Journal of the Indian Academy of Geriatrics

: 2023  |  Volume : 19  |  Issue : 1  |  Page : 3--8

Frailty as a predictor of outcome in heart failure in the elderly: An observational study at a tertiary care center

Pratap Kumar1, Minakshi Dhar1, Barun Kumar2, Vikram Jain1,  
1 Department of General Medicine, AIIMS, Rishikesh, Uttarakhand, India
2 Department of Cardiology, AIIMS, Rishikesh, Uttarakhand, India

Correspondence Address:
Minakshi Dhar
Department of General Medicine, AIIMS, Rishikesh, Uttarakhand


Introduction: Heart failure is the leading contributor to global morbidity and mortality. Frailty is an emerging prognostic factor in heart failure. There is little data on the prognostic role of frailty in patients admitted for acute heart failure as most studies have been done on stable heart failure patients. Methodology: The study included elderly (age ≥60 years) patients admitted with acute heart failure at a tertiary care center in India. Patients with dementia, cognitive impairment, and documented terminal illness were excluded. The sample size was 85 patients. Frailty assessment was done using short performance physical battery (SPPB) and Fried phenotype scales and follow-up data was collected at 3 months postdischarge telephonically. The primary objective of the study was to determine the proportion of frailty in elderly in-hospital heart failure patients. The secondary objectives were to see the agreement between the frailty assessment tools used (SPPB and Fried phenotype). Results: Eighty seven patients were included in the study. The majority were male (n = 45) and had heart failure with reduced ejection fraction (n = 56). Coronary artery disease (CAD) (n = 60) was the most common cause of heart failure. Eighty-two patients had at least one comorbidity. The proportion of frailty as per the SPPB was 43.67%, and as per the Fried phenotype was 68.9%. A total of 4 deaths and 15 re-admissions occurred during the follow-up period of 3 months. The majority belonged to the frail category as per both the frailty scales (P < 0.001 for SPPB, P = 0.087 for Fried phenotype). Fleiss's kappa coefficient for agreement between the scales was 0.373 (SE = 0.106, P < 0.001), which signifies that there was a fair agreement between the two scales. The Spearman Rank correlation coefficient was −0.691 (P < 0.01) between the two scales. Hence, the SPPB score inversely correlated with the Fried phenotype. Conclusion: Frailty is largely prevalent in elderly heart failure patients. It can be used to predict poor outcomes in these patients. Clinicians should identify these high-risk patients at the time of discharge from their facility and consider interventions (tailored rehabilitation programs) to minimize the adverse outcomes.

How to cite this article:
Kumar P, Dhar M, Kumar B, Jain V. Frailty as a predictor of outcome in heart failure in the elderly: An observational study at a tertiary care center.J Indian Acad Geriatr 2023;19:3-8

How to cite this URL:
Kumar P, Dhar M, Kumar B, Jain V. Frailty as a predictor of outcome in heart failure in the elderly: An observational study at a tertiary care center. J Indian Acad Geriatr [serial online] 2023 [cited 2023 Mar 22 ];19:3-8
Available from:

Full Text


Cardiovascular diseases (CVDs) are the leading contributor to global morbidity and mortality. CVDs lead to 17.9 million deaths globally in 2019, which accounts for 32% of all fatalities as per the WHO.[1] A major share among these is contributed by heart failure; which is the final endpoint of various cardiac disorders - Rheumatic heart disease (RHD), coronary artery disease (CAD), cardiomyopathy, etc., The European Society of Cardiology (ESC) and the American College of Cardiology/American heart Association (ACC/AHA) defines heart failure as “a clinical syndrome with current or prior symptoms or signs caused by a structural and/or functional cardiac cause and corroborated by at least one of the following - elevated natriuretic peptide levels, objective evidence of cardiogenic pulmonary, or systemic congestion.”[2] The global prevalence of heart failure was estimated to be 64.3 million in 2017.[3] In India, the prevalence of heart failure according to studies was estimated to be 1.3%–6.7%.[4],[5],[6],[7] Heart failure causes functional impairment in the patient leading to poor performance status. It also leads to a financial burden on the patient.[8] This burden is quite significant in a low-income country like India.

