If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Health Research Methods, Evidence and Impact, McMaster University, 175 Longwood Rd. S, Hamilton, Ontario L8P 0A1, CanadaCollege of Nursing, Michigan State University, 1355 Bogue St, East Lansing, MI 48824, United States
Bruyère Research Institute, 85 Primrose Ave, Ottawa, Ontario 85K1R 6M1, CanadaSchool of Public Health Science, Faculty of Applied Science, University of Waterloo, 200 University Ave W, Waterloo, Ontario N2L 3G1, Canada
Department of Emergency Medicine, Queens University, Kingston, Ontario, CanadaDivision of Emergency Medicine, Department of Medicine, McMaster University, Hamilton, 1280 Main St W, Hamilton, Ontario L8S 4L8, Canada
Department of Health Research Methods, Evidence and Impact, McMaster University, 175 Longwood Rd. S, Hamilton, Ontario L8P 0A1, CanadaMcMaster Institute for Research on Aging, McMaster University. 1280 Main St W, Hamilton, Ontario L8S 4L8, Canada
Department of Health Research Methods, Evidence and Impact, McMaster University, 175 Longwood Rd. S, Hamilton, Ontario L8P 0A1, CanadaTed Rogers Centre for Heart Research, University Health Network, 661 University Ave, Toronto, Ontario, Canada
Département de Médecine Familiale et Médecine d’urgence, Faculté de Médecine, Université Laval, 1050 avenue de la médecine, Québec, Québec G1V 0A6, Canada
Department of Health Research Methods, Evidence and Impact, McMaster University, 175 Longwood Rd. S, Hamilton, Ontario L8P 0A1, CanadaSt. Joseph’s Health System, 50 Charlton Ave E, Hamilton, Ontario L8N 4A6, Canada
To evaluate the association between frailty and post-cardiac arrest survival, functional decline, and cognitive decline, among patients receiving home care.
Methods
Frailty was measured using the Clinical Frailty Scale (CFS) and a valid frailty index. We used multivariable logistic regression to measure the association between frailty and post-arrest outcomes after adjusting for age, sex, and arrest setting. Functional independence and cognitive performance were measured using the interRAI ADL Long-Form and Cognitive Performance Scale, respectively. We conducted sub-group analytics of in-hospital and out-of-hospital arrests.
Results
Our cohort consisted of 7,901 home care clients; most patients arrested out-of-hospital (55.4%) and were 75 years or older (66.3%). Most of the cohort was classified as frail (94.2%), with a CFS score of 5 or greater. The 30-day survival rate was higher for in-hospital (26.6%) than out-of-hospital cardiac arrests (5.2%). Most patients who survived to discharge had declines in post-arrest functional independence (65.8%) and cognitive performance (46.5%). A one-point increase in the CFS decreased the odds of 30-day survival by 8% (aOR = 0.92; 95%CI = 0.87–0.97). A 0.1 unit increase in the frailty index reduced 30-day survival odds by 9% (aOR = 0.91; 95%CI = 0.86–0.96). The frailty index was associated with declines in functional independence (OR = 1.16; 95%CI = 1.02–1.31) and cognitive performance (OR = 1.24; 95%CI = 1.09–1.42), while the CFS was not.
Conclusion
Frailty is associated with cardiac arrest survival and post-arrest cognitive and functional status in patients receiving home care. Post-cardiac arrest cognitive and functional status are best predicted using more comprehensive frailty indices.
Frailty is a multidimensional syndrome characterized by a heightened vulnerability to adverse health events and diminished homeostatic recovery from stressors.
The home care population is medically complex, with a high prevalence of frailty and a greater risk for cardiac arrest, given their advanced age and high rates of cardiopulmonary disease.
In Canada, individuals receiving publicly-funded home care services in many provinces are assessed using the interRAI Resident Assessment Instrument Home Care (RAI-HC),
The prognosis of cardiac arrest is known to be worse among patients receiving home care compared to the general public, similar to that of nursing home residents.
However, little is known about how frailty influences post-cardiac arrest health outcomes in this high-risk population.
Our primary objective was to examine the association between frailty and survival following cardiac arrest in patients receiving home care using two validated frailty measures. We hypothesized that frailty would be associated with post-cardiac arrest survival using both frailty measures. Our secondary objective was to examine the association between frailty and post-cardiac arrest changes in functional independence and cognitive performance.
