Article Text

Download PDFPDF

Implications for COVID-19 triage from the ICNARC report of 2204 COVID-19 cases managed in UK adult intensive care units
  1. Claire L Shovlin1,2,
  2. Marcela P Vizcaychipi3,4
  1. 1 NHLI Vascular Science, Imperial College London, London, UK
  2. 2 Department of Respiratory Medicine, Imperial College Healthcare NHS Trust, London, UK
  3. 3 Department of Anaesthesia and Intensive Care, Chelsea and Westminster Hospital NHS Foundation Trust, London, London, UK
  4. 4 Department of Surgery and Cancer, Imperial College London, London, UK
  1. Correspondence to Professor Claire L Shovlin, London, UK; c.shovlin{at}; Dr Marcela P Vizcaychipi; Marcela.Vizcaychipi{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

On 4 April 2020, the Intensive Care National Audit and Research Centre (ICNARC) reported data from 286 adult intensive care units (AICUs) across England, Wales and Northern Ireland.1 Of 2204 patients admitted with COVID-19, 1524/2204 (69%) remained on AICU, 340 (15.4%) had been discharged and 340 (15.4%) had died.1 These survival rates emphasise the crucial importance of intensive/critical care support for patients most severely affected by COVID-19.

The 2204 COVID-19 cases were compared with 4759 patients with non COVID-19 viral pneumonia admitted to the same AICUs in the previous 3 years.1 The striking difference was that prior to their respective illnesses, the COVID-19 cohort was significantly healthier, with much lower disease burdens in the preceding 6 months (figure 1).

Figure 1

COVID-19 and non-COVID cases in the Intensive Care National Audit and Research Centre (ICNARC) report of 4 April 2020.1 Percentage of total cases with the respective disease burden within the 6 months prior to critical care, as defined by ICNARC1: immunocompromise: chemotherapy, radiotherapy or daily high dose steroid treatment in previous 6 months, HIV/AIDS or congenital immune deficiency; respiratory: shortness of breath with light activity or home ventilation; haematological malignancy: acute or chronic leukaemia, multiple myeloma or lymphoma; cardiovascular: symptoms at rest; metastatic disease: distant metastases; liver: biopsy-proven cirrhosis, portal hypertension or hepatic encephalopathy; and renal: renal replacement therapy for end-stage renal disease. For this manuscript, p values were calculated by Fisher’s exact test, and the data presented graphically as mean and SE of the mean, using GraphPad Prism 7.03 (GraphPad Software Inc, San Diego, California, USA).

While it is theoretically possible that all critically ill patients with COVID-19 were genuinely healthier, the total COVID-19 death figures of 3939 in the UK by the same date2 indicate more than 10 times as many were dying without accessing AICU. This drew our attention to AICU COVID-19 triage which, in the UK, has generally occurred on arrival in hospital, via algorithm guidance. Many proposals were available early in the pandemic, and UK hospital Trusts implemented local policies aiming to avoid overburdening AICUs as a time of unprecedented demand.3

While some triage documents are very reasonable,4 content has varied. For example, one COVID-19 decision support tool that was circulating in March 2020 (no longer available online) suggested adding points scored across four elements: age (extra points for each 5-year increments above 50 years), the 9-point Clinical Frailty Scale, comorbidities (a point each) and male sex. Implementation of such tools could prevent healthy, independent individuals from having an opportunity to benefit from AICU review/admission by protocolised counting of variables that do not predict whether they would personally benefit from AICU care. The European Very elderly Intensive Patient 2 study recently reported that the Clinical Frailty Scale was more important than age alone in models of 30-day mortality in 3920 AICU-admitted patients aged 80–104 years.5 Additionally, the extremely common states of diabetes, hypertension and male sex indicate patients requiring extra care, rather than less.

Vulnerable groups become a self-fulfilling prophecy when implemented in triage decisions. From the 4 April 2020 ICNARC report,1 UK total deaths2 and continuing AICU bed availability,2 we conclude that current triage criteria are overly restrictive and suggest review. COVID-19 admissions to critical care should be guided by clinical needs regardless of age.


These data derive from the Intensive Care National Audit and Research Centre (ICNARC) Case Mix Programme Database. The Case Mix Programme is the national clinical audit of patient outcomes from adult critical care coordinated by the ICNARC. For more information on the representativeness and quality of these data, please contact ICNARC. We would like to thank ICNARC for their comment that 'due to the relatively low proportion of patients that have completed their critical care, all outcomes should be interpreted with caution', and therefore did not perform any analyses on outcomes in the cohorts. We would also like to thank all our healthcare professional colleagues striving to improve the outcomes for patients with COVID-19 and the general population for their adherence to the difficult restrictions placed on their activity. The authors wish to note that triage documents referred to were not from their own institutions.



  • Contributors Conception and design: both authors. Analysis and interpretation: both authors. Drafting the manuscript for important intellectual content: both authors. In detail: both authors performed literature searches and designed the work based on clinical experience, particularly from MPV. CS performed the data analysis and wrote the first draft. Both authors contributed to data interpretation and manuscript revisions before joint approval.

  • Funding The study received no specific funding support. CLS acknowledges support from the National Institute of Health Research Biomedical Research Centre Scheme (Imperial BRC).

  • Competing interests None declared.

  • Patient and public involvement Focusing of data interpretation towards the triaging of patients was an outcome of inputs from British patients contacting CLS, focusing on the question 'Am I at High Risk?'.

  • Patient consent for publication Not required.

  • Ethics approval No ethics approvals were required.

  • Provenance and peer review Not commissioned; internally peer reviewed.