What every intensivist should know about intensive care unit admission criteria (2024)

Introduction

The Institute of Medicine (IOM) published a landmark report in 2001⁽¹⁾ recommending a thoughtful newhealth care delivery framework to improve the quality of care for the Americanpopulation in the 21^st century. The IOM defined six quality domains thatshould be at the foundation of how we deliver critical care services: safety,timeliness, efficacy, efficiency, patient-centeredness, and equitability. However,critical care resources are limited and are not available to all, raising majorconcerns regarding how these resources are allocated. There are myriad reasons forthe shortage of critical care services, which have led to the development ofpolicies to improve the utilization of, and in turn often ration, these scarceresources.⁽²⁾

Who needs to be in the intensive care unit?

The question of who requires intensive care support by a criticalcare team is complex. Critical care encompasses a broad variety of clinicalconditions across different specialties and environments. The question ofhow to allocate critical care services is also not easy toanswer; it depends upon many factors, which vary among individual institutions andinclude practitioner and resource availability among other considerations. It iseasy to over-triage with minimal consequences if there is a surplus of bedsavailable, but doing the same with a limited number of beds can have catastrophicrepercussions for those later denied intensive care.

Recently, a task force of the Society of Critical Care Medicine (SCCM) published anew set of evidence-based guidelines to be used as a framework for enhancing andguiding clinical operations; the group provided updated intensive care unit (ICU)admission, triage and discharge strategies. These were aimed at not only helpingpractitioners but also guiding research in these particularly complex subjectareas.⁽³⁾ The authorsextensively evaluated all the relevant factors that play a role in these threeprocesses, and through a systematic evidence-based approach highlighted the scarcityof evidence, which limited the strength of recommendations in most areas.

In regard to admission, there are several existing models to consider; the diagnosismodel, the objective parameters model, and the prioritization model are the mostcommonly used. In the diagnosis model, practitioners use policies with lists ofspecific conditions that merit care in their respective units to guide admissions.This is an approach that involves listing a series of pathologies requiring care inthe ICU. This approach is relatively straightforward (e.g., acute pulmonary edema)but is seldom used throughout Latin America. In the objective parameters model,specific thresholds are set whereby certain laboratory or physiologic parameterstrigger evaluation and, when certain objective criteria are met, evaluation foradmission occurs (e.g., sodium < 110 or > 170mEq/L). This is a more difficultapproach to implement because there are not well-defined criteria for all systems.This approach is mostly used in combination with the diagnosis model and is oftenused as criteria for assessment by rapid response teams. In the prioritizationmodel, patients are selected following a specially structured triage systemprioritizing patients according to their needs and likelihood of benefiting fromadmission. In addition to the above models, others have proposed classifyingpatients by matching their hospital needs, rather than by the parameters of theirillnesses. In this approach, patients are allocated to four different levels of careranging from 0 to 3 (0 - ward, 1 - telemetry, 2 - intermediate medical unit, and 3 -ICU) based on their need for monitoring and/or interventions.⁽⁴⁾

Each of these approaches has its advantages and drawbacks, and none has been fullyvalidated. The use of vital signs alone appears to be poorly specific and sensitiveas a measure, precluding their use alone.⁽⁵⁾ More comprehensive scoring systems based on physiology andco-morbidity have been developed both in general⁽⁶⁾ and in specific subpopulations, such as hematopoietictransplant patients,⁽⁷⁾ to aid inadmission decisions but have been validated only locally. Evidence for such scoringsystems is poor, and the recommendation was made to not use a scoring system aloneto determine eligibility for higher levels of care or discharge from ICU. Similarly,prognostic severity of illness scores are not recommended for use in making end oflife decisions in individual patients.

Making the decision to admit to intensive care

The decision to admit is multifaceted, encompassing many aspects of clinicalpractice. The initial criteria to be considered for admission are the need for anintervention that is not available elsewhere in the institution as well as clinicalinstability that places the patient at risk of dying or immediate deterioration.

To make recommendations on when to admit a patient, the SCCM task force created alist of the most important considerations, using aspects of the above models. Thefactors considered important by the task force included the identification ofinterventions that could be only provided in the ICU environment (e.g.,life-supportive therapies), available trained personnel to care for the patient(including nursing and physician ratios), prioritization according to the patient'scondition, clinical diagnosis, bed availability, objective parameters at the time ofreferral (e.g., elevated respiratory rate), potential to benefit from theinterventions needed, and the patient's prognosis (Table 1). Also important are timing (minimizing delays in care) andethical decision-making (avoiding discrimination and avoiding over- andunder-triage). Patients' autonomy and wishes must be respected as central to anydecision. Boarding (placing patients in beds not suited for their specialty) isassociated with worse outcomes and should be avoided.

