When, late in 2019, Covid-19 was identified as the microbial agent causing a severe viral pneumonia in several patients epidemiologically linked to a seafood market in Wuhan, in China’s Hubei province,1 few would have predicted that within weeks it would lead to a pandemic that is the greatest threat to global health for over a century.

It has virtually monopolised the media and public discourse, and has necessitated rapid transformation in how health services, including primary care, respond. Yet this dreadful cloud may have some silver linings, allowing us to explore different ways of working and questioning whether some existing procedures could be simplified.

In this article, we argue that the temporary changes to the law for death certification and cremation should provide the impetus for overdue reform of these archaic processes, as well as the coroner’s service. We also present and discuss two clinical cases. 

Underpinning the new guidance is the Coronavirus Act 2020. In a nod to limitations imposed by lockdown and pressures on the medical profession, the key modifications are extending the period within which the doctor signing a death certificate has attended a patient from 14 to 28 days, allowing video consultation to meet that requirement, and (with the coroner’s approval), accepting a non-attending doctor as signatory provided they have access to relevant medical records.2 Cremation is also simplified, with no Part 2 doctor and allowing, with the family’s approval, funeral directors to act as informant to the registrar.2

This legislation is particularly relevant to deaths occurring in the community, and therefore to GPs. Around 58% of deaths in the UK occur in hospitals,3 where a recent medical review can be assumed. By contrast, end-of-life care in the community involves numerous nursing, medical and ancillary professionals, and not being seen by a named doctor within a fortnight preceding death hardly equates with poor care or neglect.

Current procedures are somewhat Victorian, harking back to days of far greater doctor-centeredness and less sophisticated community-based multi-disciplinary care than is the modern reality. It seems reasonable, therefore, to argue the case for the attendance-to-death time to be permanently increased to 28 days, and to further modernise practice by allowing video-conferencing to qualify as ‘seeing’, for its use as a communication tool will certainly increase.4

Deciphering a precise cause of death in an ageing population

As a mission statement, accuracy of cause of death is desirable as a marker of quality of care, a courtesy to relatives and for national statistics. Nevertheless, deciphering a precise cause of death often mirrors the reality of clinical practice in having inherent uncertainty, dominated as it is by multi-morbidity in an ageing population.5 In most cases an acceptable standard would be that the certifying doctor acts in good faith and on the balance of probability. One can also reasonably assume that deaths of paramount interest to public health, those attributed to cancer, circulatory disease and trauma including accidents and homicides, occupy the more certain end of the spectrum.

By contrast, genuine uncertainty or suspicion necessitates referral to the coroner’s service. But this is also an archaic system, over 800 years old and clamour for its reform is not new.6 In 2018, 41% of all deaths in England and Wales were reported to coroners,7 with 39% of these (16% of all deaths) having an autopsy;7 though the figure has declined in recent years, it still seems high.

Furthermore, even a process perceived as being definitive does not always provide absolute certainty, and concern has been expressed that the increase in narrative verdicts – legalese for ‘we still don’t know’ – has under-estimated the true prevalence of suicide in particular.8 This in turn underpinned legislation lowering the level of proof for suicide at inquests to the civil standard of balance of probability.9

The process of cremation is also overdue a review. In its current form it requires involvement of three doctors; the usual medical attendant, a Part 2 doctor and the crematorium’s medical referee. Reasonable checks and balances surely do not require both of the latter two, one of which could be dispensed with. Current arrangements are, above all, time-consuming. Our own practice is a case in point. Though in an affluent commuter zone with high longevity, our data indicates that around 60% of patients die the community – a far higher percentage than the national average – of whom one-third are nursing home residents, and the absolute death rate is higher than average due mainly to the latter cohort.10

The UK is not an over-doctored country, and time devoted to bureaucracy around death is time lost to caring for the living. 

The following two cases demonstrate some of the issues discussed thus far:

Case 1: death following head injury

A man in his late 70s with a history of Parkinson’s disease, Lewy body dementia, subdural hygroma and hypertension sustained a fall leading to a scalp laceration in the nursing home in which he lived. He has admitted briefly to hospital where the laceration was sutured. Head and neck CT scans were performed, showing modest cerebral and cerebellar atrophy and widespread small vessel ischaemia, but no haemorrhage or infarction. A week after his return to the home, he was found unresponsive early one morning, and death was confirmed by paramedics.

A video-link discussion with the nursing staff took place, and whilst it was agreed that the death was not entirely unexpected, its sudden nature and that fact that it followed soon after a hospital admission with a head injury meant that referral to the coroner was inevitable.  

The salient findings at autopsy were marked left ventricular dilatation, pulmonary oedema and absence of the substantia nigra, a relevant negative being the absence of cerebral haemorrhage or infarction. Thus death was due to natural causes: 1a) pulmonary oedema due to 1b) hypertensive disease and 2) dementia.

