Epilepsy, 19 Sep 2013 – 19 Jun 2026

Epilepsy-related Prevention of Future Deaths reports

All dates19 Sep 2013 – 7 Sep 2026

Custom range

Based on PFD publication date. Dates are available up to the published snapshot: 7 Sep 2026. Preset periods end on this date.

90All reports 19SUDEP 68Non-SUDEP

Select SUDEP or Non-SUDEP to filter concerns and reports. 3 reports could not be assigned and remain in All only. Counts are PFD reports, not deaths.

About this dataset

We searched all 6,433 reports in the published corpus using a fixed text-search method, then assessed 317 candidates against source-based inclusion criteria. Ninety reports were included. Epilepsy had to be materially relevant to the death or a substantive coroner’s concern; incidental mentions were excluded.

Eligibility and SUDEP assignment were assessed with GPT-5.6 Sol and checked for valid source quotations. Assignment uses a balance-of-probabilities threshold: SUDEP or non-SUDEP must be more likely than not on the report’s evidence. The exact term SUDEP need not appear. Genuinely unclear eligible reports remain in All only. These are model-assisted documentary assessments, not independently validated clinical diagnoses.

The search can miss reports with sparse terminology or incomplete source text. This dataset does not represent all epilepsy deaths. Recurring concerns and responses are taken from the existing published evidence. Published corpus snapshot: 7 September 2026. Cohort selection updated 5 November 2026. The title shows the first and last included reports’ publication dates.

Reports over time

Matching reports published each year within the selected date range.

* 2026 is projected from reports observed to 7 Sep 2026.

Most frequent recipients

Matching reports sent to each recipient.

Betsi Cadwaladr University LHB1
Doncaster and Bassetlaw Teaching Hospitals NHS Foundation Trust1
HM Prison and Probation Service1
Lewisham and Greenwich NHS Trust1
NHS England1
Practice Plus Group1
Royal College of General Practitioners1
Royal College of Physicians1
Thameside Prison1