Heart failure patients have a variety of well-known prognostic indicators. Recently, interest has grown in frailty as an important prognostic factor in heart failure. Frailty is defined as “a biological syndrome of decreased reserve and resistance to stressors, resulting from cumulative declines across multiple physiological systems, and causing vulnerability to adverse outcomes.”[9] It is associated with aging and prolonged illness. Numerous experts have lately investigated the connection between frailty and heart failure.[10],[11],[12],[13],[14] Findings from these studies suggest a significant association between frailty and adverse outcomes in patients with heart failure. However, there are very few studies on this topic from India and most of the studies are based in western countries. Furthermore, most studies have included stable heart failure patients only, and there are only a few studies that have included in-hospital patients of acute heart failure. We, therefore, wanted to conduct a study among hospitalized patients with heart failure to find the prevalence of frailty in them and also explore the prognostic role of frailty in this subset of patients.


This observational cohort study was conducted at a tertiary care center in Northern India over 15 months from July 2021 to November 2022 after approval from the Institutional Ethics Committee. All subsequent patients with age 60 years or more, admitted in the institute for acute heart failure, as per the ESC definition were included in the study. Patients with dementia, cognitive impairment, and documented terminal illness (defined as a life expectancy of 6 months or less as per the Hospice Diagnostic Guidelines) were excluded from the study. Taking the prevalence of frailty in patients with heart failure as 44.5% from a previous study[15] and considering the average number of patients admitted to the ward over the last 15 months, the sample size was calculated as 85 patients. [Figure 1] shows the study flow.{Figure 1}

Determining the percentage of frailty among elderly heart failure patients who were admitted to the institution was the study's primary objective. Secondary objectives were to see the agreement between the frailty assessment tools used (Short performance physical battery [SPPB] and Fried phenotype). After obtaining a written informed consent, patients were screened for eligibility as per the inclusion and exclusion criteria. Participant details were collected on a predesigned format which included demographic details, comorbidities, substance use, anthropometry, and in-hospital investigations. Frailty assessment was carried out at discharge of the patients using two frailty assessment tools - SPPB[16] and Fried phenotype.[9]

Fried frailty phenotype scale was developed by Fried et al.[9] in the year 2001. It uses parameters such as unintentional weight loss, self-reported exhaustion, slowness, weakness, and low physical activity to assess frailty. It has a sensitivity of 93% and specificity of 76%. Its positive predictive value is 73% and negative predictive value is 94%. For each of the criteria, the participant is classified as frail or not frail, using the following cutoffs:

Weight loss: more than 4.5 kg lost unintentionally in the last yearExhaustion: participants stating that they felt that everything they did was an effort or that they could not get going a moderate amount of the time or most of the timePhysical activity (Physical Activity Scale for Elderly): cutoff <64 for men, <52 for womenWalk time (15-feet walk): ≥7 s (men height ≤173 cm, women height ≤159 cm) or ≥6 s (men height >173 cm, women height >159 cm)Grip strength (Jamar Dynamometer, Layfayette Instruments, USA) (average of three trials): ≤29–32 kg for men (stratified by body mass index [BMI] classifications) and ≤17–21 kg for women (stratified by BMI classifications).

Participants were instructed to use an assistive ambulatory aid for the walk test if an aid was used in their normal routine. Participants who scored below the cutoffs for three or more criteria were labeled as frail, below the cutoffs for one or two criteria as prefrail, and score below the cutoffs for any criteria as nonfrail.

SPPB: The SPPB consists of three assessments:

Repeated chair standsBalance tests (side-by-side, semi-tandem, and tandem balance tests)An eight-foot walk test.

An SPPB score of ≤ 9 has the most desirable sensitivity (92%), specificity (80%), and greatest area under the curve (AUC = 0.81) for identifying frail adults.

