Search the hub

The Patient Safety Incident Response Framework became mandatory for all health services contracted under the NHS Standard Contract, including NHS-funded care delivered by independent healthcare providers, in April 2024. It replaced the Serious Incident (SI) framework. The change in approach to investigations under PSIRF has resulted in some practical challenges to the way in which information and organisational learning evidence is presented to the coroner for inquests. Chaired by Amelia Newbold, Risk Management Lead, this Shared Insights session discussed how the PSIRF and coronial processes can work more effectively together to ensure that coroners receive the information they need for inquests while preserving PSIRF's core principle of fostering a learning culture within healthcare. Bringing together perspectives from across the system, we explored some of the key challenges and, importantly, shared positive and practical examples of how a collaborative approach across both learning and coronial processes can ensure that relevant information is shared effectively.

On 30 July 2025 an investigation was commenced into the death of Pamela Ann Honeybone, who died at Scarborough General Hospital on 19 October 2024 aged 90. The investigation concluded at the end of the inquest on 23 September 2025.  The conclusion of the inquest was that: Pamela Ann Honeybone died as a consequence of naturally occurring disease. Diagnosis of her condition was delayed when another patient was scanned in error instead of Mrs Honeybone, but it has not been possible to determine on the balance of probabilities that this contributed to her death.  On the 19 of September 2024 Pamela Ann Honeybone was admitted to Scarborough General Hospital following a fall. She required CT scanning but another patient with the same first name underwent the investigation in error and its results were attributed to Mrs Honeybone. Mrs Honeybone’s condition continued to deteriorate and a CT scan undertaken on the 15 of October 2024 revealed the presence of an abdominal mass suggestive of lymphoma. Mrs Honeybone was moved to end of life care and she died at the hospital on the 19 of October 2024. Matters of concern: It was accepted in evidence that neither the doctor who escorted the wrong patient from the Emergency Department to radiology, nor the radiographer who undertook the CT scan on her, checked the identity of the patient in question. No transfer checklist was completed, and the patient was not asked to complete and/or sign the CT scanning questionnaire herself. No member of staff inquired as to the outcome of this patient’s CT scan prior to her discharge a few hours later. The scanning error was recognised by a radiologist on the 15th of October 2024, but was not conveyed to Mrs Honeybone’s treating team until late October, by which time she had died and her death had been scrutinised by the Medical Examiner and certified by her treating doctor as wholly natural and not requiring referral to the Coroner. As a result of the aforementioned delay, a Trust investigation did not commence until late November 2024. No prompt after action review therefore occurred in the hours and days after the error was recognised. When the Trust investigation did commence, staff directly involved either could not be identified or had no recollection of events. Despite hearing evidence that it was a doctor who would have escorted the wrong patient to scanning, the Trust Investigation focussed on nursing involvement with the patients in question and did not seek to identify and question medical team members. An Action Plan was drawn up as a result of the Trust Investigation, but for various reasons no audit of compliance with patient identification processes commenced until early August 2025, some ten months after Mrs Honeybone’s death. The results of the audit thus far were made available to me at inquest and indicate that 1 in 5 audited treatment encounters between staff of all grades and specialisms still occur without the patient being positively identified. The coroner heard evidence that while radiology transfer checklists are routinely completed ‘in hours’ at Scarborough Hospital when a dedicated HCA is on duty to perform this task, no such checklist is in use at the Trust’s York site at any time of the day. Mrs Honeybone’s misidentification occurred ‘out of hours’ at Scarborough when no designated person assumes responsibility for this task at that site. The coroner considers the above represent a continuing risk to others from misidentification and delayed responses to identified errors, with clear implications for patient safety.

A witness statement provides the Coroner with important evidence to consider as part of their investigation. It is important to ensure that all of the relevant information is included in your statement and that the statement is clear, thorough, truthful and accurate. This guide from the law firm Browne Jacobson gives you advice on writing the statement for an inquest.