Most commonly raised concerns

All classifications 90 reports 19 Sep 2013 – 7 Sep 2026

  1. Unreliable seizure care and managementControls dedicated to recognising, assessing, monitoring, responding to and managing seizures are not reliably provided.9 reportsSelected
  2. Incomplete, inaccurate or unavailable clinical and care recordsClinical and care records are incomplete, inaccurate, inconsistent, unavailable or otherwise unreliable for safe care.7 reportsSelected
  3. Unsafe medication administrationA recurring failure of the medication-administration system to ensure that medicines are administered safely, accurately, promptly and as prescribed.6 reportsSelected
  4. Failure to ensure clinical investigation results are reliably available, interpreted and acted uponRecurring deficiency in the end-to-end management of clinical investigation results, including their availability, transmission, checking, review, interpretation, follow-up and escalation when needed.4 reportsSelected
  5. Inadequate driver licensing controls for road safetyDriver licensing systems do not reliably identify, record, review or restrict drivers whose inexperience, health or vehicle-related risk requires additional safeguards.4 reportsSelected
  6. Inadequate staff competence to provide first aidReports identify an unsafe first-aid capability condition across healthcare, care, custody and other staff. The recurring concern is the unreliability of staff competence to recognise emergencies and provide appropriate first aid, rather than any single training, refresher, assessment or practical-performance failure.4 reportsSelected
  7. Unreliable hospital discharge processesHospital discharge does not reliably bring together planning, clinical readiness, relevant organisations, safe execution and required follow-up.4 reportsSelected
  8. Failure to communicate clinically important information reliably between care servicesRecurring concern that clinically important guidance, decisions and patient information are not kept accurate, current and available to the clinicians and emergency services who need them.3 reportsSelected
  9. Failure to provide required medication promptly when clinically neededThe anchor identifies delayed prescribing of a patient's usual medication after an unexpected admission, while another report identifies delayed prescribing and non-provision of urgent therapeutic medication. These support a recurring concern about timely provision of clinically required medication rather than only the anchor's admission or prescribing-step failure.3 reportsSelected
  10. Failure to provide timely clinical follow-up after medication prescribingPrescribed medication is not reliably followed up for effectiveness, safety, collection, adherence and patient response.3 reportsSelected
  11. Failure to seek medical attention when a person's condition warrants itReports identify an unreliable escalation control in which care staff or other responders do not promptly seek medical or paramedic assistance when a person's injury, pain, fall or deterioration warrants clinical attention.3 reportsSelected
  12. Insufficient police training for safe recognition and response to mental-health and behaviour-related medical crisesReports identify insufficient police training to recognise and respond safely to mental-health crises and behaviour-related medical emergencies.3 reportsSelected
  13. Insufficient qualified healthcare staffing capacityHealthcare services lack enough appropriately qualified staff to provide the required clinical coverage and patient care safely for the workload, acuity and specialist needs.3 reportsSelected
  14. Unreliable DVLA medical fitness notification arrangementsThe anchor identifies limited professional awareness and uncertain effectiveness of guidance for notifying the DVLA about medically relevant driving risks. A different report identifies clinicians failing to advise patients with cough syncope to stop driving and notify the DVLA, while another identifies failure to notify the DVLA of relevant medical conditions. These support the named DVLA medical-fitness notification process rather than guidance awareness alone.3 reportsSelected
  15. Unreliable communication of patient-care information between clinical staffClinical staff responsible for the same patient's care do not reliably exchange relevant information about care, treatment and risk.3 reportsSelected
  16. Unreliable healthcare patient transfer processesHealthcare transfers do not reliably provide the procedures, responsibility, documentation and clinical information needed for safe movement and continuity of care.3 reportsSelected
  17. Unreliable mental health referral pathwaysReports identify different failures within the mental health referral process, including referrals not being made when indicated, received referrals not being actioned, and frontline practitioners lacking clarity about when and where to refer. These support a recurring concern with the reliability of mental health referral pathways rather than any single referral or awareness failure.3 reportsSelected
  18. Unreliable patient observation arrangementsPatient observation systems do not reliably set, support and deliver the observation level required by assessed safety risk.3 reportsSelected
  19. Unsafe interoperability between prison custody and healthcare proceduresReports identify failures in the bounded interface between prison custody, contracted custody providers and healthcare services, where procedures, responsibilities and safety-relevant information do not operate together reliably.3 reportsSelected
