9. Frailty Assessment in Critical Illness

Quick Recap

Cross-cutting protocol — companion to Goals of Care & Palliative Care Integration and ICU Discharge Criteria & Step-Down protocols. Addresses frailty as a distinct, increasingly important risk-stratification concept in critical illness, grounded in the large multinational VIP (Very Old Intensive Care Patients) research program, which has consistently found frailty — not chronological age alone — to be the dominant predictor of ICU outcome in older patients.

1. Definition

Frailty: a state of increased vulnerability to physiological stressors resulting from cumulative decline across multiple organ systems, leading to diminished homeostatic reserve — distinct from chronological age itself, disability, or comorbidity, though frequently overlapping with all three.

Clinical Frailty Scale (CFS): a 9-point global clinical judgment scale (1 = "very fit" to 9 = "terminally ill"), based on evaluation of mobility, energy, physical activity, and function — the most widely used frailty assessment tool in critical care research, used in 60.3% of studies assessing frailty on ICU admission per systematic review, ahead of the Frailty Index (8.6%) and Fried's frailty phenotype (10.3%). A key practical feature: it is the only major frailty scale that allows assessment via caregiver interview, since it is anchored to the patient's premorbid status approximately two weeks prior to the acute illness — making it uniquely applicable to sedated, encephalopathic, or otherwise non-interviewable ICU patients, unlike scales requiring direct patient performance testing.

Threshold definition: frailty is most commonly defined as CFS ≥5, distinguishing frail from non-frail (CFS 1–2) and pre-frail (CFS 3–4) patients in the research literature, though some studies use alternative cut-offs.

Distinction from age and comorbidity: frailty, chronological age, and comorbidity burden are related but non-identical constructs — a substantial body of evidence (Section 11) specifically demonstrates that frailty is a stronger, more consistent predictor of ICU outcome than chronological age alone, which has direct implications for how admission and treatment-intensity decisions should be framed (Section 22/23).

2. Pathophysiology

Frailty reflects reduced physiological reserve across multiple organ systems simultaneously — diminished capacity for the cardiovascular, respiratory, renal, immune, and neuromuscular systems to mount an adequate compensatory response to the acute physiological insult of critical illness. This is mechanistically distinct from single-organ dysfunction: a frail patient may have no single organ system failing outright at baseline, yet lack the aggregate reserve to tolerate the compounding physiological stress of critical illness and its treatments (mechanical ventilation, vasopressors, RRT) in the way a non-frail patient of similar age and comorbidity burden would.

This explains why frailty assessment adds prognostic information beyond illness severity scores (SOFA, APACHE) and comorbidity counts alone — those tools capture the acute physiological derangement and chronic disease burden respectively, but not the patient's underlying reserve capacity to withstand and recover from the combined insult, which is precisely what frailty assessment is designed to capture.

3. Immediate Stabilization (ABCDE) — Frailty Assessment as Part of Initial Evaluation

Not an acute stabilization scenario; frailty assessment should be incorporated into the initial evaluation framework alongside standard severity scoring:

Checklist:

4. Focused History

• Baseline functional status, mobility, and independence in activities of daily living approximately 2 weeks prior to the acute illness (the specific anchor point for CFS assessment)
• Cognitive status at baseline
• Comorbidity burden and polypharmacy
• Prior hospitalizations and trajectory of functional decline over recent months
• Patient/family understanding of and prior discussion regarding goals of care, particularly relevant in frail patients where the risk-benefit calculus of aggressive intervention may differ from a non-frail patient

5. Comprehensive System-wise Examination

CFS assessment itself functions as a global, gestalt clinical judgment rather than a traditional organ-system examination — informed by mobility, energy level, physical activity capacity, and functional independence as reported by the patient (if able) or caregiver.

Related, complementary geriatric syndrome assessment domains (per the VIP2 study's broader framework, Section 11): cognition, activities of daily living, and comorbidity/polypharmacy burden — frailty should be understood as one component of a broader "geriatric syndrome" assessment in older critically ill patients, not necessarily sufficient in isolation.

POCUS integration: not a direct component of frailty assessment; muscle ultrasound for sarcopenia assessment is an emerging, adjunct research tool but not yet standard clinical practice for frailty determination.

6. Syndrome Identification — Reframed as Frailty Classification

• Non-frail (CFS 1–2): standard risk stratification applies; frailty status does not meaningfully modify prognosis beyond standard severity scoring
• Pre-frail (CFS 3–4): intermediate risk; may warrant heightened attention to functional trajectory during and after critical illness (cross-reference Early Mobilization & Rehabilitation and Post-Intensive Care Syndrome protocols)
• Frail (CFS ≥5): independently associated with worse outcomes across multiple large studies (Section 11) — warrants explicit incorporation into shared decision-making discussions, without this alone determining a specific treatment-limitation decision

7. Differential Diagnosis — Not a Traditional Differential

Cross-reference the relevant disease-specific protocol for the acute critical illness itself; this protocol's role is risk-stratification and prognostic framing, applied alongside the diagnostic workup, not a diagnostic process in itself.

