All content

Post-market surveillance of AI health tools must be “beefed up” to protect doctors as well as patients, England’s patient safety commissioner says. Henrietta Hughes also told The BMJ it was vital to establish clarity on where clinical liability sits when, not if, AI tools harm patients. Hughes, a GP and a former medical director at NHS England, is deputy chair of the National Commission into the Regulation of AI in Healthcare. The commission was set up by the Medicines and Healthcare Products Regulatory Agency (MHRA) to help guide development of a new regulatory framework for AI medical devices. The commission published interim findings from its consultation and engagement process last week. Hughes said some clear themes had already emerged during the process of engagement with patients, the public, and doctors. Among the most pressing was the need for greater surveillance of AI tools after approval, so the MHRA can act if patients are at risk. Hughes told The BMJ, “It’s really important that real time, real life monitoring happens when a device like AI is deployed in a real life clinical environment, particularly if the population of patients may be different from the population used to feed the model.” Hughes added that while medicines have to pass an “extremely high hurdle” and evidence base to reach the market, AI—where new products are rapidly launched and updated—is different. “We know that AI can change once it’s actually deployed, and so it’s important that the regulations are able to be updated to take account of that and to ensure that all medical devices, and particularly AI, are safe across its whole life cycle,” she said. “Whether we’re using the yellow card system or other kinds of ‘always-on’ postmarket surveillance and postmarket monitoring, that side of things really needs to be significantly beefed up if we’re going to lower the hurdles for products to come onto the market.” Read full story Source: BMJ, 18 June 2026

As artificial intelligence (AI) becomes deeply embedded in triage and clinical workflows, experts are raising concerns about a growing “blind trust” where clinicians and patients alike defer to algorithmic confidence over independent medical judgment. Speaking at the HLTH Europe 2026 conference, panellists stressed that a person’s information ecosystem —who they follow on social media, the podcasts they listen to, and how they interact with AI — is becoming a dominant determinant of health outcomes. Speaking at the event, Patient Safety Learning’s Chief Digital Officer Clive Flashman defined blind trust in this new era as the moment a “clinician stops being able to think independently, independently judging what they see, feel, or hear, because the algorithm has told them something that they should believe or do.” Read full article. Source: Medscape, 21 June 2026

A ‘landmark moment’ is being celebrated in the NHS as a first-of-its-kind therapy that can delay the onset of type 1 diabetes for up to three years will be made available. The National Institute for Health and Care Excellence (Nice) has approved teplizumab, which the charity Diabetes UK said “marks the start of a new age of type 1 diabetes treatment”. Teplizumab, also known as Tzield and made by Sanofi, is approved for children aged eight and over and adults who have type 1 diabetes in its early stage before symptoms appear. It is given as a one-off course and trains the immune system to stop attacking pancreatic cells. Evidence shows the drug can delay the onset of type 1, meaning people can live a fuller life and children can have longer before having to aggressively manage their diabetes. Nice estimates that around 1,100 people could be eligible for teplizumab in the first year, dropping to around 820 patients in the coming years. Read full story Source: The Independent, 23 June 2026

A trial examining the risks or benefits of drugs that can delay puberty for gender-questioning children will help reduce harm, according to the author of a landmark review. Dr Hilary Cass said she was "absolutely convinced that more children will be harmed if we don't do the trial than if we do." Her comments follow pressure from campaigners and some politicians to have the research programme scrapped after it was announced children as young as 11 could be recruited onto the trial. The Pathways clinical trial will be run by researchers at Kings College, London (KCL). In addition to setting a minimum age, they have also increased the safeguards for participants. The puberty blockers research was recommended by Dr Cass after her 2024 review of gender medicine for children pointed to weak evidence behind their use. Speaking to the BBC, Dr Cass said she believes since then "some of the hype about risks have been exaggerated in that we genuinely don't know if there are harms." And she said the trial was "essential" to answer the question about "whether these drugs are helpful or not". She added that young people will be "closely monitored in every respect" and the drugs stopped if concerns emerge. The researchers will examine the impact of the drugs on the physical, social and emotional wellbeing of participants. This will include checks on bone density, brain function and fertility. Cass believes without a trial young people will continue to get drugs from "unregulated and dangerous routes." Read full story Source: BBC News, 22 June 2026

Oversight of advanced AI systems capable of making autonomous decisions should “mirror” the assessment of healthcare professionals, a government commission has proposed. The National Commission into the Regulation of AI in Healthcare has proposed that agentic AI systems, which can autonomously plan and execute tasks with limited human supervision, should be required to demonstrate capability over time before being allowed to undertake more complex work. The minutes to the commission’s latest meeting, seen by HSJ, stated: “Commissioners advised that approaches to deploying AI systems should mirror that of human professional style progression.” This would involve AI agents needing “to demonstrate capability over time before being exposed to higher risk activities”. The commissioners were responding to a discussion paper on agentic AI systems, “which explored regulatory approaches to oversee AI systems that are capable of autonomously planning and taking actions with limited human supervision”. The paper proposed “a tiered regulatory framework, which uses levels of agent autonomy as a basis to determine what regulation and risk controls are required”. The commissioners “welcomed the proposal for a tiered regulatory framework”, but suggested, “further work should be undertaken to identify other potential factors relevant to determine the appropriate level of regulation”. Read full story (paywalled) Source: HSJ, 22 June 2026