  20. Electronic patient records failing to make relevant clinical information available and actionableRecurring concern that electronic patient-record systems do not reliably present, preserve, share or prompt action on clinically relevant information across the patient’s care.2 reportsSelected
  21. Failure to communicate safety-critical care information effectively between care providers and familiesReports identify unsafe communication between care providers and families, including failures to convey risks, care instructions, concerns or other information needed to support safe care and decision-making.2 reportsSelected
  22. Failure to conduct required welfare checks on people in distressThe anchor identifies a risk that care home staff will not check a resident in distress, while a different canonical report identifies failure to conduct a later welfare check on a person left in potential medical distress. Together these support a recurring concern about welfare checks not being completed when a person may require assistance.2 reportsSelected
  23. Failure to ensure safe prisoner transfersRecurring concern that the end-to-end prisoner transfer process is unreliable, leaving transferred prisoners exposed to unmanaged risks or interruptions in essential support.2 reportsSelected
  24. Failure to identify and address recurring safety issues through organisational learningDifferent reports identify failures of reporting, review and learning arrangements to retain, recognise and address substantive or recurring patient-safety issues, rather than merely failures to complete one specific corrective action.2 reportsSelected
  25. Failure to reliably document the rationale for consequential decisionsDifferent reports identify missing or inadequate records explaining the rationale and evidential basis for consequential decisions, including the anchor’s crucial decisions and comparable mental-capacity, NRPF and transfer decisions. The shared concern is unreliable rationale documentation for decisions that affect safety, accountability and continuity of care, rather than any one decision type or record system.2 reportsSelected
  26. Failure to reliably learn from medication incidents and implement safeguardsThe medication incident learning process is unreliable: incidents are not consistently followed up or reviewed to identify, communicate and implement safeguards that protect patients from recurrence.2 reportsSelected
  27. Failure to reliably recognise and respond to acute clinical deteriorationClinical deterioration is not reliably recognised, monitored, escalated and acted upon with the urgency it requires.2 reportsSelected
  28. Failure to reliably refer patients to required specialist servicesThe anchor identifies a required physiotherapy referral that was not completed. A different report identifies delayed referrals to physiotherapy, occupational therapy, dietetics and Parkinson’s specialist services. Together these support an end-to-end concern with required specialist referrals not being made promptly and reliably, rather than a concern limited to one physiotherapy referral.2 reportsSelected
  29. Failure to secure and control medicationReports identify medication as remaining accessible, unsecured or inadequately controlled, creating risks of misuse, overdose, theft or access by people other than the intended patient.2 reportsSelected
  30. Inadequate control of illicit substance use and supply in secure institutionsThe anchor identifies an ongoing failure to mitigate illicit substance entry, distribution and use in a secure hospital, while assertions from different reports identify continuing illicit-drug influx, widespread access and weak supply-reduction controls in prisons. These distinct manifestations support a bounded concern about controlling illicit substances in secure institutional settings, rather than only room searches, drug testing or a prison-specific strategy.2 reportsSelected
  31. Inadequate controls on contraband mobile phones in prisonsDifferent reports identify failures in the prison control system for preventing and mitigating access to contraband mobile phones, including their easy availability to prisoners and their use to facilitate drug dealing and other prison-security risks.2 reportsSelected
  32. Inadequate recording of medication prescribing decisionsReports identify unreliable medication prescribing records that can obscure prescribing status, rationale, changes or safety-critical decisions and thereby allow unsafe medication to be prescribed or administered.2 reportsSelected
  33. Incomplete clinical history-takingReports identify an unsafe clinical assessment condition in which relevant patient history is not obtained sufficiently to support recognition of symptoms, risks and appropriate care.2 reportsSelected
  34. Unreliable Care Programme Approach care coordinationReports identify different failures within the named Care Programme Approach, including the absence of a structured medication review and failures to allocate coordinators, complete required reviews, maintain care plans and follow up patients. Together these support a recurring concern with the reliable operation of CPA care coordination rather than medication review alone.2 reportsSelected
  35. Unreliable RIO clinical-record functionsThe anchor identifies an unclear timeliness standard for RIO entries, while different reports identify failures of RIO to support complete information capture, cross-team access and usable observation records. Together these support a recurring concern with the reliability of the named RIO clinical-record system rather than only one entry-timing or documentation manifestation.2 reportsSelected
