks and alternatives. Conversely, information with clear provenance, appropriate evidence and expert oversight can support better-informed conversations between patients and clinicians.
This becomes particularly significant as AI assumes a larger role. Recent UK regulatory developments are focusing on lifecycle governance, transparency, evidence and accountability around healthcare AI. NHS England is also reviewing its DCB0129 and DCB0160 clinical-risk standards against a background of AI-supported decision-making, interconnected systems and cross-organisational responsibility.
The underlying issue is important because an AI-generated answer may sound “authoritative” – even without making the source of that authority apparent. Healthcare therefore needs to preserve a chain between information, evidence and accountable human judgement.
From information to an auditable chain of authority
One useful way of considering this is as a progression from data, to AI analysis, to evidence, to expert interpretation, to patient discussion and then accountable decision.
At each stage, the provenance or source matters and should include points such as: Where did the information originate? Which evidence was considered? Which AI system or model contributed? What limitations or uncertainty were present? Did an appropriately qualified professional accept, reject or modify the conclusion?
Healthcare Today recently highlighted a closely related issue in AI-generated clinical inferences that observed data, model output and clinical assessment need to remain distinguishable because once those stages become blurred, provenance itself becomes a patient-safety concern. Clinical authority could provide a useful framework for thinking about this wider challenge.
We currently see six important components, including provenance, evidence, expertise, independence, context and governance.
A distinguished clinician can still express an opinion unsupported by current evidence. A systematic review can be accurately quoted but applied outside the population for which its findings are relevant. Commercially funded information can be valuable, but the relationship should be transparent, and funding should not determine the clinical conclusion.
What can healthcare organisations do now?
Clinical authority does not require another regulatory framework before organisations can start improving their approach. Clinicians can begin by considering whether their digital presence demonstrates why their expertise should be trusted rather than merely listing qualifications and services. Evidence-based patient education, transparent authorship, appropriate references, declared interests and regular updating all contribute to establishing genuine authority.
Healthcare organisations can go further by examining the information surrounding important patient decisions. Does it explain realistic alternatives? Are benefits and risks placed in context? Is legitimate clinical disagreement acknowledged? Can patients distinguish independent clinical evidence from promotional claims?
And organisations introducing AI should increasingly consider whether the authority behind that answer can be reconstructed and challenged. All this may eventually require provenance, attribution, evidence, versioning and governance to become part of healthcare’s digital infrastructure rather than simply good editorial practice.
Authority cannot simply be bought
Healthcare organisations legitimately want their expertise to be discovered. Clinicians with exceptional experience should be able to communicate it, and patients should be able to find them. However, clinical authority should not become another name for content marketing. Commercial participation can fund the development and communication of authoritative clinical information but should not be able to purchase the clinical conclusion.
This separation and difference may become increasingly valuable in an information environment where producing more content is becoming almost costless. AI can produce information; however, the scarcer resource may increasingly be the evidence, provenance, expertise and accountable judgement that give that information authority.
When information is abundant, clinical authority becomes the most valuable asset of all.
