COVID-19: Frailty scores and outcomes in older people

Healthcare stressors felt during the COVID-19 pandemic have allowed us to learn much about the resilience of health and social care systems. One of the early features of the international response, at least at a clinical level, was to think about the associations between frailty and outcomes from critical illness. The British Geriatrics Society worked collaboratively with critical care colleagues and various Royal Colleges and national Societies to provide guidance on critical care decision making in people with COVID-19.

At the time the original guidance was published, there were no studies specifically examining the link between frailty and COVID-19 outcomes in the acute care setting, but over the past year and a half, there have been many examining outcomes in this population. This page was originally acting as a repository of related work, but with the slowing of publications and recent systematic reviews summarising the literature, the page will now act as a reflection upon the accumulated body of work. Specifically, the page will focus on systematic reviews reporting mortality as related to individuals with frailty diagnosed with COVID-19 in acute hospital settings, as well as pertinent related literature.

If you have any papers or data that you think should be published here, please get in touch with the curators, Professor Simon Conroy (spc3@le.ac.uk) or Kevin Boreskie (kboreskie@gmail.com). We are also happy to receive any commentary or musing related to this emerging body of evidence.

Thank you

What have we learnt?

The totality of the body of evidence examining the role of frailty in predicting outcomes related to COVID is extraordinary, and supports frailty - mostly using the Clinical Frailty Scale - as a useful tool to identify those at risk of poor outcomes. The evidence base has matured from single centre studies through to national and international multicentre evaluations. What we do not know is how it was applied in practice and how frailty influences clinical decision making. The origins of the frailty construct were in geriatricians' desire to recognise the heterogeneity of the older population, as opposed to using age as a crude measure of prognosis or to inform care planning. A criticism of many of the frailty/COVID studies is that they did not fully apply the spectrum of frailty in their estimates of prognosis. Frailty assessment and scaling allows a more nuanced and evidence-based assessment of an individual's baseline, which should inform what it is likely to be possible to achieve in the face of an illness, but such plans should always be accompanied by discussion - a shared decision making process. Frailty or indeed any other measure, should not be used in isolation to direct clinical decision making.

We have now seen that frailty is widespread in clinical practice, which is welcome. There is emerging work assessing if the frailty construct is being applied ‘correctly’, and that the clinical response is suitably holistic, in order to improve patient outcomes.

Clinical Frailty Scale

The Clinical Frailty Scale (CFS) was the tool suggested by NICE and the BGS early on in the pandemic to assist with critical care decision making given its ease of use and rapid assessment.¹ This tool was validated for use in those 65 years of age and older based on clinical judgement and functional status.² The CFS uses a 9-item scale ranging from more fit and managing well (1-3), those living with mild/moderate frailty (4-6), to those living with more advanced frailty (7-8), and terminally ill patients (9). Use of this scale is based on patient mobility, function, and cognition from two weeks prior to presentation.

Canada: Using the CFS in allocating scarce healthcare resources: https://cgjonline.ca/index.php/cgj/article/view/463

Canada: A classification tree developed to assist in CFS scoring: https://academic.oup.com/ageing/advance-article/doi/10.1093/ageing/afab006/6144822

UK: A mobile application to developed by the NHS to support clinical use of the CFS: https://apps.apple.com/gb/app/clinical-frailty-scale-cfs/id1508556286

Canada: Aging Innovation in Perioperative Medicine & Surgery (AIMS) developed online CFS training resource:  https://rise.articulate.com/share/deb4rT02lvONbq4AfcMNRUudcd6QMts3#  

Systematic Reviews

Meta-Analysis: Association of frailty with outcomes in individuals with COVID-19: a living review and meta-analysis, Dumitrascu et al. (May 28, 2021):³ https://doi.org/10.1111/jgs.17299

Measurements: 52 studies covering (14 prospective, 37 retrospective, 1 ecological) 118,373 patients. Average study age range was 56-87 years. Three of these studies were in a hospital setting and two were in long term care facilities. 80.7% of included studies used the CFS. Quality of the included studies varied with substantial heterogeneity.

Results: Frailty using CFS was significantly associated with mortality in COVID-19 patients (OR 1.79, 95% CI 1.49-2.14; HR 1.87, 95% CI 1.33-2.61) even after including adjustments from included studies. All studies adjusted for age, and most adjusted for comorbidity and sex. Some studies included additional adjustments for illness severity, laboratory values and socioeconomic status. CFS score was associated with unadjusted increased odds of delirium (OR 2.91, 95% CI 0.08-0.71), and unadjusted reduced odds of ICU admission (OR 0.24, 95% CI 0.08-0.71).

Conclusion: Frailty was associated with increased COVID-19 mortality risk and delirium, thus it may assist with patient prognosis and care. Dumitrascu et al. identify a need for data examining patient-reported outcomes to ensure relevance to the study population.

Meta-Analysis: Clinical frailty scale as a point of care prognostic indicator of mortality in COVID-19: a systematic review and meta-analysis, Kastora et al. (May 23, 2021):⁴ https://doi.org/10.1016/j.eclinm.2021.100896

Measurements: 34 studies for systematic review and 17 for meta-analysis with high heterogeneity. Average patient age ranged from 56-86 for a total cohort of 18,042 patients.

Results: Patients with CFS scores of 4-5 had increased odds of mortality (OR 1.95, 95% CI 1.32-2.87) as compared to those with CFS scores of 1-3. Those with CFS scores of 6-9 had even higher odds of mortality (3.09, 95% CI 2.03-4.71) as compared to the same reference. Furthermore, this review also identified age, sex and comorbidity as being associated with increased risk of COVID-19 mortality.

Conclusion: Even those living with mild or moderate frailty appear to be at increased risk of COVID-19 mortality, thus more comprehensive assessment and management is called for in these populations beyond those living with more advanced frailty. Knowledge of frailty status, along with additional demographic and physiological variables, may benefit outcomes for patients with COVID-19.

Meta-Analysis: The impact of frailty on COVID-19 outcomes: a systematic review and meta-analysis of 16 cohort studies, Yang et al. (May 3, 2021):⁵ https://doi.org/10.1007/s12603-021-1611-9

Conclusion: Frailty was associated with increased risk of mortality, illness severity, ICU admission, ventilation and longer length of hospital stay in patients with COVID-19. Frailty assessment may assist in the management of COVID-19 as well as resource allocation.

Meta-Analysis: Frailty as a predictor of mortality among patients with COVID-19: a systematic review and meta-analysis, Zhang et al. (March 17, 2021).⁶ https://doi.org/10.1186/s12877-021-02138-5

Conclusion: Frailty independently predicts mortality risk in COVID-19 patients. The authors recommend the use of frailty assessment as a prognostic tool to be used early in screening in order to direct interventions for high-risk patients with COVID-19.

Meta-Analysis: Clinical frailty scale and mortality in COVID-19: A systematic review and dose-response meta-analysis, Pranata et al. (March 1, 2021):⁷ https://doi.org/10.1016/j.archger.2020.104324

Conclusion: This meta-analysis showed that increase in CFS was associated with increase in mortality in a linear fashion.

Meta-Analysis: Association of frailty and mortality in patients with COVID-19: a meta-analysis, Kow et al. (March, 2021):⁸ https://doi.org/10.1016/j.bja.2020.12.002

Conclusion: Kow et al. found increased risk of mortality with more advanced frailty status in patients with COVID-19. Thus, they recommended the use of the CFS or other validated frailty tools to assist with prioritising allocation of critical care resources for COVID-19 patients.

Systematic Review: What is the relationship between validated frailty scores and mortality for adults with COVID-19 in acute hospital care? A systematic review, Cosco et al. (January 14, 2021):⁹ https://doi.org/10.1093/ageing/afab008

Conclusion: Most studies showed that more advanced frailty led to increased risk for COVID-19 mortality, but the authors note that one should not place too much emphasis on frailty alone when examining prognosis of older adults with COVID-19.

Meta-Analysis: Prevalence of frailty in patients with COVID-19: a meta-analysis, Kow et al. (December 2020):¹⁰ https://doi.org/10.14283/jfa.2020.70

Conclusion: Frail individuals may be overrepresented among the COVID-19 patient population and given a rather strong hint that the presence of frailty may lead to a higher risk of acquisition of COVID-19.

Other resources and publications

UK: NICE COVID-19 rapid guidelines covering management of COVID-19 in all care settings: https://www.nice.org.uk/guidance/ng191

UK: Ceiling of care decisions were made for the majority of patients during the COVID-19 pandemic, broadly in line with known predictors of poor outcomes in COVID-19, but with a focus on comorbidities suggesting ICU admission might not be a reliable end-point for observational studies where advanced care planning is routine¹¹. https://pubmed.ncbi.nlm.nih.gov/33430850/

Italy: Most COVID-19 survivors exhibit substantial frailty and require continuing care after discharge from acute care¹². https://pubmed.ncbi.nlm.nih.gov/33434372/

UK: Important paper that illustrates that treatments have changed over time for people hospitalised with COVID-19, which may in turn affect the performance of predictive scores¹³. https://www.sciencedirect.com/science/article/pii/S2213260020305798?via%3Dihub

UK: HFRS and CCI constructed from medical records concurrent with the start of the pandemic can be used in COVID-19 mortality risk stratification at the population level, but they show limited added value in COVID-19 inpatients¹⁴. https://agsjournals.onlinelibrary.wiley.com/doi/10.1111/jgs.17089

UK: COVID-related mortality was associated with increasing age; most deaths were anticipated and occurred in patients with advance decisions on ceilings of treatment¹⁵. https://pubmed.ncbi.nlm.nih.gov/33557238/ A helpful paper that starts to address the issue as to whether it is frailty or the system response to frailty that drives mortality? Supported by the work Straw et al, which concluded ‘Ceiling of care decisions were made for the majority of patients during the COVID-19 pandemic, broadly in line with known predictors of poor outcomes in COVID-19, but with a focus on co-morbidities suggesting ICU admission might not be a reliable end-point for observational studies where advanced care planning is routine.’¹¹ https://bmcpalliatcare.biomedcentral.com/articles/10.1186/s12904-021-00711-8

UK: An interview study with members of the public addressing the ethics of decision-making identified three core principles that should govern triage decisions: equality, efficiency and vulnerability¹⁶. https://jme.bmj.com/content/early/2021/03/08/medethics-2020-107071.