An accepted principle in intensive care medicine is that is intensive care should be provided for those with the greatest chance of benefit from the care. The demand for intensive care exceeds availability even in high income countries. So intensive should not ideally be provided those for whom death seems unavoidable, although palliative care in the intensive care unit (which sounds like an oxymoron) is growing. (1) Ironically many deaths in intensive care may be “better” than deaths on acute wards simply because the staff are so familiar with death and the ratio of staff to patients is much higher.
Decisions on whom to admit to intensive care are inevitably complex, although the chance that the patient can benefit is a prime criterion. The The Intensive Care National Audit and Research Centre (ICNARC) has been collecting data and conducting research in intensive care in England, Wales, and Northern Ireland for 25 years and has seen an increase in the number of intensive care beds and a drop in mortality.(2) The centre has developed a score that predicts accurately the chance of a patient dying in intensive care and produced for the Commission data on patients who had 80% or greater chance of dying (or 20% or less chance of surviving) in intensive care.
Overall in 2018-19 there were 163 340 total admissions to adult intensive care with 30 195 deaths (18.5%). Over the decade 2009-19, 93 870 patients with a greater than 80% chance of dying were admitted, accounting for 6.4% of all admissions; 57.5% of the admissions were for men, and the average age of those admitted was 68.6 years. The proportion of admissions varied across critical care units from 0 to 14.5% (median 6.5%, interquartile range 5.1 to 7.9%). After excluding readmissions of the same patient within the same acute hospital stay, overall acute hospital mortality for these patients was 89.5% (80,593/90,047), reflecting the accuracy of the predictive score. Over the decade the proportion of these admissions declined slightly but the absolute number increased slightly as the number of intensive care beds increased.
Nobody can say whether these rates are right or wrong, but they are interesting in providing data on the degree to which treatment continues when the risk of death is high. The variation that is seen among units is seen in almost all parts of health care, and it is reassuring that there has been no increase in the proportion of high risk patients admitted, showing that there is no increasing tendency to treat those likely to die.
We thank ICNARC for suppling these data, and their full paper is attached.