8. Severity/Risk Assessment

CFS as an independent outcome predictor: an individual-patient-data meta-analysis specifically evaluating CFS for mortality prediction in acutely admitted older ICU patients found that in patients ≥65 years, frailty (CFS ≥5) was independently associated with ICU mortality even after adjustment (HR 1.34, 95% CI 1.25–1.44, p<0.0001) — a robust, adjusted association, not merely a crude/unadjusted correlation that disappears with proper statistical control.

Contrast with unadjusted all-comer analysis: notably, in the same meta-analysis, frailty's association with ICU mortality in the univariate, all-patient analysis (not restricted to older patients) did not remain significant after adjustment — an important nuance suggesting frailty's independent prognostic value may be most robust specifically in the older ICU population where it has been most extensively studied, rather than uniformly across all ICU age groups.

VIP1 study (Flaatten et al., Intensive Care Med 2017, >5,000 patients ≥80 years, 311 European ICUs): frailty (CFS) was highly associated with 30-day mortality and with the level of care provided — the foundational large-scale study establishing frailty's prognostic dominance in this population.

VIP2 study (Guidet et al., Intensive Care Med 2020, ~4,000 emergency admissions ≥80 years): examined the combined contribution of frailty, cognition, activities of daily living, and comorbidities — frailty remained the best single predictor of outcome even when assessed alongside these other geriatric syndrome domains, reinforcing CFS's particular prognostic value relative to other geriatric assessment tools.

COVIP study (~4,000 patients, conducted during 2020 COVID-19 surges): examined predictors of hospital outcome in very old ICU patients with COVID-19 specifically — frailty's prognostic importance was confirmed in this specific pandemic context as well, extending the VIP program's findings to a distinct disease population.

Nonagenarian-specific Bayesian reanalysis: a recent Bayesian analysis pooling VIP1, VIP2, and COVIP registry data specifically examined whether the very oldest patients (≥90 years, nonagenarians) had meaningfully different mortality risk than octogenarians (80–89 years) — reflecting the field's move toward more nuanced, age-band-specific risk quantification beyond a single "very old" category, and toward Bayesian methods that can incorporate varying prior assumptions (non-informative, pessimistic, skeptical) given the inherent uncertainty in this domain.

9. Investigations

Not a laboratory-driven assessment; the relevant "investigation" is the CFS assessment itself, ideally supplemented with the broader geriatric syndrome domains (cognition, ADLs, comorbidity/polypharmacy) per the VIP2 framework where feasible and resources allow.

10. Point-of-Care Ultrasound — Not a Standard Component

Muscle ultrasound for sarcopenia quantification remains a research-phase adjunct rather than standard clinical practice for frailty assessment.

11. Evidence-Based Management

Frailty's Consistent Association with Worse Outcomes

• Across the VIP1, VIP2, and COVIP studies plus numerous single-center and meta-analytic studies, frailty is consistently associated with higher ICU/30-day mortality, greater post-ICU functional dependence, and greater likelihood of long-term institutionalization among survivors — this is one of the more consistently replicated findings across this literature, in contrast to several other topics in this library where findings are genuinely discordant across studies
• A systematic review/meta-analysis (Muscedere et al.) similarly confirmed frailty's association with adverse ICU outcomes across the broader (not exclusively very-old) critically ill population
• Organ support requirement and hospital-acquired infection: prospective observational data has specifically examined frailty's association with requirement for vasoactive support, mechanical ventilation, and RRT, as well as hospital-acquired infection rates — frailty appears to influence not just mortality but the trajectory of ICU-level intervention itself

Frailty vs. Chronological Age — The Central Clinical Message

• Clinical outcomes in older ICU patients are determined less by chronological age than by clinical phenotype — including frailty, comorbidity, polypharmacy, and functional/cognitive impairment — this is an explicitly stated conclusion from recent VIP-program-affiliated research and represents the central practical message of this entire body of evidence: age alone should not be used as a determinant of ICU admission or treatment intensity decisions; frailty and the broader geriatric syndrome profile provide meaningfully more prognostic information than age as a standalone variable
• This has direct implications for avoiding both undertreatment (denying appropriate ICU care based on age alone to a biologically "young" frail-negative very old patient) and overtreatment (pursuing maximally aggressive, potentially non-beneficial intervention in a severely frail patient of any age based on optimistic age-based assumptions)

