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
School of Medicine, Keele University, Newcastle, United KingdomKeele Cardiovascular Research Group, Centre for Prognosis Research, Keele University, Newcastle, United Kingdom
Department of Cardiology, University Hospital of Split, Split, CroatiaKeele Cardiovascular Research Group, Centre for Prognosis Research, Keele University, Newcastle, United Kingdom
Data are limited about the contemporary association between frailty and the causes and outcomes of patients admitted with cardiovascular diseases (CVD). Using the US National Inpatient Sample, CVD admissions of interest (acute myocardial infarction, ischemic stroke, atrial fibrillation (AF), heart failure, pulmonary embolism, cardiac arrest, and hemorrhagic stroke) were stratified by Hospital Frailty Risk Score (HFRS). Logistic regression was used to determine adjusted odds ratios (aORs) and 95% confidence intervals (CIs) of in-hospital mortality among different groups with frailty. The study included 9,317,398 hospitalizations. Of these, 5,573,033 (59.8%) had a low HFRS (<5); 3,422,700 (36.7%) had an intermediate HFRS (5 to 15); and 321,665 (3.5%) had a high HFRS (>15). Ischemic stroke was the most common admission for the groups with high risk (75.4%), whereas acute myocardial infarction was the most common admission for the group with low risk (36.9%). Compared with the group with low risk, patients with high risk had increased mortality across the most CVD admissions, except in patients admitted for cardiac arrest and hemorrhagic stroke (p <0.001). The strongest association with all-cause mortality was shown among patients with high risk admitted for AF (aOR 6.75, 95% CI 6.51 to 7.00, and aOR 17.69, 95% CI 16.08 to 19.45) compared with their counterparts with low risk. In conclusion, patients with CVD admissions have varying frailty risk according to cardiovascular cause of admission, with ischemic stroke being the most common among groups with frailty and high risk. Increased frailty is associated with all-cause mortality in patients with most CVD admissions, except for cardiac arrest and hemorrhagic stroke, with the strongest association seen in patients admitted with AF.
Frailty is defined as an impairment of multiple systems resulting in an increased vulnerability to stress, leading to an increased risk of adverse outcomes such as hospitalizations and mortality, and is strongly associated with age.
Similarly, the numbers of patients living with cardiovascular (CV) disease (CVD) are increasing, particularly given an improved survivorship in patients with acute or chronic CVD.
A recent meta-analysis including 31,343 patients with CVD reported that the prevalence of frailty was 17.9% and was associated with an increased risk of heart failure (HF).
Risk of cardiovascular disease morbidity and mortality in frail and pre-frail older adults: results from a meta-analysis and exploratory meta-regression analysis.
Previous studies have attempted to understand the underlying mechanisms linking older age and adverse CVD outcomes, with common mechanisms implicated being inflammation, concomitant risk factors, and co-morbidity burden.
However, there are few data investigating whether CVD admissions vary by frailty status and whether frailty is associated with in-hospital outcomes in patients admitted with acute CVD conditions. Knowledge of the specific causes of CVD admissions and their outcomes in relation to frailty status is fundamental in planning healthcare services around the growing needs of the population living with frailty. Therefore, the aim of this study was to describe the prevalence, clinical characteristics, and in-hospital mortality of patients with the CVD admissions of interest based on their frailty status, as measured by the Hospital Frailty Risk Score (HFRS).
Methods
The National Inpatient Sample (NIS) is the largest available database of US hospitalizations developed for the Healthcare Cost and Utilization Project sponsored by the Agency for Healthcare Research and Quality.
Agency for Healthcare Research and Quality. Healthcare Cost and Utilization Project (HCUP). Available at: https://www.ahrq.gov/data/hcup/index.html. Accessed on 01/09/2023.
The NIS contains anonymized data on diagnoses and procedures from >7 million hospitalizations annually, representing a 20% stratified sample of all discharges from US community hospitals, excluding rehabilitation and long-term acute care hospitals.
Agency for Healthcare Research and Quality. Healthcare Cost and Utilization Project (HCUP). Available at: https://www.ahrq.gov/data/hcup/index.html. Accessed on 01/09/2023.
Development and validation of a Hospital Frailty Risk Score focusing on older people in acute care settings using electronic hospital records: an observational study.
to establish whether older patients at risk of adverse outcomes could be identified using routinely collected healthcare data. Briefly, a cohort of older patients (aged >74 years) hospitalized with diagnoses associated with frailty was identified.
Development and validation of a Hospital Frailty Risk Score focusing on older people in acute care settings using electronic hospital records: an observational study.