Similar to Fried's phenotype method, the participant's scores on each component of the battery are compared to normative data and a score between zero and four is determined for each component. If participants are unable to complete a component of the test, a score of zero is given for that component. A final summary performance score out of 12 is calculated. In order to classify participants as frail, prefrail, and nonfrail, the following cutoffs were used: SPPB 0–6 (frail), SPPB 7–9 (pre-frail), and SPPB 10–12 (non-frail).

This test was done using the SPPB test mobile application. These patients were then followed up telephonically at 3 months postdischarge to enquire about their wellbeing: data on re-admission or death in this period were sought.

The data obtained were entered in MS Excel 2016. Analysis was done using MS Excel 2016 and SPSS version 23. Only the data from the initial admission were analyzed if patients were hospitalized more than once throughout the research period.

Demographic variables (age, BMI, etc.) were represented as median (Q1-Q3) and frailty scores as mean ± standard deviation. Categorical variables were represented as numbers and percentages. The proportion of frailty in elderly heart failure patients was calculated. The Chi-square test was used to test the significant difference of proportions of frailty between two groups and the proportions of adverse outcomes in the three frailty categories as per both scales. The agreement between the two frailty assessment tools was assessed by Fleiss's kappa coefficient. The correlation between both scales was checked by the Spearman rank correlation coefficient. Taking the confidence level as 95%, we defined statistical significance at P < 0.05.


The mean age of the participants was 67.8 ± 7.1 years with almost equal representation from males and females. 64.3% patients were suffering from heart failure with reduced ejection fraction, and coronary artery disease being the most common etiology, present in 68.9% patients (n = 60). 94.3% patients (n = 82) had at least one comorbidity in addition to heart failure, with hypertension and diabetes mellitus being more prevalent among them. Substance abuse was found in more than half of the population and smoking being more prevalent. [Table 1] shows the demographic details of the patient population.{Table 1}

The mean score of the participants as per the SPPB was 7.18 ± 1.81. The proportion of frailty as per the SPPB was 43.67% (frail - 38, prefrail - 41, and nonfrail - 8), as shown in [Figure 2] Similarly, the mean score of the participants as per the Fried phenotype was 3.08 ± 1.05. The proportion of frailty as per the Fried phenotype was 68.9% (frail - 60, prefrail - 26, and nonfrail - 1), as shown in [Figure 3].{Figure 2}{Figure 3}

[Table 2] and [Table 3] shows followup data at 3 months postdischarge revealed that a total of 4 deaths and 15 re-admissions occurred in the study participants. The majority of these patients belonged to the frail category at discharge. The difference in re-admission in the three frailty categories as per the SPPB showed statistical significance (P < 0.001). A similar result was seen in groups classified as per the Fried phenotype, but the difference failed to achieve statistical significance (P = 0.087).{Table 2}{Table 3}

The proportion of frailty as per the SPPB scale was 43.67%, while as per the Fried phenotype was 68.9%. Both scales classified the individuals in the same group for 58 patients. Fleiss's kappa coefficient was 0.373 (standard error [SE] = 0.106, P < 0.001), which signifies that there was a fair agreement between the two scales. The Spearman rank correlation coefficient between the two scales was −0.691 (P < 0.01). Hence, the SPPB score is inversely correlated with the Fried phenotype as shown in [Figure 4].{Figure 4}


This study estimated the proportion of frailty in elderly heart failure patients to be 43.67% and 68.9% as per the SPPB and Fried Phenotype Scale, respectively. Furthermore, there were more readmissions and deaths in the frail group compared to the other two groups. This shows the prognostic importance of frailty in elderly heart failure patients. Further, the study also addressed the agreement between two different scales for frailty assessment - Fried phenotype and SPPB scale and found that there was a fair agreement between the two scales (κ = 0.373, SE = 0.106, P < 0.001).

According to Denfeld et al.'s evaluation of 26 studies encompassing 6896 heart failure patients, the prevalence of frailty in this condition is 44.5% (Confidence interval = 36.2%–52.8%).[15] In addition, it was discovered that the prevalence was greater in studies utilizing multidimensional frailty measures and low in studies using physical frailty measures.[15] In our study also, we estimated a higher proportion of frailty by the Fried phenotype which is a multidimensional frailty measure compared to the SPPB which assesses the physical reserve of the patient.