Serious concerns have been raised that the delayed NHS “quality strategy” does not “prioritise patient safety”, HSJ has discovered. The government’s 2025 10-Year Health Plan stated “we will revitalise the National Quality Board (NQB) and task it with developing a new quality strategy”. The plan said the strategy would be published by March 2026, but this goal was missed, as was a second scheduled publication date soon after the May local elections. Minutes from the meeting obtained by HSJ reveal that NQB members “raised concerns” about the strategy’s lack of focus on patient safety and mental health. They also expressed a desire for the strategy to set “clearer expectations for providers”. Read full article (paywalled). Source: Health Service Journal, 23 June 2026 Related reading In this blog, Patient Safety Learning and the Advancing Quality Alliance (Aqua) set out the need for safety to serve as a golden thread woven throughout the Strategy.

The Ebola outbreak in eastern Congo has now surpassed 1,000 confirmed cases, with officials reporting 254 deaths as of Sunday evening. Congo’s Ministry of Health confirmed 1,003 cases and 100 recoveries since the epidemic was declared on 15 May in Ituri province. Caused by the rare Bundibugyo virus, for which no vaccines or treatments exist, this outbreak was the worst ever in its initial month. Officials admit more cases are likely unknown, and the peak is still ahead. Contact tracing remains a key issue, with local authorities achieving only 55 per cent coverage. The outbreak’s patient zero is yet to be identified, and over 35,000 contacts still require tracing, authorities confirmed. Read full article. Source: The Independent, 22 June 2026

The health minister has once again apologised for what he described as the "evil" perpetrated at Muckamore Abbey Hospital in County Antrim. Speaking in the assembly, Mike Nesbitt said what happened was a " true scandal". On Thursday, a long-awaited report into abuse at the hospital said a number of patients suffered physical abuse, including black eyes, broken bones, bruising and excessive restraint. Nesbitt said the weight of evidence had provided a "watershed" moment for the treatment and care of the most vulnerable in society. The Police Service of Northern Ireland has said its Muckamore investigation is the biggest criminal adult safeguarding case of its kind in the UK. In the assembly on Monday, Nesbitt said the report "helps us understand the failings of the past, and provides a road map for the work needed to address those issues". But, he said, it was "vital that we now move forward as a health and social care system, and importantly as a society, into a safer, more inclusive and accepting future for those most vulnerable in our society". Read full article. Source: BBC News, 22 July 2026

Thanks @Jules I've used AI to summarise this. Will watch the full presnetation too! “Can improvement and innovation save the NHS?” by Professor Mary Dixon-Woods Below is an AI generated concise summary report of the video “Can improvement and innovation save the NHS?”, a keynote by Professor Mary Dixon-Woods published by THIS Institute in May 2026. The lecture argues that improvement and innovation can help the NHS, but only when they are evidence-based, realistically implemented, attentive to inequality, and supported by well-functioning organisations rather than treated as universal solutions in themselves.[1][2] Executive summary The lecture presents a sober assessment of current NHS performance across access, timeliness, quality, effectiveness, and equity, using examples such as elective delays, cancer treatment delays, unwarranted variation in diabetes and breast cancer care, and persistent inequities in maternity outcomes. Professor Dixon-Woods argues that these problems are not simply deficits of effort or goodwill, but symptoms of deeper organisational, policy, and system design failures that limit the impact of improvement work.[2] Her central message is that innovation and improvement are necessary but insufficient unless they are grounded in evidence, matched to context, and protected from hype, overclaiming, and poorly designed large-scale programmes. She cautions that the NHS has often adopted interventions with excessive optimism, weak evaluation, and inadequate attention to implementation, creating cycles of enthusiasm followed by disappointment.[2] Main arguments The lecture identifies several core challenges facing the NHS: care is not consistently accessible, timely, high quality, effective, or equitable, and these deficits vary substantially by geography, deprivation, ethnicity, and sex. Examples cited include falling public satisfaction, persistent elective backlogs, non-compliance with guidance in some diagnostic testing, and marked disparities in maternal mortality and severe morbidity.[2] A major theme is that variation should not be dismissed as inevitable background noise, because it often indicates remediable organisational weakness, uneven capability, or failure to apply existing knowledge reliably. The lecture also highlights the continued use of some low-value activity alongside failures to deliver proven beneficial care, showing that both underuse and overuse coexist in the NHS.[2] Improvement lessons Professor Dixon-Woods argues that improvement succeeds least when it is treated as a slogan, a centrally imposed programme, or an assumption that any change is inherently beneficial. She emphasizes that large-scale initiatives often fail when they are oversold, under-specified, weakly evaluated, and inattentive to frontline realities, staffing pressures, and competing operational demands.[2] The lecture supports a more disciplined model of improvement: test interventions properly, understand mechanisms, use robust evidence, and distinguish genuinely effective innovation from attractive but weakly evidenced ideas. In practice, this means improvement should be designed as serious applied inquiry rather than as advocacy, branding, or policy theatre.[2] Governance implications For board and governance audiences, the lecture implies that oversight should focus not only on performance outcomes but on the organisational conditions that make safe and effective improvement possible. These conditions include the ability to identify risk early, hear uncomfortable information, respond to variation, evaluate change honestly, and sustain attention on inequity as well as aggregate performance.[2] The talk is particularly relevant to patient safety governance because it links poor outcomes to structural and cultural issues rather than isolated individual failings. It therefore supports governance approaches that emphasise system surveillance, speaking up, learning capability, and critical scrutiny of improvement claims before scale-up.[2] Actions for leaders A practical reading of the lecture suggests five priorities for NHS leaders and boards: · Treat major improvement claims as propositions requiring evidence, not as self-validating solutions.[2] · Target unwarranted variation as a governance signal of uneven quality and possible safety risk.[2] · Examine inequity explicitly, especially where deprivation, ethnicity, sex, or geography are linked to worse outcomes.[2] · Avoid adopting innovations at scale without credible implementation planning and evaluation.[2] · Strengthen organisational conditions for learning, challenge, and candour so that weak signals are detected earlier.[2] An example of the lecture’s practical relevance is its treatment of maternity inequity: disparities in mortality and morbidity are presented not as unfortunate externalities but as evidence that service design and care delivery are failing some groups more than others. That framing is directly applicable to board assurance, quality committees, and patient safety improvement programmes.[2] Would you like this converted into a more formal board paper style with headings such as background, key issues, implications, and recommendations? 1. https://www.youtube.com/watch?v=E_iCWIazGtU 2. https://support.google.com/youtube/answer/15930243?hl=en-GB

