Adopting a mechanism for allocating critical care capacity during a surge

A four-site hospital group decides, in ordinary conditions and in advance, what mechanism would allocate critical care capacity if demand exceeded it. The log is about the mechanism rather than the cl...

Published content

node: Context. The organisation is a hospital group operating four acute sites with a combined 96 adult critical care beds, in a jurisdiction where a public authority can declare a public health emergency or crisis standards of care. Modelling of respiratory surge scenarios indicates demand could exceed staffed capacity for a period of days to weeks, after all conventional surge measures have been taken. No standing allocation mechanism exists. This decision is taken in ordinary conditions, in advance of any surge.
node: What mechanism should allocate critical care capacity if demand exceeds it during a surge: bedside clinical judgement as now, published criteria applied by the treating clinician, a separate triage team at each site, or a single team across the group?
node: Benefit. Retaining bedside judgement requires nothing to be built, trained or maintained, and it is what the group does today.
node: Benefit. Publishing criteria and a threshold is materially better than the status quo on consistency and on activation clarity, and it can be implemented without any new rota.
node: Benefit. A triage team at each site is the arrangement the published frameworks describe, so the group would be adopting a documented approach rather than devising one, and could point to that provenance if the mechanism were ever examined.
node: Benefit. A single team also allows capacity to be balanced across sites, so that a patient is allocated against the group's capacity rather than the capacity of the hospital they happened to reach.
node: Decision-maker. The group board, on the recommendation of the Medical Director and the Clinical Ethics Committee. A mechanism of this kind is adopted by the governing body rather than by a clinical function, because its legitimacy rests partly on who adopted it.
node: Scope. The decision concerns the mechanism: who makes an allocation decision, on what threshold it begins to apply, and what safeguards attach to it. It does not settle the clinical criteria themselves, which are a separate clinical governance matter, and it does not alter the duty of care owed to any patient, which continues under every option. Whether reassessment may lead to a resource already in use being reallocated is a matter for the clinical criteria and is outside this decision, but the mechanism must be able to carry such a decision lawfully and with a record if the criteria provide for it.
node: Consulted. The Clinical Ethics Committee on the framework and its grounding; critical care and emergency clinicians at all four sites on workability; the patient and public involvement panel on how a decision would be explained to a family; and legal advisers on the standing of the mechanism.
node: Risk. It cannot satisfy the constraint of consistent and impartial application, because there is nothing to apply consistently. Decisions of this gravity would be made without a framework, without a record designed for review, and differently at each site.
node: Risk. It leaves the treating clinician deciding access for their own patient, which the published model policy specifically separates. The separation exists both to secure consistency and to protect the clinician, and this option delivers the first only in part and the second not at all.
node: Risk. Twenty-four trained staff must be maintained against an event that may not occur for years. Training decays, staff move, and a mechanism that is documented but not rehearsed performs like one that was never written.
node: Risk. A single team is a single point of failure and a potential queue. If referral volume during a surge exceeds what one team can consider promptly, the mechanism becomes the delay, and the time to allocation is unconfirmed for exactly that reason.
node: Action. Train the four site teams together rather than separately, and rehearse jointly at least annually. Consistency across sites is the weakness of this option relative to a single team, and joint training is the only mitigation available for it.
node: Constraint. Any mechanism must be capable of consistent, impartial and non-discriminatory application, and must be documented before it is needed. A mechanism designed during a surge cannot demonstrate that it was designed impartially.
node: Risk. Not choosing a mechanism is itself a choice, made by default and taken under the worst possible conditions when the surge arrives. The published guidance is directed precisely at avoiding that.
node: Risk. A clinician applying allocation criteria to a patient they are treating faces a conflict between two duties they hold simultaneously. Published guidance treats the duty to care as continuing regardless, which is precisely why the allocation judgement is placed elsewhere.
node: Risk. Four separate teams may diverge in how they apply common criteria, particularly if they train separately. The option secures consistency within a site more reliably than across the group.
