Intended for healthcare professionals


Making ordinary decisions in extraordinary times

BMJ 2020; 370 doi: (Published 26 August 2020) Cite this as: BMJ 2020;370:m3268
  1. Rachel Burman, consultant in palliative care,
  2. Ruth Cairns, consultant in liaison psychiatry (older adults),
  3. Sergio Canestrini, consultant in critical care,
  4. Robert Elias, consultant in renal medicine,
  5. Victoria Metaxa, consultant in critical care,
  6. Gareth Owen, consultant in liaison psychiatry,
  7. Hazel Polat, senior solicitor,
  8. Alex Ruck Keene, barrister,
  9. Emer Sutherland, consultant in emergency medicine
  1. Covid-19 Ethics Working Group, King’s College Hospital, London, UK
  1. Correspondence to: Ruth Cairns ruth.cairns{at}

Members of a covid-19 ethics working group at a large London teaching hospital argue that a lack of detail in national decision support guidelines led to fear driven anticipatory triage during the first wave of the pandemic

The pandemic arrives

The covid-19 pandemic has led to unprecedented uncertainty for governments and healthcare bodies across the world, bringing with it the very real threat of overwhelmed systems. The perennial question for ethicists about if and how to plan on a utilitarian basis to save the maximum number of lives has been brought into stark focus. In March 2020, it became an urgent operational question in the UK, needing the attention of policy makers, healthcare providers, and clinicians in a way we have not previously encountered.1 Clearly, in such a situation it was not going to be possible to get everything right. In the UK, steps were quickly taken to increase intensive care unit (ICU) bed capacity and to protect existing secondary care health structures by cancelling all elective and outpatient work. There was also a recommendation to GPs that high risk groups should have advance care plans in place for decisions around hospitalisation and ventilation with covid-19, as well as around resuscitation status.2

All these steps, however, were taking place in the context of a dearth of national guidance. Clinicians were able to look to countries affected by the pandemic ahead of the UK34 to see what might lie ahead. Guidance, most notably that produced by the Italian Society for Anaesthesia, Analgesia, and Intensive Care, suggesting that there might need to be a simple age cut-off for admission to ICU, appeared to many to be too simplistic. In the UK, an early—and very high level—attempt was made by the National Institute for Health and Care Excellence (NICE) to provide critical care guidance for the pandemic.5 Within days of publication, however, this was the subject of threatened judicial review proceedings because of concern that its reliance on the Clinical Frailty Score would discriminate against people with learning disabilities or other “stable” cognitive impairments.6 NICE and NHS England published no further detailed guidance as to under what circumstances ICU triage—deciding how to allocate limited ICU resources—might have to be considered or how it should be undertaken.

Not publishing national ICU triage guidance seems likely to have been influenced by political (including health service political) concerns about public reaction to media portrayals of ICU doctors “playing God,” as well as by the ICU expansion policy. It is striking that the first guidance seeking to guide triage decisions in an operational fashion was not published by NICE, NHS England, or the Department of Health and Social Care, but by the Intensive Care Society—and was also not published until after the peak of the first wave in London.7

As a covid-19 ethics working group at a large London teaching hospital, we believe that the lack of practical, detailed national guidance has had other unforeseen consequences that require consideration. What we have learnt from the first wave of covid-19 has implications for a potential second wave and for subsequent public health emergencies that could place NHS resources under strain.

Riding the first wave

As the first wave passes, we have seen the NHS stand up to the many demands of the pandemic without being completely overwhelmed, at least in terms of ICU beds and ventilators. It is important, however, to acknowledge that it has been affected and stretched at other points in the system. The expansion in ICU beds and ventilators also stretched the quality of intensive care such that normal Guidelines for the Provision of Intensive Care Services standards have been harder to meet, and clinical decision making changed in the process.8 We need to reflect on the impact this has had on patient and NHS staff outcomes and to learn from this before a second wave of the pandemic.

Unexpected consequences of the first wave were a reduction in emergency department attendances—at its peak, a fall of 57% in April 20209—and reduced bed occupancy in general hospitals resulting from this, together with a reduction in elective and GP admissions. In light of the high numbers of UK deaths and the spotlight on care home deaths, the question must be asked: how did this happen? Some of the explanation was undoubtedly down to people choosing not to “burden” their local hospitals, even in the presence of non-covid-19 related serious conditions, such as myocardial infarction and stroke presentations1011 or being fearful that going into hospital would infect them. This poses complex and challenging questions about the messaging that was used in relation to healthcare—the simple message of “stay home, save the NHS” unfortunately did not, in all cases, equate to “save lives.”

But another part of the explanation, we suggest, is because anticipatory triage was taking place in the community, driven by fear of an overwhelmed health system. There is increasing concern that GPs were put in the unenviable position of being asked to contact their elderly and frail patients to hastily discuss decisions about ICU admission and resuscitation. This, together with altered ambulance service thresholds for transfer to hospital,1213 led to barriers to accessing hospital for some who may have benefited from medical treatment, irrespective of whether ICU admission would have been appropriate. In short, demand suppression resulting from anticipatory triage limited access to hospital care. One hypothesis is that failure to communicate a clear decision making and triage policy resulted in professionals making up the policies in their own minds—without consistency and under conditions of fear—where the worst eventualities were expected with unchecked biases.