Coroners have an important patient safety role under Regulation 28 of the Coroner’s (Investigations) Regulations 2013. This creates a statutory duty for Coroners not just to decide how somebody came by their death but also, where appropriate, to report about that death with a view to preventing future deaths (PFD report). In certain cases you may wish to provide the Coroner with evidence to explain the outcome of any internal investigation and provide assurance that organisational learning has been, or is being, implemented. This guide from the law firm Browne Jacobson has been produced to assist with the preparation of that evidence, and supplements their previous 'inquest guide for witnesses' and 'guide to writing statements for an inquest'.

A guide from Browne Jacobson, a law firm, to support staff involved in a coroners' inquest. It covers: When does the Coroner hold an inquest? What is an inquest? The inquest hearing Court day checklist Giving oral evidence Giving evidence remotely Inquest conclusions Regulation 28 / Preventing Future Deaths Further guidance and resources

Stef Cormack, Patient Safety Specialist at Sandwell and West Birmingham Hospital NHS Trust, recently presented at the Patient Safety Management Network meeting on the work she has been doing on the Patient Safety Incident Response Framework (PSIRF) and with coroners. Attached is the template the Trust uses to supplement the evidence the Trust provides to the Coroner for the purposes of the Inquest investigation, and to summarise the investigations, findings, conclusions, learning points and actions, which form the Trust’s response to the relevant patient safety incident(s) under PSIRF. Steph is happy for other organisations to adapt and use. Patient Safety Management Network You can apply to join the Patient Safety Management Network by signing up to the hub today. When you complete the registration form you’ll see a section called ‘Join a private group’, please tick the box by the relevant Network. If you are already a member of the hub, please email [email protected].

A coronial investigation into the death of Roman Louie Barr, aged 22 who died on 14  December 2023, was opened on 20 June 2024 and concluded on 3 March 2026. The inquest was conducted without a jury. The conclusion reached was a short factual  narrative: “The deceased died as a result of an asthma attack. Information indicating the need for an  urgent ambulance response was not obtained, and because no ambulance was available for several hours, he was taken to hospital by his family. On the balance of probabilities, earlier intervention by an emergency ambulance would have prevented his death.”  On 14 December 2023, Roman Louie Barr suffered an asthma attack. His father collected him from work and took him home, where Roman used his nebuliser without improvement. Three calls were made to the ambulance service. During these calls, Roman was assessed as Category 2, and the family were twice advised that no ambulance would be available for several hours. They were asked whether they could transport him to hospital themselves and took the decision to do so. Evidence established that at the time of the first call, Roman was critically unwell, displaying symptoms including bluish lips, but this information was not elicited during triage. Roman was of mixed ethnicity and had a darker skin tone, as his father explained to the call handler. The NHS Pathways question requiring confirmation that the patient was “a deathly colour” was not understood by his father. Clearer prompts—such as asking whether the lips were blue or grey—were not asked. A recommendation made during the subsequent review to amend this NHS Pathways wording was not accepted by those responsible for the system’s content. Ambulance availability was severely constrained due to significant delays in hospital handovers, leaving no crews free to respond. On the balance of probabilities, had clearer wording been used and the relevant information obtained, Roman would have been categorised as Category 1, for which an ambulance would be expected to arrive within approximately ten minutes even during surge conditions. While being driven to hospital, Roman suffered a cardiac arrest. His mother moved into the footwell of the passenger side and commenced CPR as they continued their journey. On arrival at the hospital, the family vehicle was involved in a collision, during which Roman’s mother sustained serious injuries. Roman could not be resuscitated and died shortly after arrival. I also heard evidence that Roman had been using his blue (salbutamol) inhaler more frequently than recommended, indicating poor asthma control, and that neither he nor his family were aware of the clinical significance of this increased use. Following his death, the GP practice conducted a review and introduced measures to better identify and monitor patients with high salbutamol use, including keeping a list of such patients, automatically booking reviews when further inhalers are requested, liaising with community pharmacists, and placing alerts on patient records to support timely assessment. Notwithstanding the Drug Safety Update issued on 25 April 2025 reminding clinicians of the risks associated with increased salbutamol use, the evidence in this case indicates that the importance of excessive reliever use may still not be fully recognised by patients or by primary care. Matters of concern Limited awareness of salbutamol overuse Evidence showed that patients and families may not appreciate the clinical significance of increased use of the blue (salbutamol) inhaler or its association with poorly controlled asthma. Identification and follow-up of reliever overuse Evidence showed that excessive or repeated requests for salbutamol inhalers may not be reliably identified within existing systems, and there may be no consistent process for follow-up when such patterns occur, meaning deteriorating asthma may go unrecognised. Ambulance handover delays affecting emergency availability Prolonged ambulance handover times at local hospitals were a significant factor in no ambulance being available at the time help was sought, reducing emergency response capacity during periods of high demand. Risks when families transport critically unwell patients The absence of an available ambulance for several hours resulted in the family transporting Roman to hospital themselves, exposing both him and his family to significant risk during a time-critical medical emergency. Clarity of NHS Pathways triage wording Evidence showed that a key NHS Pathways question used during triage was not understood by the caller and did not elicit clinically significant information. This raises a concern that, given the reliance on scripted triage systems, such scripts may not always use wording that is easily understood by lay callers in distress

