New Swedish Guidelines for ICU Priority Setting in Exceptional Circumstances

This post was updated March 30, 2020 with criteria for rationing outside ICU!

During the last two weeks, I have, together with a bunch of philosophy and bioethics colleagues specialised on these kinds of questions not only in theory but also in practice, contributed to a speedy development of of new central guidelines on ICU priority setting and resource allocation from the National Board of Health and Welfare (one of the central governing authorities of the Swedish healthcare system). The guidelines address exceptional circumstances, such as an unusual high pressure of cases that exceeds normal capacity, and are, of course, prompted by the Covid19 pandemic. Lars Sandman, head of the national Center for healthcare priority setting, led the work, and it took a little less than two weeks to go from zero to end product. Besides philosophers, the National Council for Medical Ethics, representatives of the Ethics Delegation of the Swedish Society of Medicine, and ICU specialist consultants have contributed.

As the legal language of Swedish public documents is Swedish, this is also the language of the guidelines. This full version can be read and downloaded here. But, as there has been quite a bit of international interest, here is a short English summary by myself (ie. this is not an official translation from NBHW):

BACKGROUND

Health care priority setting is already regulated by special laws, where non-instrumentality, equal treatment and non-discrimination, priority based on patient need only, and ban on unnecessary waste of resources are the key principles decided by the Swedish Parliament since before.

APPLICATION IN THE GUIDELINES

First of all, ICU capacity should be strengthened by rationing less acute or important care, or care than can be postponed without severe consequences (and without disturbing management of the pandemic), so that an exceptional ICU situation is avoided as far as is practically possible. There is a specific appendix in the full guidelines setting out rationing principles to this effect (see below).

In exceptional circumstances, ICU should focus on patients where the treatment can do the most good, thus focusing more on the prognosis of the treatment than the severity of the condition (both dimensions of patient need). This means that indication for ICU may be strengthened to exclude some patients with severe conditions that are given ICU in normal circumstances to give extra time for planning and family contact.

While the legal regulation excludes chronological age as priority basis in its own right, it allows for life expectancy (if given successful care) to be part of the assessment of the patient need for a resource (besides the severity of the condition and the immediate prognosis of the treatment). Factors to assess life expectancy in the clinic are:

• Extent and severity of vital organ failure
• Extent and severity of pre-existring co-morbidities
• High chrononological age in combination with a bad prognosis after intensive care

Priority groups are patients with:

Priority 1

Severe condition or injury with >12 months life prognosis, otherwise meet indication for ICU, and don't suffer severly decreased life expectancy (see above). If there is necessary to make a choice between patients in this group, patients with a general longer life expectancy (higher biological age) should be given priority.

Priority 2
(a) One or several serious systemic diseases with extensive functional impact, and/or (b) expected survival of 6-12 months based on pre-existing underlying disease.

Priority 3

Bad prognosis in ICU, and where ICU is normally undertaken only to provide time for planning, or family contacts.

These priority groups should ground resource allocation both when assessing new patients, and when reassessing patients admitted to ICU care, and may therefore also ground discontinuation of ICU care, to free resources for new patients with a higher priority.

Patients who are denied or taken off ICU should receive best possible palliative and other care to manage their care need given the ICU resource allocation.

There are also some sections on management of staff ethical stress, and of how to involve patients and family in the decisions.

COMMENTS BY MYSELF

In the preparatory stage, it was discussed whether or not the guidelines should include the feature of being crucial for life saving societal functions as a further item to sort among those in priority group 1. There seems to have been agreement that the requirement of equal treatment for all patients with similar need of care could support this idea if the circumstances are extremely prolonged (as loss of critical staff could then undermine the treatment prospects for future patients). However, the general assessment was that the crisis possibly caused by the Covid19 pandemic for Swedish ICU care would probably not be of this nature, especially in view of the time it would take for someone to return to work after a very critical ICU episode. Should the situation change, the guidelines may be amended to this effect.

RATIONING OUTSIDE OF ICU
These are the rationing criteria for healthcare outside ICU, in order to free capacity for ICU in an Covid19 crisis, mirroring the priority setting criteria above, and based on the same legal principles. They apply on the condition that rationing can free resources/capacity for ICU, and that the rationing does not impede communicable disease control work: 
RATIONING GROUP 1: Elective procedures that may be postponed without deterioratiuon of the condition; 
GROUP 2: Elective procedures that may be postponed without threat to life, abeit with negative effect on QoL: 
GROUP 3: Elective or acute procedures that are unlikely to restore health and likely to lead to prolonged care with risk of ICU need; 
GROUP 4: Acute procedures where postponement will affect QoL during the delay, but not lead to a worse longterm prognosis; 
GROUP 5: Acute procedures where postponement will affect QoL also in the longterm, but not be a threat to survival. 

*