NHS England is being warned that the planned rollout of a new “portal” for all NHS primary dental work could lead to widespread disruption. The NHS Dental Services Portal is proposed as a new digital system for managing all NHS dental contract administration, including how dental activity is recorded, validated and paid. It is being rolled out to modernise an old, fragmented process, with the aim of improving efficiency, transparency, and consistency. In an open letter sent to NHS Business Services Authority and NHSE, and shared with HSJ, the Dental Software Suppliers Association raised concerns about the speed of implementation being imposed. Read full article (paywalled). Source: Health Service Journal, 22 June 2026

NHS England has taken enforcement action against a major health trust over multiple safety concerns, warning that it cannot be sure more patients won’t be harmed. The sanction means Northern Care Alliance NHS Foundation Trust, in Greater Manchester, could be fined or lose its license to provide care if it does not improve. It comes after a string of serious concerns were raised about patient safety, including in its gynaecological services, after an audit of hundreds of cases at Salford Royal Hospital in 2024 found dozens of women, including cancer patients, were “harmed” after their diagnosis and treatment were delayed due to admin failures. Now, a damning document, seen by The Independent, reveals NHS England found the trust has been “unable to provide assurance” that it has a clear and consistent structure “that will ensure no further patients may suffer harm”. Read full article. Source: The Independent, 19 June 2026

The report of the inquiry into the biggest maternity scandal in NHS history will outline “horrendous” failings in the care provided to women in Nottingham. A catalogue of appalling behaviour over many years by staff at the city’s two hospitals – Queen’s Medical Centre and Nottingham City hospital – included racism towards mothers, it will say. The NHS is bracing itself for the publication on Wednesday of a report by Donna Ockenden on 2,500 cases involving babies and mothers dying or being injured, and babies being stillborn, while under the care of Nottingham university hospitals NHS trust between 1 April 2012 and 31 May 2025. The document will stretch to more than 350 pages. Ockenden, a senior midwife and expert in maternity care failings, began her inquiry into Nottingham more than four years ago, in May 2022. About 2,505 families – more than in any previous maternity scandal – and approximately 850 staff and ex-staff of the NHS trust have given evidence to it. Read full article. Source: The Guardian, 22 June 2026

A highly recommended watch for anyone interested in leadership, patient safety, quality improvement and implementation. That is, if you haven'y yet come across it. Mary Dixon-Woods' lecture Can improvement and innovation save the NHS? is both thought-provoking and refreshing. It provides an opportunity to pause and reflect on where we are in our improvement journey. Among the many nuggets is a reflection on the "priority thickets" described in recent NHS literature, and the reality that health systems are often trying to improve everything, everywhere, all at once. One observation particularly resonated with me. Reflecting on decades of inquiries and reports into healthcare failings, she notes that sometimes the only thing that changes between one report and the next is the font. A sobering thought. The good news is that this lecture is not simply a critique. It also points towards practical ways forward, supported by evidence, resources and examples. An hour very well spent. Watch the video here: https://www.thisinstitute.cam.ac.uk/blog/can-improvement-and-innovation-save-the-nhs/

thank you Urmila, so please that I'm not an outlier when one looks at the evidence - there is no clear guidance - but we should be doing all we can - as you allude to