Aarav died from the consequences of a cardiac arrest caused by severe bleeding following damage to an intercostal artery during a liver biopsy which went undiagnosed and untreated at the time of the procedure. His death was contributed to by poor planning before the procedure when there was no consideration of stopping antiplatelet medication, poor written and oral communication about the complication that occurred during the procedure all of which hampered treatment after his collapse. His death was contributed to by neglect.  MATTERS OF CONCERN Prophylactic antibiotics for severely immunocompromised patients: The inquest heard evidence that patients like Aarav who are immunocompromised require additional prophylactic antibiotics for procedures. This is not covered in the current NICE guidelines. The concern is that there is currently no guidance for the use of prophylactic antibiotics in severely immunocompromised patients. Experience and competence of trainees: The inquest heard evidence that there was confusion around the experience and level of the trainee involved. He was thought to be an ST6 when he was an ST4. The concern is that there is no mechanism to evidence trainees experience and competence when they travel to various different hospital trusts as part of their training. Consent forms: The parents of Aarav were unaware that a trainee would be doing the liver biopsy. The concern is that there is currently no way to obtain consent when a trainee will be doing the procedure. Individual patient risk factors: Aarav had a complex medical background and several risk factors for any procedure. The concern is that there is currently no mechanism to identify individual patient’s risk factors so that all clinicians involved in their care are aware. Learning from deaths: The initial M&M meeting after Aarav’s death was described as inadequate. The concern is that there was no immediate learning from this tragedy and further consideration is needed to ensure a safe and effective mechanism to properly learn from deaths at the earliest opportunity. Electronic patient records: Evidence that the lack of electric medical records meant clinicians found it difficult to see all of the patient’s medication details. The concern is that critical information can be missed if clinicians do not have access to all the clinical records when planning treatment.

On 25 February 2026, healthcare leaders and stakeholders gathered in London for the Patient Safety Forum, organised by Public Policy Projects (PPP) in partnership with Patient Safety Learning. Panellists at this event investigated multifaceted impacts patient safety measures are having on wider system operations. From avoiding duplicated processes and failure demand, to cross-disciplinary learning and efficiency gains, this blog summarises discussions from two panels which highlighted the Patient Safety Incident Response Framework and productivity as focuses for a reforming healthcare system. Read the full article from PPP via the link below.