  36. Unreliable access to emergency communicationReports identify failures in dedicated emergency communication arrangements that can delay summoning or contacting emergency help.2 reportsSelected
  37. Unreliable communication of critical medication information to GPsRecurring deficiency in the medication-information exchange process between services and GPs, including information needed to maintain supplies, assess prescribing safety and support continuity of treatment.2 reportsSelected
  38. Unreliable escalation by care staff for required medical attentionCare staff do not reliably recognise and escalate deterioration, pain or other safety concerns to obtain timely medical assessment or advice.2 reportsSelected
  39. Unreliable implementation of medication changesReports identify different failures in the medication-change process, including failure to act on a prescribed reduction, failure to ensure carers acknowledge changes, and difficulty implementing specialist-directed changes accurately. Together these support the wider unsafe condition that medication changes are not reliably implemented across prescribing, communication and follow-up controls.2 reportsSelected
  40. Unreliable inter-agency information sharing for coordinated careThe anchor identifies absent agreed communication means and inaccessible shared patient information between mental health agencies. Other canonical reports identify relevant risk information not being shared between agencies and poor inter-agency information sharing preventing coordinated care, supporting a recurring concern with the reliability of inter-agency care information sharing rather than only the anchor's repository or communication-channel failure.2 reportsSelected
  41. Unreliable management of medication doses not takenThe anchor identifies missing guidance for monitoring prescribed medication that self-administering patients do not take when dispensed. A different report identifies inconsistent procedures for responding when service users refuse or do not take medication. Together these support a recurring concern with the reliability of detecting, monitoring and escalating omitted or refused medication doses, rather than a concern limited to one guidance document or one home's procedure.2 reportsSelected
  42. Unreliable medication dosage verification and communicationReports identify unsafe failures to ensure that medication dosages are accurately verified, clarified and communicated across prescribing, dispensing and administration activities.2 reportsSelected
  43. Unreliable observation of people in custodyThe anchor report identifies failure to transfer and replicate medically required observation levels when a detained person moves into GEO AMEY custody, while a different report identifies uncertainty about whether the correct observation level, including constant watch, was provided to a person in prison custody. Together these support a recurring concern about the reliability of the custody observation process rather than only one handover or level-setting failure.2 reportsSelected
  44. Unreliable recording of required observations in care and custodyDifferent reports identify inaccurate, delayed, anticipatory or falsified records of required observations, including 15-minute observations in prison custody and patient-care settings. The shared unsafe condition is that observation records do not reliably reflect when and how observations were actually conducted, rather than any one observation frequency, recording location or electronic system.2 reportsSelected
  45. Unreliable review and action on clinically significant incoming correspondenceReports identify failures across the incoming clinical-correspondence process, including failure to identify GP medication-review requests, failure to bring action-requiring correspondence to medical staff, and correspondence being filed without GP review or action. The shared unsafe condition is that clinically significant correspondence may not reliably reach the responsible clinician or trigger the required clinical response.2 reportsSelected
  46. Unreliable safety CCTV monitoring systemsDifferent reports identify failures in the operation of CCTV used as a safety control, including absent staff access arrangements, inadequate coverage, non-working cameras, insufficient monitoring and unclear staff instructions. Together these support the reliability of safety CCTV monitoring as the parent concern rather than any single access, coverage, equipment or training failure.2 reportsSelected
  47. Unreliable urgent mental health referral and assessment pathwaysUrgent mental-health referral pathways do not reliably secure timely access, coordination and specialist assessment.2 reportsSelected
  48. Unreliable welfare-check processes for people whose health is of concernThe anchor identifies unclear purpose and use of the welfare sheet for monitoring prisoners whose health is of concern. A separate report identifies inconsistent criteria for what constitutes a welfare check and when it is recorded. Together these support a recurring concern with the reliability of the bounded welfare-check process, rather than only the sheet or recording component.2 reportsSelected
  49. Unsafe cell-sharing decisions for prisonersRecurring deficiencies in the prison cell-sharing risk system can leave prisoners exposed to unsafe pairings or prevent them from understanding changes affecting their cell-sharing status.2 reportsSelected
  50. Unsafe medication prescribingReports identify unsafe prescribing decisions in which the medication, indication, dose or relevant clinical circumstances are not adequately evaluated before prescribing.2 reportsSelected

