Frailty, a multi-system decline in health status resulting from the cumulative impairment of homeostatic reserve (i.e., energy, physical ability, cognition), is increasingly recognized as an important prognostic determinant of outcome among critically ill populations [1,2,3]. Frailty predisposes to disproportionate and heightened vulnerability to adverse outcomes from acute stressors such as infection, surgery, or trauma. As many as one-third of adult critically ill patients admitted to the intensive care unit (ICU) have been shown to be screened as frail, with the prevalence increasing with older age [1, 2, 4].

While the prevalence of frailty may naturally increase with age, it is important to recognize that frailty and ageing are not synonymous. The accumulation of health impairments driving the development of frailty may occur at vastly different rates between individuals and across the adult lifespan [5]. Accordingly, frailty may represent a more robust predictor of vulnerability and “recoverability” than chronological age alone, particularly in the context of critical illness. Indeed, frail persons are susceptible to catastrophic multi-system organ failure from illnesses that are often tolerated by fit (not frail) persons of the same age. This may translate into greater rescue of frail patients by “intensive care” compared with contemporary age-matched fit patients. Frail patients admitted to ICU have worse outcomes, including higher mortality and among survivors, greater disability and impaired quality of life, and greater likelihood of long-term institutionalization. In this context, frailty is also increasingly recognized as a common end-of-life trajectory that is associated with considerable health resource use [6, 7].

Accordingly, we read with interest the recent findings of the prospective, observational cohort study of 5021 very old patients [median (IQR) age 84 years (81–86)] admitted to 311 European ICUs presented by Flaatten and colleagues [Very Old Intensive Care Patients (VIP Study)] [8]. The main objective of the VIP study was to evaluate the impact of frailty on ICU and 30-day mortality, along with its association with intensity of ICU support provided.

In the VIP study, frailty status was ascertained by screening patients using the Clinical Frailty Scale (CFS) score at the time of ICU admission [1, 2]. The CFS is a previously validated judgment-based tool that can be relatively simple to use to screen patients on an visual analogue and ordinal scale (range 1–9), with higher scores representing greater degrees of frailty [1, 2, 9]. In the VIP study, the CFS was assigned by bedside physicians and nurses and aimed to integrate patient information or surrogate reporting of patient status immediately prior to the current illness. Patients were classified as fit (CFS score 1–3), pre-frail (CFS score 4) or frail (CFS score 5–9), respectively.

The VIP study found 42.9% of patients aged ≥ 80 years old admitted to ICU were screened as frail. Description of the cohort further showed that frail patients were older, more likely female, had higher admission sequential organ failure assessment scores, were more likely acute admissions (unplanned) and were significantly more likely to have treatments withheld or withdrawn compared with fit or pre-frail patients. Frailty was significantly associated with greater 30-day all-cause mortality compared with those that were fit (41% for frail vs. 24% for fit; adjusted-hazards ratio 1.54; 95 CI, 1.38-1.73). Of interest, 23.8% of admitted patients did not receive any ICU-specific procedures.

The VIP study has several noteworthy findings. It builds further on prior work and affirms patients admitted to ICU can be screened for frailty using a relatively simple tool, to provide a global impression of pre-hospital function [1,2,3]. Moreover, the CFS score appears to reliably predict a subgroup more likely to die within 30 days—confirming construct validity.

Despite these insights, the VIP study has limitations that warrant some consideration. Despite being multi-center, the cohort may be susceptible to sampling bias given the variable contributions of participating countries [e.g., Great Britain ~ 20.9%, whereas others were represented by as few as six patients (< 1%)]. Likewise, there may be selection bias by not having an understanding of the relative proportion of frail patients ≥ 80 years of age denied ICU admission. In addition, there was limited adjustment in multi-variable analysis for multi-morbidity. Finally, to further build on prior work in this area, the VIP study would have been further strengthened by inclusion of confirmatory assessments of frailty status, and longer-term (post-hospitalization) outcomes including cognition, disability and disposition.

The findings of the VIP study have a number of potential implications. First, it was likely that very old patients were already carefully selected by intensivists for ICU admission, yet, despite this, an estimated two in five were screened as frail, illustrating the high prevalence in this demographic. This finding is of particular interest when juxtaposed with the fact that 23% of patients did not receive an ICU intervention (i.e., mechanical ventilation, vasoactive therapy, renal replacement therapy), although it is unclear how this interacts with the 31% of frail patients having treatments “withheld” or whether they simply were not indicated. Among those not receiving ICU interventions, it would be of interest to have insight into the reasons for ICU admission and whether this was reasonably warranted. Further work should aim to explore the implications of frailty for duration of organ support, and, perhaps due to diminished resilience, transition to a state of persistent critical illness [10]. In such circumstances, an appreciation of baseline functional and frailty status may help clinicians, patients and families navigate challenging decisions regarding offering and/or continuing advance life-support measures in ICU settings.

The burden of frailty in our ageing population is likely to increase and poses a significant health challenge [11]. The accrued evidence to date would imply frailty in ICU settings is important and that the integration of routine screening measures, such as the CFS, may add value by aiding in the identification of vulnerable patients, by providing incremental prognostic information, and by advancing research and innovation to improve patient-centred outcomes.