Methods
Study design
We conducted a population-based retrospective cohort study linking multiple de-identified administrative health datasets housed within ICES, a not-for-profit organization with 75+ provincial data sets in Ontario, Canada. We were granted a waiver of ethics review from the Hamilton Integrated Research Ethics Board (HiREB), as informed consent is not required to leverage this data in accordance with Section 45 of Ontario’s Personal Health Information Protection Act. We reported our findings in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.
We identified publicly funded long-stay (>60 days) home care clients using the Home Care Reporting Dataset, a population-based dataset of Ontario, Canada. RAI-HC assessments were extracted from the Home Care Dataset. The RAI-HC contains over 250 assessment items and evaluates various health domains,
RAI-HC assessments are typically conducted within two weeks of this service enrollment and upon return to services. However, we did not have access to data on the less detailed interRAI contact assessments that sometimes occur in place of the RAI-HC upon returning to home care services. We obtained follow-up data on function independence and cognition performance from the most recent home care, long-term care, or post-acute interRAI assessment post-discharge occurring within 1-year post-cardiac arrest. Where multiple post-arrest assessments were available, we used the assessment most proximal to the arrest date, irrespective of the assessment setting.
Cardiac arrests that occurred pre-hospital or within the emergency department (ED) were identified using the National Ambulatory Care Reporting System, and cardiac arrests that occurred post-hospital admission were identified using the Discharge Abstract Database. We extracted data on patient age and sex from the Registered Persons Database and used the Vital Statistics and Death database to determine the etiology and date of death. The databases used in this study are routinely checked for quality and have been validated for clinical and health services research in Ontario and Canada.
We included all adult (≥18 years) long-stay home care clients who experienced a cardiac arrest in Ontario, Canada, between 01/01/2006 and 03/31/2018. Specifically, we included two home care populations (i) individuals who arrested pre-hospital and were transported to a hospital and those who arrested within the ED (out-of-hospital), and (ii) individuals who arrested within 72 hours of ED registration post-hospital admission (in-hospital), excluding those who received surgery in-hospital prior to their cardiac arrest.
We included ED arrests within the out-of-hospital cohort as we were unable to delineate the location of arrest from diagnostic and intervention codes available in the dataset. We selected a 72-hour time frame to mitigate the risk of health decline found with lengthy hospital admissions,
The ED registration time was utilized as the reference point for time-based measures. We excluded observations without a valid Identification Key Number and birth or death date. We also excluded patients without a RAI-HC assessment in the six-month (180-days) pre-arrest to ensure assessment data accurately depicts patient features. To identify episodes of cardiac arrest, we used a validated set of Canadian Classification of Health Interventions codes (IHZ30JN, IHZ3OJY) and International Classification of Disease codes (I46.1, I46.2, I46.8, I49.0, 149.01, I49.02, R960, R96.1, R98, R99).
We used the first event in cases where two arrests occurred within the study period.
Patient characteristics
We measured age as a categorical variable due to data privacy limitations within ICES, with years of age collapsed to 18–49, 50–64, 65–74, 75–84, and 85 + years. We measured cognitive performance using a validated measure, the Cognitive Performance Scale,
Validation of the interRAI Cognitive Performance Scale against independent clinical diagnosis and the Mini-Mental State Examination in older hospitalized patients.
with scores ranging from zero (intact) to six (very severe impairment) based on performance with decision-making, verbal expression, and short-term memory. The Cognitive Performance Scale has been validated against gold standards for cognitive assessment like the Mini-Mental State Exam, the Montreal Cognitive Assessment,
Validation of the interRAI Cognitive Performance Scale against independent clinical diagnosis and the Mini-Mental State Examination in older hospitalized patients.
We defined caregiver distress as the expression of stress, anger, or depression from a non-professional caregiver. We defined polypharmacy as taking five or more medications simultaneously.
Frailty. Frailty was measured using the Clinical Frailty Scale (CFS), and a valid frailty index derived for use within the RAI-HC. The CFS is a 9-item ordinal scale that ranges from one (very fit) to nine (terminally ill), is commonly used to predict survival post-cardiac arrest, and has been validated for retrospective calculation.
The frailty index was calculated as a health deficit accumulation model using 43 items (see supplemental file) from the RAI-HC and is predictive of mortality and admission to long-term care.
We operationalized frailty measures in their full forms to leverage the benefits of their granularity and adhere to best practices in frailty measurement.