The HFRS was then created by grouping the patients identified according to their International Classification of Diseases (ICD) Tenth Revision codes into 3 groups: low risk (HFRS <5), intermediate risk (HFRS 5 to 15), and high risk (HFRS >15).
Development and validation of a Hospital Frailty Risk Score focusing on older people in acute care settings using electronic hospital records: an observational study.
Development and validation of a Hospital Frailty Risk Score focusing on older people in acute care settings using electronic hospital records: an observational study.
Each component of the HFRS and the associated weighting is outlined in Supplementary Appendix 1.
Using ICD Tenth Revision codes (Supplementary Table 1), all adult discharge records with a principal diagnosis of an acute CVD admission between October 2015 and December 2019 were identified. This sample was further filtered by focusing on the 7 CVD admissions of interest: acute myocardial infarction (AMI), atrial fibrillation/flutter (AF), ischemic stroke, HF, pulmonary embolism (PE), cardiac arrest, and hemorrhagic stroke. The sample was further stratified according to their frailty status measured by the HFRS into 3 groups: low risk (HFRS <5), intermediate risk (HFRS 5 to 15), and high risk (HFRS >15), as defined by Gilbert et al.
Development and validation of a Hospital Frailty Risk Score focusing on older people in acute care settings using electronic hospital records: an observational study.
Cases were excluded owing to missing data for the following variables: age, gender, elective admission, in-hospital mortality, primary expected payer, total charges, and length of stay. These cases accounted for no more than 1.0% of the original dataset. Cases not pertaining to 1 of the 7 diagnoses of interest were also excluded (Supplementary Figure 1). This observational study was appraised according to the Strengthening The Reporting of OBservational Studies in Epidemiology recommendations (Supplementary Appendix 2).
Continuous variables such as age, length of stay, and total charges were summarized using median and interquartile range. Categorical variables were compared using the chi-square test and summarized as percentages (%). Multivariable logistic regression was performed to determine the adjusted odds ratio (aOR) for all-cause mortality. Regression was adjusted for the following variables: age, gender, race, weekend admission, primary expected payer, median household income, bed size of hospital, region of hospital, location/teaching status of hospital, smoking status, previous myocardial infarction, previous percutaneous coronary intervention, previous coronary artery bypass graft, dyslipidemia, and Elixhauser co-morbidities (anemias, coagulopathy, diabetes mellitus, liver disease, metastatic cancer, peripheral vascular disorders, and chronic renal failure). Results were presented as aORs with 95% confidence intervals (CIs). Results were determined significant at the level of p <0.05. All statistical analyses were weighted and performed using SPSS version 27 (IBM Corp, Armonk, New York).
A total of 9,317,398 discharges had 1 of the 7 CVD diagnoses of interest (AMI, ischemic stroke, AF, HF, PE, and hemorrhagic stroke) (Supplementary Figure 1). Overall, 5,573,033 discharges (59.8%) had an HFRS of <5; 3,422,700 (36.7%) had an HFRS of 5 to 15; and 321,665 (3.5%) had an HFRS of >15 (Table 1).
Table 1Patient characteristics for all cardiovascular admissions according to HFRS
Patients with an HFRS >15 were more likely to be older (median age 75 vs 73 years for HFRS 5-to- 15 group and 68 years for HFRS <5 group) and female (54.1% vs 48.7% for HFRS 5-to-15 group and 43.0% for HFRS <5 group) and to have a higher prevalence of hypertension, coagulopathy, and thrombocytopaenia, in addition to a lower prevalence of previous AMI, previous percutaneous coronary intervention, previous coronary artery bypass graft, HF, and diabetes, than were patients with an HFRS <5 and HFRS 5 to 15 (p <0.001 for all) (Table 1).
The most common cause of admission was AMI (28.7%), followed by ischemic stroke (23.8%), AF (21.0%), HF (16.2%), hemorrhagic stroke (5.9%), PE (4.0%), and cardiac arrest (0.4%). The cohort admitted with ischemic stroke had the highest proportion of patients with an HFRS >15 (10.8%), followed by hemorrhagic stroke and cardiac arrest (9.1% and 2.5%). Similarly, cohorts admitted with cardiac arrest had the highest proportion of patients with an HFRS 5 to 15 (70.4%), followed by ischemic stroke and hemorrhagic stroke (66.9% and 54.9%, respectively). The cohort admitted with AF had the highest proportion of patients with an HFRS <5, followed by AMI and PE (80.3% vs 76.7% and 74.1%) (Figure 1, Supplementary Table 2-8 ).