Terrence Frost’s death was verified at 00:26 on 18 July 2024, at the Ipswich Hospital, in Ipswich, Suffolk, although Terrence’s death had occurred earlier at approximately 22:20 on 17th July 2024. On the 11 July 2024 Terrence was admitted to the Ipswich Hospital for an elective surgery (angioplasty) to improve the blood flow to his left leg and foot. Terrence was discharged on the following day 12 July 2024. On the 14 July 2024 Terrence was admitted again to the Ipswich Hospital with abdominal pain and rectal bleeding. No diagnosis was made, and as this settled spontaneously, Terrence was discharged again on the 15 July 2024. On the 16 July 2024, due to concerns raised by his family, a GP’s Paramedic conducted a home visit, and following subsequent concerning blood test results Terrence was told to go back to Ipswich Hospital as a failed discharge. After a prolonged period in the Accident and Emergency department Terrence was readmitted to the Ipswich Hospital. Despite testing, no definitive diagnosis was made during Terrence’s final admission, and Terrence appeared reasonably stable until he suffered a sudden collapse and cardiac arrest at 21:22 on the 17th July 2024. A subsequent postmortem examination identified that Terrence suffered from significant cardiac disease (cardiomegaly and coronary artery disease) and significant vascular disease (systemic atherosclerosis). The pathologist identified that his clinical markers identified that sepsis played a factor in Terrence’s death, although evidence of any infection could not be found. MATTERS OF CONCERN Evidence was heard that prior to his attendance in the Accident and Emergency department on the 16 July 2024, Terrence had been seen at home by a paramedic from his surgery, who was concerned by Terrence’s presentation and wanted to admit him to hospital. However, Terrence was reluctant so it was agreed that urgent blood tests would be taken in the first instance. The results of these tests were seen by a GP, and due to the findings (which indicated a possible serious infection or inflammation) the GP called Terrence and told him to go straight to hospital, and whilst enroute she would speak to the Medical Assessment Unit. In evidence the GP said she then spent 30 minutes on the telephone trying to contact the Medical Assessment Unit as is the required procedure, to discuss Terrence’s admission. After being unable to contact the Medical Assessment Unit, the GP contacted Terrence, via a family member, and told him that as she could not contact the Medical Assessment Unit he should head to the Accident and Emergency department instead. The GP told Terrence she would pre- alert the Accident and Emergency department to his arrival. The GP then spent a further period of time telephoning the Accident and Emergency department but again could not get through. As such upon arrival, a patient who was considered by their GP to be significantly unwell enough to warrant either admission to the Medical Assessment Unit, or that Accident and Emergency should be pre-alerted to their arrival, was unable to speak to either unit prior to the patient’s arrival. Terrence endured a 5 hour wait in Accident and Emergency before being seen. Although observations taken at the time of his subsequent admission suggest he had not developed sepsis at this stage, I am concerned that the inability of a GP to be able to promptly communicate with either the Medical Assessment Unit or Accident and Emergency department may lead to future deaths in cases where suspected sepsis or other life threatening conditions have been differentially diagnosed, especially if those conditions have progressed further than Terrence’s had at the time of his arrival. I am further concerned that evidence was heard from a clinician based at the Ipswich Hospital itself, that they too found contacting the Medical Assessment Unit extremely difficult, with internal hospital telephone calls frequently going unanswered.