The primary outcome for our study was 30-day survival, using the date of arrest as the reference point. Our secondary outcomes for this study include 1-year survival and declines in functional independence or cognitive performance compared to pre-arrest status. We defined a decline in physical independence as an increase in ADL Long-Form Scale from the pre-arrest score. The ADL Long-Form is the most sensitive of all interRAI scales when assessing functional independence over time, with scores ranging between 0–28 based on independence in tasks of eating, bathing, toilet use, personal hygiene, dressing, locomotion, and bed mobility.
We defined a decline in cognitive performance as an increase in the Cognitive Performance Scale from pre-arrest assessment. Those with a maximum score pre-arrest ADL Long-Form and Cognitive Performance Scale were excluded from outcomes beyond survival to avoid possible ceiling effects (see Fig. 1). Outcomes for this study are patient-important and recommended for evaluation by the Core Outcome Set for Cardiac Arrest (COSCA) initiative and the International Consortium for Health Outcome Measurement.
We reported descriptive statistics using measures of frequency and central tendency. We calculated odds ratios and 95% confidence intervals using univariable and multivariable logistic regression. Adjusted models controlled for age, sex, and arrest setting (in-hospital versus out-of-hospital) within the overall cohort. These factors can influence survival and the likelihood of bystander response to cardiac arrest.
In-hospital versus out-of-hospital cardiac arrest: Characteristics and outcomes in patients admitted to intensive care after return of spontaneous circulation.
Associations between post-cardiac arrest outcomes and the CFS were evaluated and reported using a one-point increase as the unit of measurement and a 0.1 unit increase for the frailty index. We conducted an unadjusted sub-group analysis of associations within distinct cohorts of in-hospital and out-of-hospital arrests. Missing data were scant (<1%) for pre-arrest assessment data and were deleted within each analysis. Data were managed and analyzed in R version 3.1.
Results
Our cohort contained 7,901 home care clients who experienced a cardiac arrest and had a RAI-HC assessment in the six months prior to arrest. Most arrests occurred out-of-hospital (55.5%), and one-fifth (26.0%) survived to hospital discharge. The median time between cardiac arrest and pre-arrest assessment was 66 days and 70 days between arrest and post-arrest assessment. Fig. 1 displays the patient flow diagram, highlighting exclusions and post-arrest outcomes beyond survival.
Pre-cardiac arrest, roughly one-half of patients had mild-to-severe cognitive impairment (49.9%) or needed assistance with ADLs (46.7%). One-fifth (23.1%) of patients received a CHESS score of three or greater, indicating pre-arrest health instability. The median CFS score was six, and the median frailty index score was 0.3 (range = 0–0.64). Table 1 describes the pre-arrest features for the overall cohort and compares them between in-hospital and out-of-hospital arrests.
Table 1Pre-Arrest Features in 7,901 Patients Receiving Home Care in Ontario, Canada between 2006–2018.
Fig. 2 displays a forest plot of the associations between frailty scores post-cardiac arrest outcomes, adjusting for age, triage acuity, and arrest setting. Unadjusted estimates can be found in the supplement. A total of 1,165 (14.8%) survived to 30 days post-cardiac arrest and 744 (9.8%) survived to 1 year. The 30-day survival rate was higher for those who arrested in-hospital (26.6%) than out-of-hospital (5.2%). Similarly, the 1-year survival rate was higher for in-hospital (17.6%) than out-of-hospital (2.8%) cardiac arrest. Sub-group analytics of outcomes between in-hospital and out-of-hospital arrests can be found in Table 2. A 1-point increase in the CFS resulted in a 9% reduction in the odds of 30-day survival (aOR = 0.91; 95%CI = 0.86–0.96) and a 12% reduction in the odds of 1-year survival (aOR = 0.88; 95%CI = 0.83–0.94), after adjusting for age, sex, and arrest setting. Similarly, a 0.1-unit increase in the frailty index reduced the odds of 30-day survival by 8% (aOR = 0.92; 95%CI = 0.87–0.97) and 1-year survival by 13% (aOR = 0.87; 95%CI = 0.82–0.85).
Fig. 2Association Between Frailty Scales and Survival to 30-Days after Cardiac Arrest Note: Statistics reported as odds per 1 point increase in Clinical Frailty Scale and 0.1 unit increase in frailty index.