Reports matching this recurring concern

19 Sep 2013 – 7 Sep 2026 · 4 reports

Classification: all classificationsRecurring concern: Unreliable hospital discharge processes
Reset

Only concerns related to the selected recurring concern and their linked actions and positions are shown. Other concerns are counted within each report.

Actions and positions describe what recipients stated in published responses. Select a statement to read its source wording.

  1. Inner South London

    AI-generated summary

    Mark Robert Smith · Prevention of Future Deaths report

    This summary was generated using AI from the published report. Please read the original report for the complete account.

    Report summary

    Mark Robert Smith, who was detained at HMP Thameside, died after an epileptic seizure led to cardiac arrest. The report describes concerns about medication prescribing and administration, hospital-to-prison discharge and communication, access to his cell, observation arrangements and records, staffing, and the lack of equipment for resuscitation.

    Read the report on judiciary.uk

    Source evidence

    How this individual concern was interpreted

    PFD Monitor created a concise, searchable interpretation from the report wording shown below.

    PFD Monitor interpretation

    Failure to coordinate prison discharge through an MDT and relevant prison healthcare liaison

    Wider context from the report

    “(2) I remain concerned a similar situation could arise as did in Mark’s case regarding a discharge back to prison from QEH without an MDT meeting despite repeated requests from prison healthcare and there could be a risk to life. I note PFD evidence provided regarding improved liaison between QEH and the prison GP but note no reference to liaison between QEH and the head of the healthcare in prison, who visited QEH twice in Mark’s case. Further, as recently as 2025 an incident occurred when there was a lack of understanding at QEH around the limited provision in healthcare in prison. I am concerned as to how awareness of the limits of prison healthcare will be disseminated on a continuing basis to new and locum staff at QEH. (QEH and Practice Plus Group) ”

    Source location

    Mark Robert Smith · Prevention of Future Deaths report
    Page 4 · concerns

    Open source report
  2. East London

    AI-generated summary

    Andrew Ewin-Ripp · Prevention of Future Deaths report

    This summary was generated using AI from the published report. Please read the original report for the complete account.

    Report summary

    Andrew Ewin-Ripp, who had epilepsy, suffered a fit at home on 1 November 2022, was found in cardiac arrest, and died in hospital on 4 November 2022. The reported cause of death was sudden unexpected death in epilepsy (SUDEP). Concerns included lengthy waits for neurology care, the absence of clear guidance and systems for monitoring, discharge information, and urgent review after seizure recurrence.

    Read the report on judiciary.uk

    Source evidence

    How this individual concern was interpreted

    PFD Monitor created a concise, searchable interpretation from the report wording shown below.

    PFD Monitor interpretation

    Failure to require practitioners to pass essential epilepsy information to patients on discharge

    Wider context from the report

    “There are believed to be around 500-600 SUDEP deaths in the UK each year. SUDEP deaths are common in young adults. The waiting times for outpatient neurology appointments is in the region of 9 months for the trust concerned. The inquest heard from an independent expert that it is not unusual to have waiting times of more than 6 months for outpatient neurology appointments. In the context of these lengthy waiting times, the following matters were of particular concern: (1) GP practices are not required to carry out annual reviews of epilepsy patients, as they are, for other chronic diseases. The independent consultant neurologist considered that annual reviews by general practitioners would provide an excellent safety net to prevent future SUDEP deaths. The reviews could incorporate checks on compliance with medication; reviews of any seizure activity and reminder of ways to reduce the risk of seizures. (2) There is clear national guidance in relation to how quickly patients should be seen following a first seizure, but no clear guidance around the longer-term monitoring of patients with epilepsy. How soon after the last seizure is it safe to discharge a patient? There is no clear guidance on this. (3) After discharge from the secondary care team, there was no clear guidance provided in relation to the importance of maintaining full compliance with medication even if seizure free for a very long period; the importance of notifying the GP and/or the secondary care team about the recurrence of any seizure activity or clear guidance on how best to make contact with the secondary care team in the event of recurrence of seizures. There was no system in place, or guidance, requiring practitioners to ensure that this essential information is passed to patients on discharge. (4) There was no care pathway for incorporating urgent reviews in neurology clinics in response to patients reporting concerns, such as a return of seizures or not tolerating medication. ”

    Source location

    Andrew Ewin-Ripp · Prevention of Future Deaths report
    Page 3 · concerns

    Open source report
  3. North Wales (East and Central)

    AI-generated summary

    Kate Louise Pierce · Prevention of Future Deaths report

    This summary was generated using AI from the published report. Please read the original report for the complete account.

    Report summary

    Kate Louise Pierce became unwell on 29 March 2006 and was discharged from hospital after being diagnosed with viral tonsillitis. She returned approximately 36 hours later in a deteriorated condition, was diagnosed with pneumococcal meningitis, and later died in Florida on 19 March 2013 after suffering brain damage and other health conditions. The principal concerns were uncertainty about when a sick child should receive senior paediatric assessment before discharge, and a lack of clearly defined and consistently applied processes for identifying and acting on learning opportunities after adverse events or re-presentation.