NHS England recently met with the Chief Coroner to explore how the Patient Safety Incident Response Framework (PSIRF) and coronial processes can work more effectively together. The discussion focused on a shared goal: ensuring coroners receive the information they need for inquests while preserving PSIRF’s core principle of fostering a learning culture within healthcare. A newsletter has been published following the meeting which has been copied below. Background on PSIRF Following more than four years of development, testing and evaluation, PSIRF was introduced to the NHS in England in August 2022, and became mandatory within the NHS Standard Contract from April 2024. PSIRF replaced the Serious Incident Framework (SIF), addressing weaknesses in incident response highlighted by patients and families, health professionals, regulators, academics and Parliament. PSIRF moves away from a single, linear method - Root Cause Analysis (RCA) -towards a systems-based approach, widely regarded across safety-critical industries as best practice for learning and improvement. RCA often oversimplified complex events and failed to identify interacting systemic factors, leading to narrow and ineffective actions. In contrast, PSIRF enables proportionate responses using varied evidence-based methods to generate learning, fostering openness and a culture of continuous improvement. While this approach strengthens the ability to learn from incidents, it has also introduced new challenges in how healthcare providers interact with coronial processes. Current challenge The challenge between PSIRF and inquests arises because the two processes serve fundamentally different purposes. PSIRF is designed to support organisational learning and improvement, and deliberately excludes activities such as apportioning blame, determining liability, assessing preventability, or identifying cause of death. PSIRF learning responses take a “window on the system” approach, exploring how work happens in everyday practice rather than focusing solely on a single incident. In contrast, coroners are legally required to answer four statutory questions, including how someone came by their death. This often involves establishing causation and examining the circumstances surrounding a specific death. This difference means PSIRF outputs, which focus on systemic insights rather than direct causation, may be less directly useful for coronial purposes. Some coroners, accustomed to Serious Incident investigation reports that provided clear chronologies and RCA now find that PSIRF outputs while richer in systemic insight are lacking the causation detail they expect. Action Both NHS England and the Chief Coroner agreed on the importance of continued collaboration to ensure that relevant information can be shared to support both processes. Because PSIRF and inquests serve different purposes, evidence gathering for a PSIRF learning response and for an inquest must remain distinct so that each achieves its intended aim. This means coroners may need causation to be established through other means and should no longer expect or require an RCA in place of a learning response, as this is no longer the nationally endorsed approach. Importantly, PSIRF outputs, including the rationale for the chosen response and any improvement actions, can provide valuable context about wider circumstances and system changes. Coroners may continue to use learning response outputs as supplementary information when available; however, these should not be relied upon as the primary or sole evidence for an inquest. By working together, both parties can uphold the integrity of the coronial process while fostering a culture of learning and improvement across healthcare.

NHS England recently met with the Chief Coroner to explore how the Patient Safety Incident Response Framework (PSIRF) and the coronial process can work more effectively together. There is currently a critical challenge with the two processes because they serve fundamentally different purposes. PSIRF supports organisational learning and deliberately avoids apportioning blame or identifying cause of death, whilst coroners must answer four statutory questions, including how someone came by their death.  A newsletter has been published following the meeting, confirming that coroners should no longer expect or require Root Cause Analysis (RCA) reports in place of PSIRF learning responses, as this is no longer the nationally endorsed approach. Evidence gathering for PSIRF and inquests must remain distinct, with causation potentially needing to be established through other means. This article from law firm Browne Jacobson summarises the new guidance. In their joint newsletter, both NHS England and the Chief Coroner agreed that evidence gathering for a PSIRF learning response and for an inquest must remain distinct so that each achieves its intended aim. This means coroners may need causation to be established through other means and should no longer expect or require an RCA in place of a learning response, as this is no longer the nationally endorsed approach. PSIRF outputs, including the rationale for the chosen response and any improvement actions, can provide valuable context about wider circumstances and system changes. Coroners may continue to use learning response outputs as supplementary information when available; however, these should not be relied upon as the primary or sole evidence for an inquest. Takeaways for NHS trusts and other healthcare providers: Plan for separate causation evidence for the inquest. Establish early coordination between legal and patient safety teams. Use PSIRF outputs strategically. Maintain the integrity of both processes. Challenge coroners requesting RCA reports.