Figure 1Distribution of each HFRS category within each of the selected cardiovascular admission causes.
The most common cause of CVD admission in the HFRS <5 cohort was AMI (36.9%), followed by AF (28.2%) and HF (17.4%). Ischemic stroke was the most common CVD admission for the HFRS 5-to-15 group (43.3%), followed by AMI (17.8%) and HF (15.6%). Similarly, ischemic stroke was the most common CVD admission for the HFRS of >15 groups (75.4%), followed by hemorrhagic stroke (15.5%) and AMI (4.3%) (Figure 2).
Figure 2Distribution of selected cardiovascular admission causes within each HFRS category.
Patients with HFRS >15 had higher unadjusted rates of all-cause mortality than did their counterparts with lower frailty (10.3% vs 7.6 for HFRS 5-to-15 group and 2.2% for HFRS <5 group, p <0.001). Increased unadjusted rates of all-cause mortality for patients with high-risk frailty were also observed in patients admitted with AMI, ischemic stroke, AF, HF, and PE but not for patients admitted with cardiac arrest or hemorrhagic stroke (all p <0.001) (Supplementary Figure 2, Table 2).
Table 2Prevalence of the cardiovascular admission diagnoses and associated in-hospital deaths based on HFRS
On adjustment for baseline covariates, increasing frailty risk was associated with increased odds of all-cause mortality. Patients with an HFRS of 5 to 15 or HFRS >15 admitted for AF had the highest odds of all-cause mortality (aOR 17.69, 95% CI 16.08 to 19.45 for HFRS >15 group, aOR 6.75, 95% CI 6.51 to 7.00 for HFRS 5-to-15 group). Increased odds of mortality were observed with worsening frailty status across a broad range of different CV causes for admission. The HFRS >15 group was associated with in an increased odds of mortality for patients admitted with AMI, ischemic stroke, HF, and PE admission diagnoses (p <0.001). Interestingly, a decreased odds of mortality in the HFRS >15 group was observed in patients admitted for cardiac arrest and hemorrhagic stroke only (aOR 0.46, 95% CI 0.39 to 0.55 for patients with cardiac arrest with an HFRS >15, aOR 0.86, 95% CI 0.83 to 0.88 for patients with hemorrhagic stroke with an HFRS >15) (Figure 3, Table 3).
Figure 3Adjusted mortality rates for different frailty risk category and selected cardiovascular admission causes. Reference group is low HFRS score <5 for each CVD admission diagnosis.
To the best of our knowledge, this is the first study to examine the prevalence, clinical characteristics, and in-hospital mortality of patients admitted with a broad range of acute CV presentations on a nationwide scale based on their frailty status. We report several important findings. Firstly, we report the most common CVD admissions across frailty categories, with AMI being the most common in patients with a low frailty score, and ischemic stroke being the most common in patients with an intermediate and high frailty score. Secondly, we report important frailty-based differences in baseline characteristics across patients with different CVD admission diagnoses. Finally, patients with an intermediate and high frailty score had increased all-cause mortality compared with their counterparts with lower risk across most CVD admission diagnoses, except in cardiac arrest and hemorrhagic stroke categories.
The association between frailty and CVD has been widely explored in the literature.
Risk of cardiovascular disease morbidity and mortality in frail and pre-frail older adults: results from a meta-analysis and exploratory meta-regression analysis.
Fried LP, Cardiovascular Health Study. Frailty and activation of the inflammation and coagulation systems with and without clinical comorbidities: results from the cardiovascular Health Study.
Villa Pini Stroke Data Bank Investigators. Inflammation, hemostatic markers, and antithrombotic agents in relation to long-term risk of new cardiovascular events in first-ever ischemic stroke patients.
Which biomarkers are predictive specifically for cardiovascular or for non-cardiovascular mortality in men? Evidence from the Caerphilly prospective study (CaPS).
The pathways in conjunction with low physical activity and poor nutrition could lead to decreased physiological reserves and increased susceptibility to stress, leading to frailty.
In addition, these factors can lead to a prothrombotic state and increased levels of inflammation, leading to increased risk of CV events and adverse outcomes.
Fried LP, Cardiovascular Health Study. Frailty and activation of the inflammation and coagulation systems with and without clinical comorbidities: results from the cardiovascular Health Study.
Villa Pini Stroke Data Bank Investigators. Inflammation, hemostatic markers, and antithrombotic agents in relation to long-term risk of new cardiovascular events in first-ever ischemic stroke patients.