In a study by Aung et al. which included 3881 ASIAN-HF registry participants, the prevalence of frailty was estimated to be 69%. Furthermore, frailty was associated with adverse outcomes during the 1-year follow-up period.[13] The findings of our study further support the notion that there is a link between frailty and increased hospitalization and mortality in heart failure patients. This knowledge can help clinicians in predicting the patients who are at increased risk of adverse outcomes and hence planning a more intensive monitoring and therapy plan for these patients. The risk to these patients can further be reduced by enrolling them in exercise programs to improve physical function and frailty. According to Peng et al.'s review and meta-analysis of 12 research, group-based exercise programs are beneficial for enhancing physical function, frailty, and health status.[17] A transitional, customized, progressive rehabilitation strategy with four physical-function domains was tested in a multicentric randomized control trial by Kitzman et al. (strength, balance, mobility, and endurance). They discovered that early, transitional, personalized, progressive rehabilitation interventions with various physical-function domains improved physical function more in older heart failure patients than standard therapy.[18]

In a research by Nowak et al., patients 65 years and above who had been hospitalized for acute coronary syndromes were allowed to compare the predictive value of four different frailty assessment instruments. The authors found that frailty assessed by these tools was associated with all-cause mortality and re-hospitalization in the patients. The hazard ratio of the fried phenotype was the highest. Furthermore, there was significant agreement between the scales (Fried phenotype, FRAIL scale, EFS, CFS).[19]

A study by Danilovich et al. studied the feasibility of carrying out a frailty assessment and agreement between SPPB and SHARE-FI. In this study, the agreement was assessed by Cohen's kappa coefficient which was κ = 0.264 (P < 0.001). Hence, the scales' agreement was fair but statistically significant.[20] In our study also, there was a fair agreement between the SPPB and Fried phenotype. However, the agreement was a weak one. The reason for this may be the basic difference between the components in each scale. The SPPB scale is an objective method and fully based on the physical testing of the patients-it consists of a balance test, gait speed, and chair stand test. On the other hand, Fried phenotype has 5 components. These include weight loss, exhaustion, weakness, slowness, and low physical activity levels; out of which 2 components are subjective (exhaustion and low physical activity levels). Patients who are admitted to the hospital often give a positive response to these components. Hence, only 1 patient in the study had a Fried phenotype score of 0 and was classified as non-frail. Rest all the patients were classified as pre-frail and frail.

There were certain limitations of our study. It was conducted at a single-center with a small sample size, and hence, generalization of the result is not possible. As the study was a part of thesis work, the follow-up period was only 3 months, which is a small period. The study was conducted in hospitalized patients who are ill and have poor physical reserve and hence are expected to have a high proportion of frailty. Hence, the results cannot be applied to the community setting.

As there has been very little research on frailty in heart failure in India, this study would serve as a reference for the estimation of frailty in heart failure patients of northern India. The study is unique to have covered various aspects of frailty altogether-proportion in hospitalized patients, its prognostic significance, and agreement between the frailty assessment tools.


Frailty is quite prevalent in hospitalized patients with heart failure. The proportion is high with multidimensional frailty assessment tools like Fried phenotype compared to SPPB which measures the physical reserve of the patient. Adverse outcomes are quite prevalent in frail individuals. Clinicians should identify these high-risk patients at the time of discharge from their facility and consider interventions (tailored rehabilitation programs) to minimize the adverse outcomes.