On 14 November 2024, an investigation commenced into the death of Master Avery Jake Hall, who died in Sunderland on 13 November 2024 aged 4 days. The Investigation concluded at the end of the Inquest on 23rd January 2026.  The medical cause of death was confirmed as: Ia Hypoxia Ischaemia and diffuse alveolar damage  Ib Olygo/anhydramnios and foetal distress  Ic Premature rupture of membranes, small placenta with distal villous maldevelopment and low grade foetal vascular malperfusion. Avery Jake Hall died at Sunderland Royal Hospital on 13th November 2024 having developed global hypoxia and diffuse alveolar damage with hyaline membranes in the lung following his birth as his development in pregnancy had been compromised by reduced amniotic fluid leading to poor lung development and impairment of urine production by the kidneys. During pregnancy Avery’s mother had continued to take Candesartan which had previously been prescribed to her to treat migraines. She did not receive definitive advice from clinicians to stop taking it despite various opportunities to do so and this is a medication contraindicated in pregnancy due to risks including foetal renal failure and pulmonary hypoplasia. Matters of concern Avery’s mother suffered from migraines which were increasing in severity, so she sought advice from her GP when aged 21 years old. She was prescribed Candesartan 4mg by her GP shortly before her 22nd birthday. This was to be taken daily and was placed on a repeat prescription of 28 tablets. The dose was increased to 8mg after 3 months and following a referral, the treatment was endorsed by a Consultant Neurologist at a consultation 4 months after the initial prescription. The evidence revealed that no advice was provided as to the risks of this medication should she be considering having a child. Following a positive pregnancy test in April 2024, Avery’s mother sought advice from her GP about which of her prescribed medications were safe to use during pregnancy. During the telephone consultation with her GP on 11 April 2024 she was given specific advice to avoid using 3 of 6 prescriptions. However the evidence highlighted that Avery’s mother was given only generic advice that it was best to avoid all medication during pregnancy but was not given specific advice to stop using Candesartan, and the risk of continuing to take this medication in pregnancy was not identified during this consultation. Although Avery’s mother had a number of attendances with clinicians throughout her antenatal care, the evidence revealed that she was given no additional advice regarding the safety of her medication and, whilst she was advised to seek advice from her GP as the prescriber, she did not feel it was necessary to do so having already had such a consultation in April 2024. Avery’s mother continued to suffer from migraines during her pregnancy and was unaware of the risk posed by taking Candesartan in pregnancy due to a lack of clear and definitive advice about the risk. I am concerned that she was able to resume taking Candesartan approximately 14 days after her initial GP consultation as the medication remained on a repeat prescription which she was able to continue to request during her pregnancy, and each request was approved without a detailed review. The last repeat prescription being approved only 12 days prior to Avery’s birth. The coroner's concern is that despite advice from the GP that it was best to stop all medication during pregnancy, Candesartan remained as a repeat prescription and, in addition to that, there were no warnings placed on the system which would have alerted the clinician approving the request for the repeat prescription that the patient was pregnant thus necessitating a review.

In 2014 an investigation was commenced into the death of Yousef Al-Kharboush (born 23 May 2014, died 1 June 2014, aged 8 days), Oscar Barker (born 27 May 2014, died 29 June 2014, aged 1 Month) and Aviva Otte (born 10 October 2013, died 2 January 2014, aged 2 months). The investigation concluded at the end of the inquest on 23 October 2023.  Aviva’s death (January 2014) was in hospital where she had received TPN provided and compounded by the NHS establishment under a section 10 exemption. That TPN had, on balance, been contaminated by Bacillus cereus (subsequently identified as type BC.38). The Trust undertook a root cause analysis together with involving the UKHSA and its own infection and microbiological teams, but no definitive source for the outbreak was found. In June 2014 Oscar Barker and Yousef Al-Kharboush received TPN, compounded by a commercial provider, which it turned out was also contaminated by Bacillus cereus (subsequently typed as Bc.44). The compounder having positive finger dab testing for the Bacillus within its laboratory/environmental testing. This outbreak also affected other babies in other Trusts. Bacillus cereus is resistant (because it is spore forming) to the spray and wipe cleaning methods used (with alcohol) and sporocides are required to decontaminate the outside of, for example, ampoules containing one of the constituents. This was the information and a conclusion that the Trust had reached in early 2014 and therefore prior to the outbreak in May/June 2014. It had not passed on those findings either within other section 10 units compounding TPN or the wider market. Subsequently, the MHRA brought in further advice for the use of sporocides in 2015. Matters of concern There is no requirement for a section 10 exempt entity to report any of its findings to the MHRA or indeed to other Trusts or the industry in general if an adverse event occurs. The current reporting structures (for a section 10 entity) involve reporting to NHSE and the CQC but the threshold or necessity for such reporting appears unclear and, in essence, up to the Trust. There may be times when section 10 entities reach conclusions which would assist the wider industry and help to assist both other Trusts and commercial organisations in assessing their own risks and improving the provision of highly specific medication to a group of vulnerable patients. The same may also be true of commercial organisations but they have the power of the MHRA controlling and effecting recalls and actions and the wider dissemination of information. Response from NHS England