Which biomarkers are predictive specifically for cardiovascular or for non-cardiovascular mortality in men? Evidence from the Caerphilly prospective study (CaPS).
The HFRS was nationally validated using a cohort of >1 million patients in the United Kingdom, of whom 37.6% had an intermediate risk of frailty and 20% had a high risk of frailty.
Development and validation of a Hospital Frailty Risk Score focusing on older people in acute care settings using electronic hospital records: an observational study.
In our cohort of 9 million US patients hospitalized with CVD, there was a lower proportion of patients at intermediate or high risk of frailty. The prevalence of frailty among overall patients with CVD was previously estimated between 15% and 19%,
Risk of cardiovascular disease morbidity and mortality in frail and pre-frail older adults: results from a meta-analysis and exploratory meta-regression analysis.
Variable prevalence of frailty in the literature could be partially explained by differences in the cohorts studied but also and importantly by differences in definitions used and what is considered to represent frailty, with studies using definitions derived from Rockwood et al,
Risk of cardiovascular disease morbidity and mortality in frail and pre-frail older adults: results from a meta-analysis and exploratory meta-regression analysis.
Development and validation of a Hospital Frailty Risk Score focusing on older people in acute care settings using electronic hospital records: an observational study.
There is a challenge to defining frailty because there is no standardized measurement, but the HFRS represents a potentially advantageous option owing to its dependence on the widely available ICD coding system.
The HFRS follows the deficit model (combining impairments) and has been validated against the Rockwood and Fried scores; however, it can be quite challenging for clinicians to calculate because of a lack of automated computation.
The Hospital Frailty Risk Score and its association with in-hospital mortality, cost, length of stay and discharge location in patients with heart failure short running title: frailty and outcomes in heart failure.
Similarly to other studies, we observed that patients with increasing frailty are likely to be older and female, and to have longer hospital stays and total costs.
Risk of cardiovascular disease morbidity and mortality in frail and pre-frail older adults: results from a meta-analysis and exploratory meta-regression analysis.
Development and validation of a Hospital Frailty Risk Score focusing on older people in acute care settings using electronic hospital records: an observational study.
The Hospital Frailty Risk Score and its association with in-hospital mortality, cost, length of stay and discharge location in patients with heart failure short running title: frailty and outcomes in heart failure.
This study found important variations in frailty status across different CVD admissions. Among patients with intermediate or high frailty risk, the most prevalent CVD admission was acute ischemic stroke. This may be explained by the inclusion of the sequelae of stroke in the HFRS, but other contributors, such as older age, which increases the risk of both stroke and frailty, with 70% of strokes occurring after the age of 65, are also important.
AF has been reported to be prevalent in 15% of the population with frailty, although we report an interesting pattern of a decreasing proportion of AF admission with increasing frailty.
Studies using the HFRS indicated a similar distribution of frailty status to this study among patients with AF, with most patients at low risk of frailty and only a small percentage at high risk of frailty.
This study shows admission for HF decreases with increasing frailty yet is still common, with >3 in 10 patients admitted with HF at intermediate or high risk of frailty. Our findings are supported by other studies that show intermediate or high risk of frailty is present in up to 1 in 5 patients hospitalized with HF and is associated with a longer length of stay and increased total charges.
The Hospital Frailty Risk Score and its association with in-hospital mortality, cost, length of stay and discharge location in patients with heart failure short running title: frailty and outcomes in heart failure.
This could be explained by an increased number of co-morbidities in the population with HF, with a high prevalence of dyslipidemia, anemia, and hypertension.
The Hospital Frailty Risk Score and its association with in-hospital mortality, cost, length of stay and discharge location in patients with heart failure short running title: frailty and outcomes in heart failure.
This agrees with multiple studies that showed patients with increased HFRS have a higher Charlson co-morbidity score, in line with the HFRS being based on the total co-morbidity burden of patients.
Development and validation of a Hospital Frailty Risk Score focusing on older people in acute care settings using electronic hospital records: an observational study.
The Hospital Frailty Risk Score and its association with in-hospital mortality, cost, length of stay and discharge location in patients with heart failure short running title: frailty and outcomes in heart failure.
There are no studies describing the prevalence of PE, cardiac arrest, and hemorrhagic stroke among patients with frailty who are hospitalized.
Interestingly, AF was a rare cause of admission in the high-risk group but was associated with the worst prognosis in these patients. The association between AF and mortality has been reported in multiple studies because the prevalence of AF increases with age and co-morbidity burden, and increases the risk of stroke and its associated complications.