Associations between frailty and Post-Arrest decline in functional and cognition
Of those who survived to hospital discharge, 936 (45.7%) had interRAI assessments post-discharge. Pre-arrest frailty and health instability were similar between those with (46%) and without (47%) RAI-HC assessment data post-discharge (see supplemental file). Those without assessments had greater mortality rates in the first few days post-discharge and had a lower median time to death post-discharge (8 vs 136 days), indicating those without post-cardiac arrest RAI-HC assessments likely died before revaluation post-discharge (see supplement).
Of those who survived to receive a RAI-HC assessment post-discharge, most survivors experienced a functional decline (65.8%), and roughly half (46.5%) had cognitive decline. For the out-of-hospital cohort, 67.3% of those who survived to discharge had a decline in functional independence, with similar results found among the in-hospital cohort (68.3%). Rates of cognitive decline were virtually identical between the out-of-hospital and in-hospital cohorts (46.4% versus 46.6%, respectively).
A 0.1-unit increase in the frailty index increased the odds of post-cardiac arrest functional decline by 16% (aOR = 1.16; 95%CI = 1.02–1.31) and the odds of cognitive decline by 24% (aOR = 1.24; 95%CI = 1.09–1.42), after adjusting for age, sex, and arrest setting. Upon sub-group analysis, the frailty index was only associated with functional (OR = 1.21; 95%CI = 1.04–1.41) and cognitive decline (OR = 1.38; 95%CI = 1.09–1.44) for the in-hospital cohort. The CFS was not associated with these outcomes in the primary and sub-group analytics.
Discussion
In this study, we documented that frailty, regardless of how it was measured, was independently associated with a significant decrease in 30-day and 1-year survival. These associations remained significant regardless of the arrest setting. Fewer than 15% of the overall cohort survived to 30 days, and the majority assessed post-discharge had functional or cognitive decline post-arrest (77.4%) and greater health instability (53.7%). The frailty index was associated with declines in cognition and function post-cardiac arrest in the primary analysis and within the in-hospital cohort.
Comparison to prior relevant studies
This study builds on our prior work, highlighting that patients receiving home care have a worse overall prognosis of survival post-cardiac arrest than community-dwelling older adults not receiving home care services.
Our work validates prior studies demonstrating a relationship between frailty and survival following cardiac arrest. However, most studies focus solely on in-hospital cardiac arrest, a single frailty measure, and are at greater risk for statistical fragility given the relatively small sample sizes.
Is frailty associated with long-term survival, neurological function and patient-reported outcomes after in-hospital cardiac arrest? - A Swedish cohort study.
Our study reported higher survival rates than previously reported population-based studies of in-hospital cardiac arrest. We attribute this to the exclusion of cardiac arrests after 72 hours of hospital admission, knowing health instability and decline are positively associated with hospital length of stay.
Is frailty associated with long-term survival, neurological function and patient-reported outcomes after in-hospital cardiac arrest? - A Swedish cohort study.
found no difference between frailty and cognitive status at admission and discharge for those arresting in-hospital. However, this study differed from ours in that they only included those who survived to 30 days post-cardiac arrest (i.e., healthier) and utilized the Cerebral Performance Category scale to operationalize cognition. Almost all prior resuscitation studies used the CFS to operationalize frailty, with a cut-off of five or greater used in all studies but one.
We evaluated all cut-offs of the CFS, and using a higher cut-off of six or seven significantly improved the discriminative accuracy when predicting 30-day survival (see supplemental file).
Implications
Home care clients are known to have worse health outcomes post-cardiac arrest than community-dwelling individuals not receiving home care.
A worse prognosis of survival and a high prevalence of frailty in patients receiving home care underscores the importance of pragmatic discussions and shared decision-making about end-of-life preferences upon service enrollment and during follow-up assessments. While survival rates were significantly higher for in-hospital arrests, foreknowledge of arrest setting is not available during advance care planning in the primary care setting where these discussions are most appropriate.
Most patients had functional and cognitive decline post-discharge, further emphasizing the importance of advance care planning, as the quality of life of cardiac arrest survivors is likely to be negatively influenced. This finding also showcases the importance of reassessing the functional and cognitive status, along with their care goals upon return from the hospital post-cardiac arrest. Further, reassessment of support needs (e.g., personal support, hours of care) and referrals (e.g., memory clinic, rehabilitation) should be considered by health care providers and home care services post-cardiac arrest.
Both frailty measures were associated with post-cardiac arrest survival for both out-of-hospital and in-hospital cardiac arrest, suggesting that evaluating frailty for advance care planning is informative and can support patient-specific shared decision-making about end-of-life care directives. We found that frailty is associated with declines in functional independence and cognitive performance, two outcomes reported to be important by patients, geriatricians, and resuscitation scientists.