Matilda Gwen Pomfret-Thomas was born on 29 October 2023 at Queen Alexandra Hospital following a difficult labour at home. Hypoxic ischaemic encephalopathy had developed over a period of hours. Meconium had been observed, decelerations were later observed. On 15 November 2023 an investigation into the death of Matilda Gwen Pomfret-Thomas aged 15 days commenced. The investigation concluded at the end of the inquest on 4 December 2025 and the medical cause of death was hypoxic ischaemic encephalopathy. The birth of the family’s first child had been traumatic and, for the birth of their second child, Matilda, they were focussed on achieving a different birth experience and elected to use a doula to provide them with support at a home birth. The hospital’s preference was for a hospital delivery, there was discussion as to what circumstances would result in the mother being blue lighted to hospital. Signs of fetal distress developed but the mother was not immediately transferred to hospital. A difficult atmosphere had developed, the midwives felt access was being restricted by the doula: the coroner found that she did not actively discourage midwife access but that she was seen as, in effect, a buffer by members of the midwifery team. The doula was following the birth plan. The doula was supporting the parents per the birth plan, and this appears to have been perceived as grounds for hope that a home birth was still possible. Matters of concern Doulas provide continuity of care and give emotional, informational and practical support throughout pregnancy, labour and after the birth of a baby: those words come from Doula UK’s website. Doula UK is the largest representative body for Doulas, but it is not a regulatory body, it does not represent all doulas, indeed many doulas are not members of Doula UK. Doula UK have put in place membership requirements, training offers and much guidance, but the role of a doula is clearly diffuse in practical terms and capable of multiple understandings not just by doulas but their clients and midwives. It appears that doulas have been increasingly used and increasingly offer services – as here – on a paid basis. As MNSI (Maternity & Newborn Safety Investigations – formerly HSIB) put it in their report into this birth, “MNSI acknowledges that there is no regulation of doula care or any guidance on how the two services interact with each other. MNSI considers the dynamics of a situation, where a third party are involved can provide additional challenges for staff, such as making clinical recommendations against personal recommendations or views and providing usual care that could be viewed as interference rather than surveillance.” MNSI have identified 12 cases in which there was evidence that doulas worked outside of the defined boundaries of their role and in which the care or advice provided by the doula was considered to have potentially had an influence on the poor outcome for the family. There was evidence given at the inquest by experienced midwifery professionals highlighting that provision of guidance would be helpful for all involved with a birth at which a doula was present. The issues of doula registration, regulation and training are therefore points of concern the coroner would commend for review.