Studies have suggested that patients with frailty also have a larger left atrial volume, which is one of the main cardiac abnormalities linked to the development of AF and systolic dysfunction.
Furthermore, patients with incident AF are commonly on anticoagulant medications, which increases risk of bleeding and further complications such as hemorrhage.
However, studies report that patients with AF and increasing frailty are less likely to be treated with oral anticoagulants, which can lead to increased likelihood of downstream thrombotic events and poorer outcomes.
The Hospital Frailty Risk Score and its association with in-hospital mortality, cost, length of stay and discharge location in patients with heart failure short running title: frailty and outcomes in heart failure.
Our findings of cardiac arrest and hemorrhagic stroke, although different to the other CVD admissions of interest, could have several explanations. These conditions have substantial mortality per se, with little modification by HFRS. This analysis encompassed only patients who were admitted owing to cardiac arrest, leading to potential selection bias, so it is possible that many of the patients with frailty did not survive to admission, and only those with the most favorable prognosis survived to admission. It is possible that the poor outcomes in these patient groups occur independently of frailty status.
There are important clinical implications of this study. This study indicates that patients with intermediate-to-high frailty risk represent a substantial portion of the population admitted for CVD and raises the importance of frailty assessment by cardiologists. A co-existence of frailty and CVD is becoming even more important owing to an aging population with higher morbidity burden. Patients with frailty admitted with CVD have higher mortality rates and burden the healthcare system, and knowledge of the trends in CVD admission is fundamental to improve the outcomes of this clinically at-risk population. This study may support the early identification and management of CVD in patients with frailty, particular in primary care, although whether this would affect acute admissions is not known.
This study has several limitations inherent to the use of the NIS database. Firstly, coded data for the NIS could be subject to selection bias because of inaccurate coding or missing data. Secondly, detailed clinical information such as pharmacological treatment that can mediate outcomes could not be investigated because it was not available from the NIS. The impact of differential pharmacological management in the population with frailty on outcomes may be an area for further research. Thirdly, because this is an observational study, confounding bias could not be fully eliminated despite the broad scope of conditions covered by the NIS, and therefore, causality between frailty, CVD admission, and mortality cannot be proved. Finally, the NIS only captures information on in-hospital events, and therefore, a more detailed analysis of longitudinal outcomes could not be assessed.
In conclusion, the causes of CVD admission vary with frailty status, with AMI being the most common in patients with a low risk of frailty, whereas ischemic stroke is the most common in patients with intermediate or high risk of frailty. Increasing frailty in patients admitted for AMI, ischemic stroke, AF, HF, and PE is associated with an increased all-cause mortality. Future, more granular studies are necessary to guide care and improve the CVD outcomes in patients with frailty in an ever-aging population.
Disclosures
The authors have no conflicts of interest to declare.
Risk of cardiovascular disease morbidity and mortality in frail and pre-frail older adults: results from a meta-analysis and exploratory meta-regression analysis.
Agency for Healthcare Research and Quality. Healthcare Cost and Utilization Project (HCUP). Available at: https://www.ahrq.gov/data/hcup/index.html. Accessed on 01/09/2023.
Development and validation of a Hospital Frailty Risk Score focusing on older people in acute care settings using electronic hospital records: an observational study.
Fried LP, Cardiovascular Health Study. Frailty and activation of the inflammation and coagulation systems with and without clinical comorbidities: results from the cardiovascular Health Study.
Villa Pini Stroke Data Bank Investigators. Inflammation, hemostatic markers, and antithrombotic agents in relation to long-term risk of new cardiovascular events in first-ever ischemic stroke patients.
Which biomarkers are predictive specifically for cardiovascular or for non-cardiovascular mortality in men? Evidence from the Caerphilly prospective study (CaPS).
The Hospital Frailty Risk Score and its association with in-hospital mortality, cost, length of stay and discharge location in patients with heart failure short running title: frailty and outcomes in heart failure.
This work was supported by the Wolfson Intercalated Award (United Kingdom) (Balamrit Singh Sokhal), the Midlands Racial Equality in Medicine Elevate (Warwick, United Kingdom) research grant (Balamrit Singh Sokhal), National Institute for Health Research Applied Research Collaboration West Midlands (United Kingdom) (Professor Christian Mallen), and the National Institute for Health Research School for Primary Care Research (United Kingdom) (Professor Christian Mallen).
01320-0&id=ga1.jpg){kind=link}
01320-0&id=gr1.jpg){kind=link}
01320-0&id=gr2.jpg){kind=link}
01320-0&id=gr3.jpg){kind=link}