Triage

The existence of objective and pre-defined triage criteria is an essential componentof disaster management plans, as recommended by the European Society of IntensiveCare Medicine's (ESICM) Task Force on ICU triage during an Influenza Epidemic orMass Disaster.⁽⁸⁾ Triage mustendeavor to be as accurate as possible, but the SCCM recommends that over-triage ispreferable to under-triage (that is, admitting patients to ICU who may not requireit). Although slight over-triage may be preferred, significant over-triage may bedeleterious if the service becomes overwhelmed, especially in times of crisis orextraordinarily high demand. Under-triage has been associated with increasedmortality.⁽³⁾

A Task Force of the Council of the World Federation of Societies of Intensive andCritical Care Medicine (WFSICCM) has also commented on triage decisions for ICUadmission.⁽⁹⁾ In theirrecent consensus statement, the group highlighted the role that (1) triage has inoptimizing and making equitable critical care resources available, (2) thelimitations of algorithms and protocols, (3) the importance of the collaborativeintensivist approach in making the final decision to admit, and (4) the need for theefficient and organized use of resources at local and regional levels.

The SCCM's task force delineated a specific approach to prioritizing admissionsduring the triage process, proposing 5 levels. First priority is given to patientswho are critically ill, can benefit from the intensive care/life supportinterventions in an ICU, and do not have limitations of care. Second priority isgiven to patients who are similar but have a questionable chance of benefiting frominterventions because of advanced underlying diseases reducing their long-termsurvival or those who have specific limitations of care. Third priority is given topatients who are critical but can receive their needed therapies outside the ICUenvironment, such as non-invasive ventilation in an intermediate medical unit.Fourth priority is given to patients with priority three criteria but with lowerchances of survival or with limitations of care. The fifth priority addressespatients with terminal conditions, who are moribund, or who would benefit frompalliative care rather than inappropriately aggressive or heroic interventions.

The final category includes a controversial patient population for whom consensus hasnot been achieved. Nonbeneficial care provided in the ICU continues to occur, andthere remains no universally accepted solution to this problem. The delivery of this(which some continue to inaccurately call futile) care in the ICUcontinues, and its potentially deleterious impact on others needing these resourcesremains unclear. Rationing has been undertaken in many countries with limitedresources.

Intensive care unit outreach/intensive care without walls

Because not all critically ill patients can be admitted to the ICU and some ICUadmissions can potentially be prevented, many hospitals have recently implementedrapid response systems whereby clinicians trained in critical care assess patientswho may be at risk of deteriorating on the ward and initiate suitable interventions.Rapid response systems have been shown to reduce ICU admissions and out-of-ICUcardiac arrests, though evidence for such programs has considerable methodologicalissues.⁽¹⁰⁾ The SCCM taskforce recommended using rapid response systems for the early review of acutely illpatients outside of the ICU to prevent unnecessary ICU admissions.

Critical care consult teams have been developed by several hospitals to aid indischarge and transfer to the ICU and assist in managing critically ill patients onthe ward. As with rapid response systems, structures of such systems and theirimplementation vary widely and make detailed assessment difficult. Nonetheless, theSCCM recommends the use of critical care consult teams to help facilitate admissionsand discharges from the ICU.

Conclusion

The decision to admit or triage to the ICU is a complex everyday practice that can beoverwhelming in times of high demand. We believe that by following a comprehensiveapproach based on the principles described above, any clinician can be capable ofcutting-edge decisions. These principles are based on recommendations and strategiesproposed by the IOM, SCCM, and consensus statements of the WFSICCM and ESICM. Wemust remember that ICU admission, discharge, and triage criteria are all continuallyevolving.

Regardless of the above guidance, intensivists must consider the local policies ofthe hospital and country in which they work. If such policies do not exist, it isessential to start by creating a detailed protocol in which everyone who takes careof critically ill patients knows when to call the ICU consulting team. This protocolwould ensure that admission to, triage of and discharge from intensive care aretransparent processes. These protocols must take into account the recommendationsfrom various expert bodies that have indicated methods by which to optimize theseprocesses. Once transfer to the ICU is agreed upon, it must be undertaken in a safe,efficient and timely manner to ensure high-quality patient care for all.

Footnotes

REFERENCES