In 2023, Jennifer Cahill was pregnant with her second child. Her antenatal care was managed by Manchester Foundation Trust (“MFT”) community midwives. In 2021 her first pregnancy had resulted in complications at the time of delivery. She had a Post Partum Haemorrhage for which she received an iron and also a blood transfusion. She was also positive for Group B Streptococcal. On the 26 June 2024 an investigation into the deaths of Jennifer and Agnes Cahill was carried out. The Inquests concluded on the 27 October 2025. The conclusion of the Inquests was: Jennifer Rose Cahill died as a result of complications arising from the delivery of her second child, contributed to by neglect. Agnes Lily Wren Cahill died as a result of complications during birth, such complications contributed to by neglect. The medical causes of death were recorded as:  Jen: 1a) Multiorgan failure with disseminated intravascular coagulation 1b) Cardiac arrest due to post-partum haemorrhage 1c) Perineal tear and atony during term delivery. Agnes: 1a Multi-organ insult following hypoxic ischaemic encephalopathy 1b. Cord compression and meconium aspiration syndrome leading to pulmonary hypertension. Key findings Jen had not made an informed decision to have a home birth and if the out of guidance plan had been completed and all the relevant information provided to her, it is more likely than not she would have given birth in an alternative setting and both Jen and Agnes would have survived. If the fetal heart rate monitoring had been conducted correctly and every 5 minutes, it was more likely than not an abnormal fetal heart rate would have been noted up to an hour before Agnes was born and an urgent transfer to hospital would have occurred. The coroner found emergency services would have been on scene when Agnes was born and effective resuscitation would have been administered which would likely have prolonged her life. Had this call been made it is more likely than not Jen would have survived as the after care delivered to her would have noted a perineal tear and administered syntemetrine immediately. The coroner heard evidence that since the deaths, MFT have completely overhauled the home birth service provision. The new service became operational in April 2025. In the six month period within the MFT area of GM they have received requests from 34 women for out of guidance home deliveries. Five of these could not be supported due to safety issues. Of the 29 out of guidance home births, 15 (50%) required transfer to hospital for varying degrees of obstetric emergency. Matter of concerns There is no national guidance in respect of home births. Specifically, robust evidenced based guidance on home birth care, similar to that which is in place for intrapartum care in a hospital setting. There is an increase in the number of women with ‘high risk pregnancies’ requesting home births where required interventions cannot take place or would be significantly delayed and there is no robust framework for midwives supporting home birth care. There is no national guidance to support consistent practice across the country including, for example, details of clinical scenarios where women, following robust assessment, have been considered too high risk to safely receive care in a home-setting. The lack of national guidance means there are differing models of care and unlike other specialities home births are not a specialist commissioned service. There is no national guidance considering the ethical responsibility and proportionality of offering a home birth model under the NHS framework. Even though there is a very small risk of death, this is not something which is discussed with women particularly in relation to maternal death, even if the woman has a recognised risk such as a post-partum haemorrhage. There is no guidance to ensure the risk of death to both mother and baby is discussed with any woman considering a home birth irrespective of being considered high or low risk. NICE guidance on intrapartum care (2023 updated June 2025) Section 1.3.3 only refers to the potential risk of death to a baby. There is no mention in the guidance of risk to the mother. Terminology around pregnancies describes them as ‘high’ or ‘low risk pregnancy’ and leads women to consider that pregnancy encompasses all stages through to delivery of a child. Practice does not personalise or individualise risk so women can fully understand what the level of risk is for them in actually being pregnant, or what the level of risk is for them in giving birth. In order to maintain their skills, there is no set number of deliveries a community midwife must conduct following qualification. There is no mandated number of deliveries that any midwife (irrespective of the settings in which they are working) must complete once they have qualified as a midwife in order to maintain their registration. The level of experience of community midwives in conducting deliveries is not information routinely provided to women to inform their decision whether to have a homebirth. No bespoke training needs analysis has been conducted focusing on midwives practicing in home birth teams. The lack of national data collection means there is no data to evidence the number of women who are transferred in during labour or after birth, maternal or neonatal outcomes, number of women who are considered out of guidance. The no national guidance on the model of staffing, training and experience for midwives providing home birth care. See also: NHS England's letter responding the Prevention of Future Deaths report.