
==== Front
J Frailty Sarcopenia Falls
J Frailty Sarcopenia Falls
Journal of Frailty, Sarcopenia and Falls
2459-4148
HYLONOME PUBLICATIONS Greece

JFSF-9-227
10.22540/JFSF-09-227
Short Communication
Examining the Cycle of Physical Frailty in Falls Clinic Attendees Through Structural Equation Modeling
Ortuño Román Romero 12
Duggan Eoin 12
1 FRAILMatics research group, Discipline of Medical Gerontology, School of Medicine, Trinity College Dublin, Ireland
2 Falls and Syncope Unit, Mercer’s Institute for Successful Ageing, St James’s Hospital, Dublin, Ireland
Corresponding author: Román Romero Ortuño, Discipline of Medical Gerontology, Mercer’s Institute for Successful Ageing (MISA), St James’s Hospital, James’s Street, Dublin 8, Ireland E-mail: romeroor@tcd.ie ORCID: 0000-0002-3882-7447
The authors have no conflict of interest.

9 2024
01 9 2024
9 3 227231
01 7 2024
Copyright: © 2024 Hylonome Publications
2024
https://creativecommons.org/licenses/by-nc-sa/4.0/ All published work is licensed under Creative Commons Attribution NonCommercial - ShareAlike 4.0 International
Abstract

In 1998, Fried and Walston introduced the Cycle of Frailty (CF) as a foundational concept for defining the physical frailty phenotype (FP). While the FP has been extensively validated, the CF hypothesis lacks equivalent support. This study aimed to internally validate the CF using structural equation modeling (SEM) in a clinical dataset of adults aged 50 or older attending an outpatient falls clinic. Measures included: age, morbidity, nutrition, sarcopenia by bioelectrical impedance, VO2max, handgrip strength, basal metabolic rate (BMR), 5-times chair stand test (5CST), physical activity, and total energy expenditure (TEE). The SEM, incorporating CF hypothesized causal pathways, was tested using IBM® SPSS® Amos 27.0.0 (maximum likelihood method) with a sample of 102 adults (mean age 69.8 years, 58.8% women). Overall, the SEM was supported by the data (χ2 = 44.4, df = 37, p = 0.189), with significant (p < 0.05) regression weights for morbidity→sarcopenia, age→sarcopenia, sarcopenia→VO2max, sarcopenia→handgrip strength, handgrip strength→5CST, physical activity→TEE, TEE→nutrition, and BMR→TEE. However, nutrition→sarcopenia, sarcopenia→BMR, VO2max→5CST, and 5CST→physical activity were not significant. Although the SEM was limited by inclusion of surrogate CF measures (e.g., 5CST instead of gait speed, VO2max based on age-predicted maximal/resting heart rate), it provided some internal support for the CF hypothesis.

Falls clinic
Physical frailty phenotype
Sarcopenia
Structural equation modeling
Validation studies
==== Body
pmcIn 1998, Linda P. Fried and Jeremy Walston proposed the Cycle of Frailty (CF) as the conceptual foundation for defining the physical frailty phenotype (FP)[1]. While the FP has undergone extensive validation and utilisation[2,3], the CF hypothesis lacks comparable support. This study aimed to internally and cross-sectionally validate the CF using a structural equation model (SEM) in a clinical dataset of patients attending a falls clinic.

Adults aged 50 years or older attending an outpatient falls and syncope unit (FASU) in St. James’s Hospital, Dublin, Ireland, were recruited between January and November 2022. The clinical setting has been described elsewhere[4,5].