Case 2: bacterial chest infection

A lady in her early 90s with a history of Alzheimer’s disease was steadily and predictably declining from the condition in a nursing home. Staff reported a mild fever, nausea and reduced desire for food and drink. She had no cough, nor any other symptoms likely to localise a site of infection. As the Covid-19 outbreak was established by then, a video-link was set up with the staff and patient where it was established that her oxygen saturations were dropping, though she remained comfortable.

She had a DNAR form in place, and though Covid-19 was recognised as a possibility, there were as yet no confirmed cases in the locality, so after consideration a bacterial chest infection was deemed much more likely. She was prescribed antibiotics, but did not respond and a week later, following further discussion with the nursing staff and relatives, it was decided to manage her palliatively.  She died peacefully a further week later.

Loosening regulation bucks the trend post-Shipman 

The first case demonstrates the not infrequent instance of sudden death occurring in a frail elderly person, an event not surprising per se, but in which the absence of immediate antecedent symptoms and signs makes it difficult to opine on the likely cause.

The recent head injury, which raised the distinct possibility of a delayed intracranial bleed, meant the case had to be referred to the coroner, though the latter would surely have been involved had that fall had not occurred. Similarly, the autopsy conclusions, though not in themselves surprising, could not reasonably have been predicted by the story of his last few weeks of life.

By contrast, in the second case the broad diagnostic categories of the immediate and underlying cause of death – respiratory infection and dementia respectively – were clearer. The spectre of Covid-19, however, raises the possibility that, in the event of the pandemic taking on a more chronic rather than transient form [and assuming no immediate prospect of a vaccine], the public health implications may deem that its precise diagnosis or exclusion is mandatory. There is no precedent for this: bronchopneumonia, after all, appears on countless death certificates based on clinical rather than bacteriological criteria, so such an additional hurdle would only be feasible with near-universal testing, instant results and generous government funding. 

We acknowledge that our case for loosening rather than tightening regulation bucks the trend generally applied to healthcare in recent decades. In the wake of the Harold Shipman debacle, Dame Janet Smith, the High Court judge who led the inquiry, called for draconian tightening of the regulation around death certification and cremation in her report.11

That stance was understandable in the context of the public fury stirred in the aftermath. Thankfully, common sense prevailed and, apart from the appointment of medical examiners to check death certificates12 (which only started in April 2019), the processes have remained largely intact.

We must keep a sense of perspective: Shipman was an extreme statistical outlier even in the annals of mass murderers, and hardly symptomatic of major flaws in procedure. Simplification of the processes of death certification and cremation would have many benefits, including being quicker for professionals and bereaved families, making fewer referrals to coroners and taking less time away from clinical care.   

Edin Lakasing and Sukaina Hirji, General Practitioners and Trainers, Chorleywood Health Centre, 15 Lower Road, Chorleywood, Hertfordshire WD3 5EA.

Email: edin.lakasing@nhs.net

Competing interests: none.




  1. Zhu N, Zhang D, Wang W et al. A novel coronavirus from patients with pneumonia in China, 2019. N Engl J Med 2020; 382: 727-733.
  2. Luce T. Covid 19: death certification in England and Wales. BMJ 2020; 369:
  3. Gomes B, Higginson I. Where do people die? [1974-2030]: past trends, future projections and implications for care. Palliat Med 2008; 22(1): 33-41.
  4. Greenhalgh T, Wherton J, Shaw S, Morrison C. Video consultations for Covid-19. BMJ 2020; 368:
  5. Lakasing E, Minkoff S. Uncertainties in death certification. British Journal of General Practice2012; 62 (605): 658-659.
  6. Pounder D. The coroner service. A relic in need of reform. BMJ 1999; 318: 1502-3.
  7. Ministry of justice and Office for National Statistics 2019. Coroners Statistics Annual 2018 (England and Wales). https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/810303/Coroners_Statistics_Annual_2018.pdf
  8. Gunnell D, Hawton K, Kapur N. Coroners’ verdicts and suicide statistics in England and Wales. BMJ 2011; 343:
  9. Appleby L, Turnbull P, Kapur N, Gunnell D, Hawton K. New standard of proof for suicide at inquests in England and Wales. BMJ 2019; 366:
  10. Lakasing E, James AE, Parry SB, Mutero F. A practice-based survey of mortality. GM 2018; 48(6). https://www.gmjournal.co.uk/a-practice-based-survey-of-mortality  (accessed 14 May 2020).
  11. Smith J 2003. The Shipman Inquiry. Third report - Death Certification and the Investigation of Deaths by Coroners. London, HMSO. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/273227/5854.pdf (accessed 14 May 2020).
  12. Luce T. Death certification reform in England. BMJ 2018; 361: