Baseline Frailty as a Predictor of Survival After Critical Care

Abstract

BackgroundThe extent to which the degree of baseline frailty, as measured using standardized multidimensional health assessments before hospital admission, predicts survival among older adults after admission to an ICU remains unclear.Research QuestionIs baseline frailty an independent predictor of survival among older adults receiving care in an ICU?Study Design and MethodsRetrospective cohort study of community-dwelling older adults (age, ≥ 65 years) receiving public home services who were admitted to any ICU in Ontario, Canada, between April 1, 2009, and March 31, 2015. All individuals underwent an interRAI Resident Assessment Instrument—Home Care (RAI-HC) assessment completed within 180 days of ICU admission. These assessments were linked to hospital discharge abstract records. Patients were categorized using frailty measures each calculated from the RAI-HC: a classification tree version of the Clinical Frailty Scale; the Frailty Index—Acute Care; and the Changes in Health, End-Stage Disease, Signs, and Symptoms Scale. One-year survival models were used to compare their performance. Patients were stratified based on the receipt of mechanical ventilation in the ICU.ResultsOf 24,499 individuals admitted to an ICU within 180 days of a RAI-HC assessment, 26.4% (n = 6,467) received mechanical ventilation. Overall, 43.0% (95% CI, 42.4%-43.6%) survived 365 days after ICU admission. In general, among the overall cohort and both mechanical ventilation subgroups, mortality hazards increased with the severity of baseline frailty. Models predicting survival 30, 90, and 365 days after admission to an ICU that adjusted for one of the frailty measures were more discriminant than reference models that adjusted only for age, sex, major clinical category, and area income quintile.InterpretationSeverity of baseline frailty is independently associated with survival after ICU admission and should be considered when determining goals of care and treatment plans for people with critical illness. The extent to which the degree of baseline frailty, as measured using standardized multidimensional health assessments before hospital admission, predicts survival among older adults after admission to an ICU remains unclear. Is baseline frailty an independent predictor of survival among older adults receiving care in an ICU? Retrospective cohort study of community-dwelling older adults (age, ≥ 65 years) receiving public home services who were admitted to any ICU in Ontario, Canada, between April 1, 2009, and March 31, 2015. All individuals underwent an interRAI Resident Assessment Instrument—Home Care (RAI-HC) assessment completed within 180 days of ICU admission. These assessments were linked to hospital discharge abstract records. Patients were categorized using frailty measures each calculated from the RAI-HC: a classification tree version of the Clinical Frailty Scale; the Frailty Index—Acute Care; and the Changes in Health, End-Stage Disease, Signs, and Symptoms Scale. One-year survival models were used to compare their performance. Patients were stratified based on the receipt of mechanical ventilation in the ICU. Of 24,499 individuals admitted to an ICU within 180 days of a RAI-HC assessment, 26.4% (n = 6,467) received mechanical ventilation. Overall, 43.0% (95% CI, 42.4%-43.6%) survived 365 days after ICU admission. In general, among the overall cohort and both mechanical ventilation subgroups, mortality hazards increased with the severity of baseline frailty. Models predicting survival 30, 90, and 365 days after admission to an ICU that adjusted for one of the frailty measures were more discriminant than reference models that adjusted only for age, sex, major clinical category, and area income quintile. Severity of baseline frailty is independently associated with survival after ICU admission and should be considered when determining goals of care and treatment plans for people with critical illness. Clinical frailty is a state characterized by a reduced ability to maintain or restore physical, psychological, or cognitive homoeostasis when subject to stressors.1Clegg A. Young J. Iliffe S. Rikkert M.O. Rockwood K. Frailty in elderly people.Lancet. 2013; 381: 752-762Abstract Full Text Full Text PDF PubMed Scopus (4217) Google Scholar In response to the COVID-19 pandemic, some jurisdictions2Joebges S. Biller-Andorno N. Ethics guidelines on COVID-19 triage—an emerging international consensus.Critical Care. 2020; 24201Crossref Scopus (65) Google Scholar and several professional societies3Rockwood K. Theou O. Using the Clinical Frailty Scale in allocating scarce health care resources.Can Geriatr J. 2020; 23: 210-215Crossref PubMed Google Scholar,4Rockwood K. A global clinical measure of fitness and frailty in elderly people.CMAJ. 2005; 173: 489-495Crossref PubMed Scopus (3927) Google Scholar recommend using frailty (instead of chronological age) as part of a holistic assessment to identify people who are more likely or less likely to benefit from critical care.3Rockwood K. Theou O. Using the Clinical Frailty Scale in allocating scarce health care resources.Can Geriatr J. 2020; 23: 210-215Crossref PubMed Google Scholar Should surging demand for critical care be overwhelming, organizations like the United Kingdom’s National Institute for Health Care and Excellence have recommended that a Clinical Frailty Scale (CFS)4Rockwood K. A global clinical measure of fitness and frailty in elderly people.CMAJ. 2005; 173: 489-495Crossref PubMed Scopus (3927) Google Scholar score of ≥ 5 be the threshold for considering critical care inappropriate among adults.5National Institute for Health and Care Excellence (NICE). COVID-19 rapid guideline: critical care in adults. March 20, 2020. Accessed October 15, 2021. https://www.ncbi.nlm.nih.gov/books/NBK566886/Google Scholar,6Valiani S. Terrett L. Gebhardt C. Prokopchuk-Gauk O. Isinger M. Development of a framework for critical care resource allocation for the COVID-19 pandemic in Saskatchewan.CMAJ. 2020; 192: E1067-E1073Crossref PubMed Scopus (5) Google Scholar To date, most research studying the association between baseline frailty and mortality after ICU admission have scored frailty using the CFS at ICU admission from patient and proxy interviews and medical record review.7Bagshaw S.M. Stelfox H.T. McDermid R.C. et al.Association between frailty and short- and long-term outcomes among critically ill patients: a multicentre prospective cohort study.CMAJ. 2013; 186: E95-E102Crossref PubMed Scopus (297) Google Scholar,8Brummel N.E. Bell S.P. Girard T.D. et al.Frailty and subsequent disability and mortality among patients with critical illness.Am J Respir Crit Care Med. 2017; 196: 64-72Crossref PubMed Scopus (153) Google Scholar Although established as feasible, reliable, and predictive when scored in this way,9Pugh R.J. Battle C.E. Thorpe C. et al.Reliability of frailty assessment in the critically ill: a multicentre prospective observational study.Anaesthesia. 2019; 74: 758-764Crossref PubMed Scopus (33) Google Scholar the reliability of the assessment may be suboptimal when applied by intensivists compared with geriatricians,10Surkan M. Rajabali N. Bagshaw S. Wang X. Rolfson D. Interrater reliability of the Clinical Frailty Scale by geriatrician and intensivist in patients admitted to the intensive care unit.Can Geriatr J. 2020; 23: 235-241Crossref PubMed Google Scholar suggesting the potential for bias during point-of-care discussions about prognosis and goals of care. Prognostic estimation is particularly important for triage, should a surge in demand for critical care exceed system capacity. In Canada and elsewhere, interRAI standardized and multidimensional health assessments are mandated in publicly funded home care and residential long-term care.11Carpenter I. Hirdes J.P. Using interRAI assessment systems to measure and maintain quality of long-term care.A Good Life in Old Age? Monitoring and Improving Quality in Long-Term Care. OECD Publishing, Paris2013: 93-139Google Scholar In Ontario, these assessments are completed by trained home care coordinators and measure numerous domains of health and wellbeing, including functional and cognitive performance, mood, behavior, and pain.12Morris J.N. Fries B.E. Steel K. et al.Comprehensive clinical assessment in community setting: applicability of the MDS-HC.J Am Geriatr Soc. 1997; 45: 1017-1024Crossref PubMed Scopus (441) Google Scholar Information from these home care assessments is used to guide care planning and to inform resource allocation at the client level,13Heckman G.A. Gray Leonard C. Hirdes J.P. Addressing health care needs for frail seniors in Canada: the role of interRAI instruments.CGS Journal of CME. 2013; 3: 8-16Google Scholar,14Sinn C.-L.J. Jones A. McMullan J.L. et al.Derivation and validation of the Personal Support Algorithm: an evidence-based framework to inform allocation of personal support services in home and community care.BMC Health Serv Res. 2017; 17775Crossref Scopus (6) Google Scholar while quality and to support research and at the system et of care in home care using the interRAI Care Quality Scopus Google C. Fries B.E. J.N. should be used for PubMed Scopus Google Scholar To frailty assessment in the particularly for triage, from assessments completed before a measure of baseline D. Bagshaw S. D. Hirdes J. frailty measures in the Frailty Google Scholar Frailty measures that be calculated the Frailty Index—Acute Care M. et al.Derivation of a frailty from the interRAI care Scopus Google M. Gray A. Rockwood K. Frailty at admission to hospital predicts 2017; PubMed Scopus Google Scholar and a classification tree version of the O. Rockwood K. A classification tree to with of the Clinical Frailty PubMed Scopus Google Scholar study mortality hazards within of admission to an ICU stratified by severity of baseline frailty and receipt of mechanical ventilation in the ICU among community-dwelling older adults receiving home care services in Ontario, the of these baseline frailty measures at to their potential for discussions and treatment planning in critical care. a cohort study of community-dwelling older adults (age, ≥ 65 years) receiving publicly home care in Ontario, Canada who were admitted to an ICU between April 1, 2009, and March 31, in with of Personal Health Information Ethics for by the of Ethics used linked from the including the Care National Care and the Care in of hospital care and older adults receiving home care Hirdes J.P. et among home care associated with or a cohort Health Serv Res. 2013; Scopus Google Hirdes J.P. Frailty of home care with and in the of Am 2020; Full Text Full Text PDF PubMed Scopus Google Scholar including home care client outcomes after ICU A. et al.Frailty is associated with at home after critical a Care Med. 2020; PubMed Scopus Google Scholar these were linked using The clinical assessment on home care The are with the interRAI Resident Assessment Instrument—Home I. Hirdes J.P. Using interRAI assessment systems to measure and maintain quality of long-term care.A Good Life in Old Age? Monitoring and Improving Quality in Long-Term Care. OECD Publishing, Paris2013: 93-139Google J.N. Fries B.E. Steel K. et al.Comprehensive clinical assessment in community setting: applicability of the MDS-HC.J Am Geriatr Soc. 1997; 45: 1017-1024Crossref PubMed Scopus (441) Google Scholar In Ontario, during the study the RAI-HC assessment completed at home care and to or in the of a in health for who were to and personal support services for days or Hirdes J.P. et home care and in Am 2020; Full Text Full Text PDF PubMed Scopus Google Scholar RAI-HC assessment and by to the interRAI Care assessment in April and assessment the of in Scopus Google Scholar The reliability of the on the J.N. Fries B.E. Steel K. et al.Comprehensive clinical assessment in community setting: applicability of the MDS-HC.J Am Geriatr Soc. 1997; 45: 1017-1024Crossref PubMed Scopus (441) Google Scholar and the reliability of used in measures for and of J.N. I. K. Jones measures for with home care J Google Scholar are The reliability of admission and discharge and most in the is to D. C. et Institute for Health Information A Institute for Clinical Scholar as is ICU J. intensive care in Full Text Full Text PDF PubMed Scopus Google Scholar All patients in the cohort were admitted to an ICU within 180 days of the RAI-HC assessment patients who were admitted to during hospital or a admission for assessment, or patients admitted to an ICU more than during the of the ICU admission that the and To for the that patients in the cohort were admitted to for in the of critical using cohort The cohort of patients the patients who were from the ICU in less than The cohort of patients with a ICU of of for both and based on receipt of mechanical ventilation. using of Health from the record A study of the that of Health for mechanical ventilation were both and Institute for Health Quality of the Institute for Health Scholar ventilation are mandated in used the of Health extent to identify patients who received mechanical ventilation for more than frailty measures were calculated using from the most RAI-HC assessment before ICU admission. Although the CFS is used in hospital S. Rockwood K. Theou O. A of the Clinical Frailty 2020; Scopus Google Scholar associated with the pandemic in frailty K. care in PubMed Scopus Google Scholar a CFS classification tree to and reliability of O. Rockwood K. A classification tree to with of the Clinical Frailty PubMed Scopus Google Scholar The CFS classification tree scored using RAI-HC assessment In CFS from the classification tree may be based on clinical K. care in PubMed Scopus Google Scholar with the used patients in the cohort were from score of to score of these a for the to on the M. et al.Derivation of a frailty from the interRAI care Scopus Google M. Gray A. Rockwood K. Frailty at admission to hospital predicts 2017; PubMed Scopus Google Scholar Frailty of of A. et al.Frailty is associated with at home after critical a Care Med. 2020; PubMed Scopus Google et and health of frailty in a older home care a of Scopus Google Scholar of and categorized individuals using of used the Changes in Health, End-Stage Disease, Signs, and Symptoms Scale Although as a frailty J.P. The a measure to mortality in older Am Geriatr Soc. PubMed Scopus Google Scholar in predicting and mortality in home care et and health of frailty in a older home care a of Scopus Google Scholar from health score of to health score of of survival 365 days after ICU admission, stratified by receipt of mechanical ventilation in the ICU. To the of discharge in the and The of used when in both a community or and a hospital or were to with a to that of or home care discharge both and to each survival to March 31, 2015. The reliability of for who is J. et the measure A validation Health Serv Res. Scopus Google to for the of in one of the hospital among people who were home care at the of age, sex, Scale J.N. Fries B.E. et PubMed Scopus Google Scholar before hospital admission, of before hospital admission, and home care with from the RAI-HC assessment to ICU admission. were using the RAI-HC Hirdes J.P. G.A. S. for in interRAI assessments in home home and health care a Health Serv Res. 2013; Scopus Google Scholar both and used in the days before the RAI-HC assessment The most were from the abstract record using the Institute for Health S. C. of for and Health Google Scholar The using from et et and the and score for in hospital discharge using from J 173: PubMed Scopus Google Scholar calculated using abstract as survival stratified by each of the frailty measures of at 30, 90, and 365 days from the of ICU admission. were used to the survival hazards models were used to mortality hazards among of the CFS classification the and adjusted for age, sex, major clinical based on hospital and area income quintile. health or to with the used in the CFS classification tree and the the using the and the a and to Scholar All the frailty measures that were among the overall cohort and both mechanical ventilation were used in the of for used of to compare the in area the at and from the of ICU admission a reference a frailty the and and a using the These were for both mechanical ventilation In this the is the that the of a of and individuals at a are in medical and 2017; Scopus Google predictive and 2005; PubMed Scopus Google Scholar Models with than the reference that the baseline frailty measure at the All were and used a threshold of to All the of were using version The were calculated in version for using the The using the 24,499 community-dwelling older adults receiving home care who underwent a RAI-HC assessment within 180 days before ICU admission. The from the RAI-HC assessment to ICU admission days in the of the cohort received mechanical ventilation for or of patients received mechanical for more than of the and the = (n = (n = from RAI-HC assessment to ICU admission, Scale score or to to before admission or or of before care services before admission Personal support at hospital admission or and with major of on mechanical care are as or to each of the in the RAI-HC = Resident Assessment Instrument—Home Care. in a are as or to each of the in the RAI-HC = Resident Assessment Instrument—Home Care. The CFS classification tree most patients as with or frailty before ICU admission. most patients of Patients were among from health to health of the patients in the cohort a score of more than to health These by receipt of mechanical ventilation of the and the = (n = (n = Frailty Scale to with 5 with with with with Index—Acute Care ≥ health health health health health 5 health are as = Changes in Health, End-Stage Disease, Signs, and Symptoms Scale. in a are as = Changes in Health, End-Stage Disease, Signs, and Symptoms Scale. on survival 43.0% (95% CI, 42.4%-43.6%) survived to after ICU admission patients in the (95% CI, survived to this survival to (95% CI, for patients who received mechanical ventilation survival for the overall cohort stratified by each of the baseline frailty measures are in These survival are for both mechanical ventilation in for were less than both the overall cohort and both mechanical ventilation subgroups, patients who were and frail mortality hazards as the reference of frail with patients who were frail mortality hazards increased with the severity of baseline frailty in both the overall cohort and the mechanical ventilation among patients who received mechanical for the patients who were ill CI, before admission to the ICU for Models 365 ICU = (n = (n = (95% (95% (95% Frailty Scale to with 5 with with with with Index—Acute Care health health health health health 5 health models adjusted for age, sex, major clinical category, and income quintile. All adjusted for and the models (n = = Changes in Health, End-Stage Disease, Signs, and Symptoms Scale. in a All models adjusted for age, sex, major clinical category, and income quintile. All adjusted for and the models (n = = Changes in Health, End-Stage Disease, Signs, and Symptoms Scale. the overall cohort and the mechanical ventilation for patients with of the mortality hazards increased with severity of frailty. for the mechanical ventilation that were only among patients with of hazards increased with in the models for the overall cohort and the of patients who mechanical ventilation. hazards were for patients with health and health in the for the that received mechanical ventilation CI, the for the patients with health CI, All hazards models were the overall the that adjusted for the CFS than the reference at each for models that adjusted for the and at the and A among the of patients who mechanical that the that adjusted for score than the reference at the patients who received mechanical the for models that adjusted for baseline frailty to the reference at the at the models for CFS and the reference 365 this for models that adjusted for one of the frailty measures the for Models at and ICU = (n = (n = Frailty Index—Acute Frailty Index—Acute Frailty Index—Acute models for age, sex, major clinical category, and income quintile. All calculated using = area the = Changes in Health, End-Stage Disease, Signs, and Symptoms Scale. in a All models for age, sex, major clinical category, and income quintile. All calculated using = area the = Changes in Health, End-Stage Disease, Signs, and Symptoms Scale. All of that in hospital were in the of in the the record that the client for The survival at 30, 90, and 365 days after ICU admission between the study cohort and the The adjusted mortality hazards for each baseline frailty measure and the were between the study cohort and the community-dwelling older adults who were admitted to with and for mechanical baseline frailty measures applied to a of interRAI assessment were to people at mortality Although the of baseline frailty for survival after ICU admission is models for and long-term survival that adjusted for baseline frailty were more discriminant than reference models that adjusted for age, sex, major clinical category, and income quintile. these that baseline frailty may inform discussions and treatment plans for older home care critical illness. Although have critical care survival between frail ≥ and study in mortality hazards between of J.N. et al.Frailty in critically ill patients in and a cohort J 2019; PubMed Scopus Google S.M. et al.Frailty and associated outcomes and resource among older ICU patients with Care Med. 2019; PubMed Scopus Google S.M. et al.Frailty and mechanical association with and Care Med. 2019; 45: PubMed Scopus Google M. S. et al.Frailty as a predictor of short- and long-term mortality in critically ill older medical Crit Care. 2020; PubMed Scopus Google J. I. et al.Frailty is associated with an increased mortality among patients in PubMed Scopus Google Scholar The in survival after ICU admission by severity of baseline frailty be used to is the that with should be for Frailty is intensive medical treatment on the of frailty J 2020; PubMed Scopus Google Scholar determining the of allocation of critical care baseline frailty may be when considered patient with survival as severity of and J. M. L. D. for treatment during the COVID-19 J 2020; Full Text Full Text PDF PubMed Scopus Google L. et for ICU from the of the of of and Care Crit Care. PubMed Scopus Google J. M. for the in the of 2020; Scopus (5) Google J.N. M. et al.Frailty and outcomes from in critical a cohort J 2020; Full Text Full Text PDF PubMed Scopus Google Scholar measures that are used as in adjusted models for ICU A. N. A global on using to intensive care Am Soc. PubMed Scopus Google Scholar and et during the COVID-19 pandemic at associated with of the of Med. 2020; 173: PubMed Scopus Google Scholar These measures the of mortality this may be Stelfox H.T. assessment in pandemic Care Med. PubMed Scopus Google Scholar and in with COVID-19 S. of the score for determining the mortality of patients with mechanical PubMed Scopus (65) Google Scholar Using the CFS for in critical care a of to score the particularly for determining an baseline health K. Theou O. Using the Clinical Frailty Scale in allocating scarce health care resources.Can Geriatr J. 2020; 23: 210-215Crossref PubMed Google Scholar CFS score by intensivists to be and M. A. et frailty in the intensive care a reliability and Crit Care. 45: PubMed Scopus Google M. and assessments of frailty in critically ill older J Crit Care. 2019; PubMed Scopus (6) Google Scholar may be by a and to that be a M. Rajabali N. Bagshaw S. Wang X. Rolfson D. Interrater reliability of the Clinical Frailty Scale by geriatrician and intensivist in patients admitted to the intensive care unit.Can Geriatr J. 2020; 23: 235-241Crossref PubMed Google Scholar clinical is to score the based on and standardized to be a and to support critical care at the of care.3Rockwood K. Theou O. Using the Clinical Frailty Scale in allocating scarce health care resources.Can Geriatr J. 2020; 23: 210-215Crossref PubMed Google Scholar cohort patients admitted to the may have some critically ill patients who before the ICU. using that are to the that some patients may have admitted to an ICU in the of critical for of home care a in cognitive or functional a Hirdes J.P. et home care and in Am 2020; Full Text Full Text PDF PubMed Scopus Google Scholar these may the severity of frailty before The of the CFS classification tree among this cohort of home care about the of these to the of older adults admitted to the ICU. based on the the severity of frailty in this cohort with the of community-dwelling older adults in J. et al.Frailty among and older for the frailty using the on PubMed Scopus Google Scholar that the from this study may be to older adults who home care Information using home care interRAI clinical assessments may be used by critical care to a more of baseline frailty, which home care on the of survival after admission to an ICU. clinical in the ICU. research should this in as residential long-term care and should to the of baseline frailty on outcomes as in and health after ICU Is baseline frailty, measured using standardized multidimensional health assessments before admission to an independent predictor of survival among older home care with critical of baseline frailty are associated independently with adjusted mortality hazards after ICU admission, and survival models that for baseline frailty are more discriminant than reference models that a baseline frailty Severity of baseline frailty should be considered when determining goals of care and treatment plans for people with critical illness. Is baseline frailty, measured using standardized multidimensional health assessments before admission to an independent predictor of survival among older home care with critical of baseline frailty are associated independently with adjusted mortality hazards after ICU admission, and survival models that for baseline frailty are more discriminant than reference models that a baseline frailty Severity of baseline frailty should be considered when determining goals of care and treatment plans for people with critical illness. L. A. is the of the of the including the and L. A. the study and and the A. A. and N. M. S. the A. A. N. M. J. K. O. and C. M. clinical to support the and of The and by The have to the J. D. in for in K. of the Clinical Frailty Scale and is for and health care. to or the A. A. A. O. C. The for on the study for on care classification and for and and for on of this are from the on and with from the Institute for Health The used in the of this study is on to the Clinical Frailty Scale using the RAI-HC assessment is as a The and be in the of the with Improving by Frailty in a with on the study in this of by et which the independent predictive of baseline frailty on the survival in critically ill older have used in the critical care for in a of systems and study the of the is to that the baseline frailty is associated independently with short- or long-term survival PDF are for in and on The a that to frailty the of survival models that baseline frailty. Although baseline frailty is more discriminant than of measures that patient the of an ICU admission likely that be in survival PDF

Topics & Concepts