Frailty in Shared Decision-Making — Integration with Goals of Care

• The ESICM has published consensus-based recommendations specifically for the management of very old patients in intensive care, explicitly motivated by the heterogeneity of this population and the prevalence of complex geriatric syndromes that classical evidence-based medicine methods (built around more homogeneous younger populations) struggle to adequately inform
• Frailty assessment should feed into, not replace, the individualized goals-of-care and shared decision-making framework detailed in the Goals of Care & Palliative Care Integration protocol — a high CFS score is prognostically important information to share with patients/families and incorporate into decision-making, not itself a categorical treatment-limitation trigger
• Time-limited trials (cross-reference Goals of Care & Palliative Care Integration protocol's Section 11 for the full evidence and structure) are a particularly relevant tool in frail patients specifically, given the genuine prognostic uncertainty this population presents — allowing a defined trial of intensive care with pre-agreed reassessment criteria, rather than either reflexive full commitment or reflexive limitation based on frailty status alone

Practical Approach

1. Assess CFS (or equivalent) within 24 hours of ICU admission, anchored to premorbid status, using caregiver interview where the patient cannot reliably self-report
2. Incorporate frailty status alongside illness severity (SOFA/APACHE) and, where feasible, broader geriatric syndrome domains (cognition, ADLs, comorbidity) into the overall prognostic picture
3. Explicitly avoid using chronological age alone as a proxy for expected outcome or treatment appropriateness — frailty status provides meaningfully more prognostic precision
4. Incorporate frailty information into shared decision-making and goals-of-care discussions (cross-reference Goals of Care & Palliative Care Integration protocol), considering a time-limited trial approach where prognostic uncertainty remains substantial despite frailty assessment
5. Recognize that frail ICU survivors face a genuinely elevated risk of functional decline and require particular attention to early mobilization (cross-reference Early Mobilization & Rehabilitation protocol) and post-discharge functional trajectory (cross-reference Post-Intensive Care Syndrome protocol), given their reduced baseline reserve to recover from ICU-acquired weakness and other PICS domains

12. Organ Support

Not a distinct organ-support topic; frailty status is a modifying/contextual factor for organ support decisions made per the relevant disease-specific and Vasopressor & Inotrope Selection & Titration, Timing of Renal Replacement Therapy Initiation, and other organ-support protocols throughout this library, rather than a separate organ-support pathway itself.

13. Disease-Specific Therapy — Not Applicable

This is a risk-stratification/prognostic framework protocol rather than a pharmacotherapy one.

14. Consultation Matrix

15. Monitoring Framework

• Admission: CFS assessed and documented within 24 hours
• During ICU stay: functional trajectory monitored alongside standard clinical monitoring, with particular attention to mobilization (cross-reference Early Mobilization & Rehabilitation protocol) given frail patients' reduced reserve
• At discharge: functional status reassessed for comparison to baseline, informing post-discharge planning and PICS risk stratification (cross-reference Post-Intensive Care Syndrome protocol)

16. ICU Bundle Checklist

17. Complications

Early:

• Frail patients show increased requirement for organ support (vasoactive agents, mechanical ventilation, RRT) and higher hospital-acquired infection rates in observational data

Late:

• Greater functional dependence and long-term institutionalization likelihood among frail survivors compared to non-frail survivors
• Frailty may progress even in previously non-frail patients following critical illness (per FORECAST pilot study data showing frailty remained static in previously-frail survivors but progressed in previously-non-frail survivors) — an important, somewhat counterintuitive finding suggesting critical illness itself can be frailty-inducing, not simply revealing of pre-existing frailty

Prevention: appropriate incorporation of frailty status into care planning; aggressive early mobilization and PICS-prevention bundle adherence given frail patients' particular vulnerability to functional decline

Rescue: not directly applicable; management follows the relevant disease-specific and supportive-care protocols, informed by frailty context

18. Escalation & De-escalation — Not Directly Applicable

Cross-reference Goals of Care & Palliative Care Integration protocol for the escalation/de-escalation framework this informs; frailty status is an input to, not a standalone trigger for, escalation or de-escalation decisions.

19. ICU Discharge Criteria (Frailty-Relevant Context)

Cross-reference ICU Discharge Criteria & Step-Down protocol. Frailty-specific consideration: functional status at discharge compared explicitly to premorbid baseline, informing both discharge destination planning and PICS risk communication (cross-reference Post-Intensive Care Syndrome protocol), given frail patients' documented elevated risk of functional decline.