However, the association was only found when using the frailty index, suggesting more detailed assessments or assessment items to the frailty index (e.g., symptomology, nutrition, mood, etc.) are more sensitive to prognosticating health declines post-cardiac arrest (supplemental file). The frailty index was not associated with these outcomes in the out-of-hospital cohort, possibly due to a lack of statistical power.
Strengths and limitations
Our study is novel in providing a population-based evaluation of frailty on a broad range of post-cardiac arrest outcomes among patients receiving home care services.
For in-hospital arrests, we could not provide data on all arrests, as data were only available for in-hospital arrests that occurred within a 72-hour window. However, this allowed us to assess a cohort less likely to be influenced by health declines associated with the length of hospital stays in older adults.
We also lacked data on those not transported to the hospital from the community (i.e., termination of resuscitation, dead upon arrival, etc.). We were unable to delineate cardiac arrests that occurred in the ED from pre-hospital arrests, resulting in contamination of results for OHCA estimates. However, this likely represents less than 15% of our OHCA population, considering roughly 1-in-10 cardiac arrests that occur on a hospital premises are in the ED.
American Heart Association National Registry of Cardiopulmonary Resuscitation. Cardiac arrest in the Emergency Department: a report from the National Registry of Cardiopulmonary Resuscitation.
Finally, we excluded patients who underwent surgery pre-arrest during their index hospital; this likely decreased the generalizability of our findings to the overall in-hospital cardiac arrest population.
Post-arrest interRAI assessments were only available for roughly half of those who survived hospital discharge. However, pre-arrest frailty and health instability measures were similar between those with and without post-cardiac arrest RAI-HC assessments. Mindful that we had a population-based sample and that the median time to death was significantly lower for those without post-cardiac arrest assessment data, it is likely that these patients died before reassessment. Baseline function and cognition are significantly associated with overall mortality among patients receiving home care,
Predicting death in home care users: derivation and validation of the Risk Evaluation for Support: Predictions for Elder-Life in the Community Tool (RESPECT).
resulting in conservative estimates regarding declines in these domains post-cardiac arrest.
Conclusion
Frailty is associated with survival and post-cardiac arrest declines in cognition and function. Frail patients were less likely to survive at 30 days and 1 year when evaluated using two valid frailty measures, regardless of arrest setting (in-hospital versus out-of-hospital). Most patients assessed post-discharge experienced a decline in functional independence or cognitive performance from pre-arrest status, and the frailty index was associated with these declines. Advance care planning efforts and prognostic models will likely benefit from considering pre-arrest frailty.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
This study was supported by ICES, which is funded by an annual grant from the Ontario Ministry of Health (MOH) and the Ministry of Long-Term Care (MLTC). This study also received funding from the Canadian Institute for Health Research. Parts of this material are based on data and/or information compiled and provided by CIHI and the Ontario MOH. The analyses, conclusions, opinions and statements expressed herein are solely those of the authors and do not reflect those of the funding or data sources; no endorsement is intended or should be inferred. Parts of this report are based on Ontario Registrar General (ORG) information on deaths, the original source of which is ServiceOntario. The views expressed therein are those of the author and do not necessarily reflect those of ORG or the Ministry of Public and Business Service Delivery.
References
Rockwood K.
Song X.
MacKnight C.
Bergman H.
Hogan D.B.
McDowell I.
et al.
A global clinical measure of fitness and frailty in elderly people.
CMAJ Can Med Assoc J J Assoc Medicale Can.2005; 173: 489-495
Validation of the interRAI Cognitive Performance Scale against independent clinical diagnosis and the Mini-Mental State Examination in older hospitalized patients.
In-hospital versus out-of-hospital cardiac arrest: Characteristics and outcomes in patients admitted to intensive care after return of spontaneous circulation.
Is frailty associated with long-term survival, neurological function and patient-reported outcomes after in-hospital cardiac arrest? - A Swedish cohort study.
American Heart Association National Registry of Cardiopulmonary Resuscitation. Cardiac arrest in the Emergency Department: a report from the National Registry of Cardiopulmonary Resuscitation.
Predicting death in home care users: derivation and validation of the Risk Evaluation for Support: Predictions for Elder-Life in the Community Tool (RESPECT).
00079-5&id=gr1.jpg){kind=link}
00079-5&id=gr2.jpg){kind=link}