node: Risk. Balancing capacity between sites entails transferring patients, which carries its own clinical risk and is not always possible for the sickest. The benefit is smaller in practice than the arrangement suggests.
node: Action. Commission the queueing analysis that would confirm or exclude the group-wide team, so that the option remains genuinely available rather than being abandoned for want of one piece of work.
node: Constraint. The mechanism must not be capable of activation on the judgement of a single clinician or a single site. The model policy this log follows requires both exhausted surge capacity and a declaration by an appropriate authority.
node: Action. Publish the policy, including the activation threshold and the appeal route, before any surge. A mechanism whose existence is disclosed only when it is invoked cannot demonstrate that it was designed impartially.
node: Professional guidance. The Society of Critical Care Medicine's guidelines on the allocation of critical care resources to adults during crisis-level shortages set out that allocation under crisis standards depends on a documented system of triage established in advance and applied consistently.
node: Model policy. Published model policy on the allocation of scarce critical care resources during a public health emergency provides that triage decisions are not made by the patient's treating clinicians, but by a designated triage team, in order to secure consistency and to separate the person deciding access from the person providing care.
node: Review trigger. If the queueing analysis shows a single group team can meet referral volume at the modelled peak, reconsider the group-wide option. It is better on staffing and secures consistency across sites without depending on joint training; the unconfirmed timing is the principal reason it is not recommended.
node: Model policy. The same published framework provides that triage arrangements are enacted only when critical care capacity is, or will shortly be, overwhelmed despite appropriate surge measures, and when a relevant authority has declared a public health emergency. Both conditions are required; neither alone is sufficient.
node: Review trigger. Review after any activation, and after any annual rehearsal, against what the teams actually did. A framework adopted and never examined against practice is a document rather than a mechanism.
node: Ethical guidance. Published guidance on the ethical allocation of scarce resources grounds such frameworks in the duty to care, the duty to steward resources, distributive and procedural justice, and transparency, and requires decision-making that is consistent, impartial, neutral and non-discriminatory. Procedural justice is the reason the mechanism, and not only the criteria, is the subject of this decision.
node: Review trigger. Review if the published professional guidance this policy follows is revised. The mechanism's principal justification is that it reflects that guidance, so a change to it is a change to the basis of this recommendation.
node: Research. Published work on frameworks for critical care triage during a major surge in critical illness describes reassessment of allocations already made as a component of the framework rather than an addition to it, since an allocation made at one point in a surge may not remain appropriate as circumstances change.
node: Assumption. A surge sufficient to exhaust conventional capacity is possible but not probable in any given year. Every option is therefore assessed on the staff it must keep trained and ready as well as on how it performs when used, because it will spend most of its life not being used.
node: Assumption. Staff trained for a triage role remain available to perform it during a surge. This is the least secure assumption in the decision: a surge that exhausts critical care capacity is also a surge that places demands on senior clinicians elsewhere.
node: Risk. Allocation arrangements may be challenged as discriminatory, as several published model policies were in 2020 and revised afterwards. Arrangements adopted in advance by the governing body, with a recorded rationale, published criteria and an independent appeal, are easier to examine and defend than decisions made case by case without a record.
node: Retain bedside clinical judgement with no standing mechanism
node: Publish criteria and an activation threshold, applied by the treating clinician
node: Standing policy with a designated triage team at each site
node: Standing policy with a single triage team across the group, with capacity balanced between sites
node: Recommendation. Adopt a standing policy with a designated triage team at each site, activated on the dual threshold of exhausted surge capacity and a declaration by an appropriate authority. It follows the published frameworks rather than departing from them, separates the decision about access from the clinician providing care, and carries reassessment and appeal. The group-wide team is not recommended at this stage principally because its time to allocation is unconfirmed, and a mechanism that becomes the bottleneck during a surge fails at the one moment it exists for.