Measures collected in the clinic included: age in years; sex (male vs. female); disease burden (morbidity) as per Cumulative Illness Rating Scale-Geriatric (CIRS-G)[6] score; nutritional status as per Mini Nutritional Assessment® Short-Form (MNA®-SF)[7] ordinal classification (0: normal; 1: at risk of malnutrition; 2: malnourished); sarcopenia status (0: robust; 1: probable sarcopenia; 2: sarcopenia) determined by bioelectrical impedance using TANITA® DC-430 MAP Body Composition Analyser following European Working Group on Sarcopenia in Older People guidelines, as described elsewere4,5; VO2max calculation (ml/kg/min) as per the following formula: 15.3 x [208 - (0.7 x age)] / resting heart rate[8,9], where resting heart rate was obtained using a Finapres® Nova (Finapres Medical Systems, Amsterdam, The Netherlands) by averaging the mean heart rate from 60 to 30 seconds while lying down supine before an active stand test[5]; maximum handgrip strength (kg) defined as the highest value obtained from four consecutive measurements while seated, two from the right hand and two from the left hand, using a Jamar Hydraulic Hand Dynamometer (Performance Health, Wisconsin, USA)[5]; basal (resting) metabolic rate (BMR, kcal/day), measured by the TANITA® device[4]; time in seconds to complete a 5-times chair stand test (5CST)5; self-reported physical activity assessed by the question: “How often do you engage in activities that require a low or moderate level of energy such as gardening, cleaning the car, or doing a walk?” (1: Hardly ever or never; 2: One to three times a month; 3: Once a week; 4: More than once a week)[10]; total energy expenditure (TEE) (kcal/day), defined as BMR x activity factor, where the following activity factors were assigned to the four aforementioned physical activity categories: 1.2 (hardly ever/never); 1.375 (one to three times/month); 1.55 (once a week); 1.725 (more than once a week)[11]; self-reported food intake in the last month (0: diminution in desire for food and/or eating less than usual; 1: no change in desire for food and/or eating the same as usual; 2: increase in desire for food and/or eating more than usual)[10]; and self-reported weight loss in the past month (0: no; 1: yes).

IBM® SPSS® Statistics version 27 was used to compute descriptive statistics in the included sample of 102 adults. The mean age was 69.8 years (standard deviation [SD] 10.3; minimum 50; maximum 93), and 58.8% were women. The median CIRS-G score was 6 (interquartile range [IQR]: 5). According to the MNA®-SF, 39.2% were at risk of malnutrition and 5.9% were malnourished. As per TANITA® assessment, 30.4% had probable sarcopenia and 13.7% were sarcopenic. The mean VO2max was 35.6 ml/kg/min (SD 6.2), and the mean maximum handgrip strength was 25.8 kg (SD 9.9). Mean BMR was 1487.7 kcal/day (SD 317.4) and mean 5CST time was 16.0 seconds (SD 6.2). In terms of engagement in physical activities, 15.7% reported hardly ever or never; 1.0% one to three times a month; 5.9% once a week; and 77.5% more than once a week. In terms of self-reported food intake, 2.0% reported diminution in desire for food and/or eating less than usual; 11.8% no change in desire for food and/or eating the same as usual; 86.3% increase in desire for food and/or eating more than usual; and 15.7% self-reported weight loss.

The SEM was tested using IBM® SPSS® Amos 27.0.0 (maximum likelihood method). Overall, the SEM was supported by the data (χ2 = 44.4, df = 37, p = 0.189). In terms of model fit, the root mean square error of approximation (RMSEA) was 0.044 (95% confidence interval: <0.001 – 0.087, p = 0.546). Figure 1 displays the SEM results with standardized estimates (SE), illustrating the main CF hypothesized causal pathways[1] depicted by bold arrows. Table 1 shows the full model estimates.

Table 1 Structural equation model estimates.