20. Documentation & Medicolegal Checklist

• CFS score and assessment method (patient self-report vs. caregiver interview) documented within 24 hours of admission
• Frailty status explicitly incorporated into documented goals-of-care discussion, distinct from being used as an unstated, informal basis for treatment decisions
• Rationale for treatment intensity decisions documented with reference to the full clinical picture (illness severity, frailty, patient/family preferences), not age or frailty status alone
• Functional status at discharge compared to baseline, documented for continuity of care

21. Key Guidelines

• ESICM consensus-based recommendations for the management of very old patients in intensive care: current authoritative guidance specifically addressing this population's heterogeneity and the limitations of classical evidence-based medicine approaches for this group
• ICU-5Ms framework (Geen et al., Can J Anaesth 2022): adaptation of the geriatric "5Ms" (Mind, Mobility, Medications, Multicomplexity, Matters Most) framework specifically to critical care, offering a structured approach to applying geriatric principles in the ICU context

22. Landmark Trials

23. Controversies

• Frailty's role in ICU triage/admission decisions remains ethically and clinically sensitive: while frailty is a robust prognostic marker, using it as a categorical basis for denying ICU admission raises genuine ethical concerns distinct from its legitimate use in individualized, shared decision-making — the evidence supports frailty as one input into nuanced, patient-centered discussion (cross-reference Goals of Care & Palliative Care Integration protocol), not as a validated triage threshold for resource allocation, and conflating the two uses is a genuine risk in how this evidence gets applied in practice.
• CFS's generalizability outside the very-old population it has been most extensively studied in: the strongest, most consistently replicated evidence comes from patients ≥80 years (VIP program); the meta-analysis finding that frailty's association with mortality was not significant in an unadjusted all-comer (all-age) analysis suggests some caution is warranted before assuming identical prognostic value in younger frail patients, even though frailty as a construct is increasingly assessed across the full adult ICU age spectrum.
• Optimal frailty assessment method remains genuinely unstandardized: despite CFS's dominance in the literature (60.3% of studies), 19 different frailty assessment methods have been used across the broader research base, with heterogeneity in cut-offs and timing of assessment — a systematic review specifically concluded that no meta-analysis was possible due to this methodological heterogeneity, representing a genuine, unresolved standardization gap rather than a settled best-practice choice beyond CFS's practical popularity.
• Frailty as a consequence, not only a predictor, of critical illness: the FORECAST pilot finding that frailty can progress in previously non-frail ICU survivors complicates a purely predictive framing of frailty and suggests critical illness itself may causally contribute to frailty development — an area requiring further, larger-scale confirmation.

24. References

1. Flaatten H, De Lange DW, Morandi A, et al; VIP1 study group. The impact of frailty on ICU and 30-day mortality and the level of care in very elderly patients (≥ 80 years). Intensive Care Med. 2017;43(12):1820-1828.
2. Guidet B, de Lange DW, Boumendil A, et al; VIP2 Study Group. The contribution of frailty, cognition, activity of daily life and comorbidities on outcome in acutely admitted patients over 80 years in European ICUs: the VIP2 study. Intensive Care Med. 2020;46(1):57-69.
3. Jung C, Flaatten H, Fjolner J, et al; COVIP study group. The impact of frailty on survival in elderly intensive care patients with COVID-19: the COVIP study. Crit Care. 2021;25(1):149.
4. Bertschi D, Waskowski J, Schilling M, Donatsch C, Schefold JC, Pfortmueller CA. Methods of assessing frailty in the critically ill: a systematic review of the current literature. Gerontology. 2022;68(12):1321-1349.
5. Muscedere J, Waters B, Varambally A, et al. The impact of frailty on intensive care unit outcomes: a systematic review and meta-analysis. Intensive Care Med. 2017;43(8):1105-1122.
6. The Clinical Frailty Scale for mortality prediction of old acutely admitted intensive care patients: a meta-analysis of individual patient-level data. [journal citation per source].
7. ESICM consensus-based recommendations for the management of very old patients in intensive care. Intensive Care Med. 2025.
8. Geen O, Perrella A, Rochwerg B, Wang XM. Applying the geriatric 5Ms in critical care: the ICU-5Ms. Can J Anaesth. 2022;69(9):1080-1085.
9. The frailty, outcomes, recovery and care steps of critically ill patients (FORECAST) study: pilot study results. Intensive Care Med Exp. 2022.
10. Prognosis of nonagenarian ICU patients: a Bayesian analysis of prospective European studies. 2025.
11. The Washington Manual of Critical Care, 4th ed. 2025 — relevant geriatric critical care content.
12. ICU Protocols: A Step-wise Approach, 2nd ed. — relevant comprehensive ICU care content.