The authors express their heartfelt thankfulness to the residents, staff, and faculty of the Department of General Medicine of the Institute for their support in the study concerning the data collection and guidance in the study.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1World Health Organization. Cardiovascular diseases (CVDs); 11 June, 2021. Available from: [Last accessed on 2022 Nov 24].
2Bozkurt B, Coats AJ, Tsutsui H, Abdelhamid CM, Adamopoulos S, Albert N, et al. Universal definition and classification of heart failure: A report of the heart failure society of America, heart failure association of the European society of cardiology, Japanese heart failure society and writing committee of the universal definition of heart failure: Endorsed by the Canadian heart failure society, heart failure association of India, Cardiac society of Australia and New Zealand, and Chinese heart failure association. Eur J Heart Fail 2021;23:352-80.
3GBD 2017 Disease and Injury Incidence and Prevalence Collaborators. Global, regional, and national incidence, prevalence, and years lived with disability for 354 diseases and injuries for 195 countries and territories, 1990-2017: A systematic analysis for the Global Burden of Disease Study 2017. Lancet 2018;392:1789-858.
4Ponikowski P, Anker SD, AlHabib KF, Cowie MR, Force TL, Hu S, et al. Heart failure: Preventing disease and death worldwide. ESC Heart Fail 2014;1:4-25.
5Guo Y, Lip GY, Banerjee A. Heart failure in East Asia. Curr Cardiol Rev 2013;9:112-22.
6Martinez-Amezcua P, Haque W, Khera R, Kanaya AM, Sattar N, Lam CS, et al. The Upcoming epidemic of heart failure in South Asia. Circ Heart Fail 2020;13:e007218.
7Savarese G, Becher PM, Lund LH, Seferovic P, Rosano GM, Coats AJ. Global burden of heart failure: A comprehensive and updated review of epidemiology. Cardiovasc Res 2023;118:3272-87.
8Clark KA, Defilippis EM, Morris AA. Breaking the bank: The financial burden of living with heart failure. J Card Fail 2022;28:1434-6.
9Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: Evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001;56:M146-56.
10Pandey A, Kitzman D, Reeves G. Frailty is intertwined with heart failure: Mechanisms, prevalence, prognosis, assessment, and management. JACC Heart Fail 2019;7:1001-11.
11Keeney T, Jette DU, Cabral H, Jette AM. Frailty and function in heart failure: Predictors of 30-day hospital readmission? J Geriatr Phys Ther 2021;44:101-7.
12Sato R, Vatic M, da Fonseca GW, von Haehling S. Sarcopenia and frailty in heart failure: Is there a biomarker signature? Curr Heart Fail Rep 2022;19:400-11.
13Matsue Y, Kamiya K, Saito H, Saito K, Ogasahara Y, Maekawa E, et al. Prevalence and prognostic impact of the coexistence of multiple frailty domains in elderly patients with heart failure: The FRAGILE-HF cohort study. Eur J Heart Fail 2020;22:2112-9.
14Aung T, Qin Y, Tay WT, Binte Salahudin Bamadhaj NS, Chandramouli C, Ouwerkerk W, et al. Prevalence and prognostic significance of frailty in Asian patients with heart failure: Insights from ASIAN-HF. JACC Asia 2021;1:303-13.
15Denfeld QE, Winters-Stone K, Mudd JO, Gelow JM, Kurdi S, Lee CS. The prevalence of frailty in heart failure: A systematic review and meta-analysis. Int J Cardiol 2017;236:283-9.
16Ronai P, Gallo PM. The short physical performance battery (ASSESSMENT). ACSMs Health Fit J 2019;23:52-6.
17Peng Y, Yi J, Zhang Y, Sha L, Jin S, Liu Y. The effectiveness of a group-based Otago exercise program on physical function, frailty and health status in older nursing home residents: A systematic review and meta-analysis. Geriatr Nurs 2023;49:30-43.
18Kitzman DW, Whellan DJ, Duncan P, Pastva AM, Mentz RJ, Reeves GR, et al. Physical rehabilitation for older patients hospitalized for heart failure. N Engl J Med 2021;385:203-16.
19Nowak W, Kowalik I, Kuzin M, Krauze A, Mierzyńska A, Sadowy E, et al. Comparison of the prognostic value of frailty assessment tools in patients aged ≥65 years hospitalized in a cardiac care unit with acute coronary syndrome. J Geriatr Cardiol 2022;19:343-53.
20Danilovich MK, Diaz L, Johnson C, Holt E, Ciolino JD. Evaluating frailty in medicaid home and community-based services clients: A feasibility and comparison study between the SHARE-FI and SPPB. Pilot Feasibility Stud 2019;5:48.