Despite not needing to ration ICU beds, the sheer volume of service rearrangement, staff sickness and self-isolation, redeployments, and complex clinical decision making, particularly at the height of the first wave, proved extremely burdensome for doctors. The lack of data available about covid-19 outcomes at the beginning of the pandemic required a generic approach to management that compounded the challenges involved in managing an illness about which we still have a lot to learn. These factors may have contributed to increased levels of stress and reduced confidence in decision making. Additionally, pressures both to get patients out of hospital to create bed capacity and the anticipation of needing to make triage decisions, in preparation for what it was feared lay ahead, may have also been a factor in altered decision making. The lack of open dialogue and clarity about the tipping point between normal treatment escalation decisions and ICU triage decisions may have played a part in systems imposing higher than usual thresholds for accessing medical treatment. It seems likely that raised thresholds of transfer to hospital had unanticipated consequences for patients in high risk groups. Put bluntly: what proportion of the “excess” 18 000 care home residents who died in the UK during the first wave of the pandemic14 might have gone to hospital for medical care and survived in non-covid-19 times? And—even more bluntly—what proportion of those residents chose not to go to hospital, and what proportion had the choice made for them?

The core of the challenge

We know from existing evidence that difficult decisions around levels of and access to treatment are psychologically challenging and associated with moral distress in healthcare professionals.1516 Many hospitals, including our own, have responded to the first wave by developing decision making guidance and support to aid the process and share its burden. Work is still required to embed supported decision making within hospitals, enabling doctors to move away from the (often self-imposed) culture of independent decision making in complex clinical situations. On reflection, it may be that the main challenge for doctors during the first wave was adjusting to ordinary decision making (decision making not based on triage) in extraordinary times (when an encounter with triage loomed) and that this compounded the high volume of decisions and lack of time to make them, as well as any psychological distress that arose in the process. If this is an accurate analysis of the situation, it may be that having more decision making support in place for the second wave will allow ordinary decision making that is better adjusted and confident. In doing so, this may in turn allow a change in the psychological language away from “distress” to “adjustment” and “resilience” instead.

How to do it better

We now have time to pause and reflect on the UK’s first experience of the pandemic and think about what else needs to change in preparation for a second wave, which may be upon us later this year. With ICU expansion, we did not run out of ICU beds. Frontline staff were, however, placed under significant strain and other resources, including some drugs, syringe drivers, and renal replacement therapy, were stretched across the community and hospitals. We suggest that political reluctance to tackle the possibility of ICU and ventilator triage has had a damaging effect on more widespread decisions around access to healthcare. The worst did not come to pass but the vacuum of open dialogue about carefully considered and detailed ethical ICU triage guidance—and a failure to delineate when ordinary decision making should stop and extraordinary decision making start—is likely to have contributed towards implicit triage in the community, on the wards, and by patients themselves. This pre-hospital and pre-ICU decision making was, in many cases, influenced by fear rather than the reality of the situation.

It is hoped that the starting position will be different, with more knowledge and systems in place, in facing a second wave of the pandemic. We suggest that in addition to this, patient and staff outcomes could be improved by introducing transparent ethical guidance (with a degree of detail) and decision making support at all access points to healthcare, and also by ensuring good quality information sharing between hospital and community that is visible to individuals and teams making clinical decisions. The latter could involve sharing hospital bed availability with primary care and ambulance services to prevent unnecessary pre-hospital demand suppression. Responsibility for implementing these changes lies at a local level, but ultimately within a national framework so as to ensure consistency. These changes, that we suggest were an omission from the Third Phase of the NHS Response to Covid-19,17 would also be crucial to ensure that the backlog of other NHS procedures that were put on hold in the first wave does not become any longer.

The pandemic may last for another year—and for years without a vaccine. We need to think longer term about how to ensure that a fatigued system is supported.

Key messages

  • There was a lack of detailed practical national decision support guidance during the first wave of the covid-19 pandemic in the UK

  • Intensive care bed capacity was not exceeded but resources were stretched and the authors suggest fear driven anticipatory triage impacted admission and escalation of treatment thresholds

  • Explicit guidance and open dialogue could have facilitated ordinary decision making in extraordinary times and a renewed emphasis on the fundamentals of such decision making is required

  • Transparent ethical guidance and good quality information sharing of hospital bed states with primary care and ambulance services could help reduce unnecessary pre-hospital triage in a second wave of the pandemic


We thank Leonie Penna and other King’s Health Partner colleagues at King’s College Hospital NHS Foundation Trust, Guy’s and St Thomas’ NHS Foundation Trust, and Lewisham and Greenwich NHS Trust. We also thank members of the pan-London Sustainability and Transformation Partnership ethics consortium.


  • Contributors and sources: This article is written by a covid-19 ethics working group made up of consultants from intensive care units, renal medicine, emergency medicine, palliative care, liaison psychiatry, and two lawyers. The group worked in a large London hospital to develop decision support guidance at speed during the height of the pandemic in the UK in spring 2020. The paper reflects on the experiences of this group, as well as presenting its central hypothesis that having explicit national triage criteria could have prevented some of the worst fear driven anticipatory triage seen at this time. We offer some recommendations for minimising these consequences for a potential second wave.

  • Patient involvement: Because of the nature of this article, the authors did not seek patient involvement.

  • Competing interests: We have read and understood BMJ policy on declaration of interests and have no relevant interests to declare.