    Read the report on judiciary.uk

    Source evidence

    How this individual concern was interpreted

    PFD Monitor created a concise, searchable interpretation from the report wording shown below.

    PFD Monitor interpretation

    Lack of clarity about senior paediatric assessment requirements before discharge

    Wider context from the report

    “(1) There remains uncertainty about the circumstances in which a sick child should be seen by a senior Paediatrician (Registrar or above) prior to discharge. During the hearing I was shown a document headed ‘Guidance to Paediatric Junior Doctors on Discharge [sic] Children From Assessment Unit’, which I was given to understand reflects current practice and represents an improvement on the position in 2006. The information in the letter of 4 October 2017 is broadly consistent with it. Both documents are silent as to whether a parental request for a second opinion should automatically lead to an examination of the child by the senior doctor, as was opined in court, where the importance and significance of parents’ views were noted. . I am concerned that a lack of clarity about the Health Board’s expectations in this respect may continue to allow for the possibility of a child being discharged without a sufficient (and sufficiently senior) assessment having been made. ”

    Source location

    Kate Louise Pierce · Prevention of Future Deaths report
    Page 2 · concerns

    Open source report

    Source evidence

    How this respondent action was interpreted

    PFD Monitor created a concise, searchable interpretation from the published response wording shown below.

    PFD Monitor interpretation

    Alter parent discharge information to state explicitly that parents may request escalation for a consultant review.

    Verbatim wording from the response

    “Whilst we are confident that the culture of the department is such that a parental request or indeed concern from any member of staff would lead to a consultant review, we have taken the decision to alter the parent discharge information to explicitly state this.”

    Source location

    2017-0312-Response-by-University-Health-Board
    Page 1 · response
    Published 28 November 2017

    Open published response
  4. South Yorkshire (Eastern)

    AI-generated summary

    ROSEMARY BRONWYN FERGUSON · Prevention of Future Deaths report

    This summary was generated using AI from the published report. Please read the original report for the complete account.

    Report summary

    ROSEMARY BRONWYN FERGUSON, who had a long-standing history of epilepsy, was admitted after a head injury from a fall and was assessed as unfit for discharge because of risks associated with further falls. Despite recommendations that she remain in hospital, she was discharged to the care of a friend and was found deceased alone at home on 11 March 2013; the medical cause of death was recorded as sudden unexpected death in epilepsy. The principal concerns were that Social Services were not notified of the discharge, the friend’s expected role was unclear, and hospital records did not adequately document key communications and the date of discharge.

    Read the report on judiciary.uk

    Source evidence

    How this individual concern was interpreted

    PFD Monitor created a concise, searchable interpretation from the report wording shown below.

    PFD Monitor interpretation

    Failure to notify Social Services of discharge contrary to their recommendations

    Wider context from the report

    “(1) The social workers left work on Friday 8th March, 2013 believing that, following their recommendations, Ms Ferguson would remain in hospital over the weekend and accordingly they did not put into place any support measures for her. The clinician’s decision to discharge her before support measures could be put in place was not communicated to Social Services. If it had been, this would have given an opportunity for them to take urgent supportive action. As it transpired, Ms Ferguson died from Natural causes rather than, for example, Injuries sustained in a further fall, or a deterioration of her earlier head injury, but I apprehend danger in the future if discharge occurs contrary to Social Service recommendations without the discharge being notified to them. ”

    Source location

    ROSEMARY BRONWYN FERGUSON · Prevention of Future Deaths report
    Page 2 · concerns

    Open source report

Export data

Your Excel file will include all 4 matching reports.

Classification: all classifications19 Sep 2013 – 7 Sep 2026 (Publication date)Recurring concern: Unreliable hospital discharge processes
Choose the tabs to include