Hypothesized effect	Unstandardized Estimate	Standard Error	p	
VO2max ← age	-0.161	0.061	0.008	
Sarcopenia status ← CIRS-G score	0.053	0.018	0.003	
Physical activity ← CIRS-G score	-0.055	0.029	0.061	
Sarcopenia status ← age	0.020	0.006	0.002	
BMR ← age	-9.783	3.018	0.001	
Weight loss ← food intake	-0.554	0.076	<0.001	
VO2max ← CIRS-G score	-0.283	0.167	0.090	
MNA®-SF status ← food intake	-0.772	0.118	<0.001	
5CST ← CIRS-G score	0.338	0.132	0.011	
Handgrip strength ← age	-0.206	0.089	0.021	
5CST ← Handgrip strength	-0.144	0.052	0.005	
TEE ← Physical activity	258.959	6.242	<0.001	
VO2max ← sarcopenia status	2.631	0.905	0.004	
TEE ← BMR	1.592	0.022	<0.001	
5CST ← VO2max	-0.053	0.075	0.477	
Sarcopenia status ← weight loss	0.013	0.193	0.947	
Weight loss ← MNA®-SF status	0.058	0.051	0.260	
Sarcopenia status ← MNA®-SF status	0.136	0.121	0.260	
MNA®-SF status ← TEE	<0.001	<0.001	0.006	
Handgrip strength ← sarcopenia status	-5.189	1.271	<0.001	
Physical activity ← sarcopenia status	-0.494	0.177	0.005	
VO2max ← MNA®-SF status	-1.869	0.966	0.053	
BMR ← sarcopenia status	-77.651	43.261	0.073	
5CST ← sarcopenia status	3.840	0.755	<0.001	
Physical activity ← 5CST	<0.001	0.021	0.998	
Age ←→ CIRS-G score	11.456	3.968	0.004	
CIRS-G score ←→ food intake	-0.289	0.157	0.065	
Age ←→ food intake	-0.547	0.425	0.197	
e4 ←→ e3	1429.013	280.688	<0.001	
CIRS-G: Cumulative Illness Rating Scale-Geriatric; BMR: basal (resting) metabolic rate; 5CST: 5-times chair stand test; TEE: total energy expenditure; MNA®-SF: Mini Nutritional Assessment® Short-Form; ←: regression; ←→: covariance. Main CF hypothesized effects1 are highlighted in bold; statistically significant (p < 0.05) values are highlighted in bold.

Figure 1 Structural equation model depicting the cycle of physical frailty incorporating data from 102 falls clinic attendees. The cycle of frailty main hypothesized pathways1 are depicted by bold arrows. MNA status: MNA®-SF status; CIRS-G: Cumulative Illness Rating Scale-Geriatric; Grip_Max: maximum handgrip strength; 5CST: 5-times chair stand test; Activity: physical activity; e: error term; df: degrees of freedom.

In terms of the main CF hypothesized causal pathways1, we observed significant (p < 0.05) regression weights for CIRS-G score → sarcopenia status (SE = 0.27, p = 0.003), age → sarcopenia status (SE = 0.28, p = 0.002), sarcopenia status → VO2max (SE = 0.31, p = 0.004), sarcopenia status → handgrip strength (SE = -0.38, p < 0.001), handgrip strength → 5CST (SE = -0.23, p = 0.005), physical activity → TEE (SE = 0.46, p < 0.001), BMR → TEE (SE = 0.82, p < 0.001), and TEE → MNA®-SF status (SE = -0.23, p = 0.006).

However, MNA®-SF status → sarcopenia status (SE = 0.11, p = 0.260), sarcopenia status → BMR (SE = -0.18, p = 0.073), VO2max → 5CST (SE = -0.05, p = 0.477), and 5CST → physical activity (SE < 0.01, p = 0.998) were not significant in the SEM. In this same clinical cohort, consistent with the SEM analysis presented herein, we previously demonstrated using multiple binary logistic regression that bioimpedance-measured appendicular skeletal muscle mass was not significantly associated with abnormal MNA®-SF but with frailty status measured by the frailty phenotype-based SHARE-FI tool[4]. Regarding sarcopenia status → BMR, the SEM regression approached significance (SE = -0.18, p = 0.073), suggesting potential statistical underpower in the SEM, especially considering that other larger studies previously reported a significant association between sarcopenia and BMR[12,13]. Furthermore, the lack of significance observed in the relationships VO2max → 5CST and 5CST → physical activity could be attributed to the fact that in this analysis, 5CST served as a surrogate for usual gait speed, with the latter originally included in the CF hypothesis[1]. Although it has been previously proposed that 5CST might be useful as a substitute for gait speed when diagnosing sarcopenia[14], this modification could impact on its predictive ability in the context of the FP operationalisation[15].

In terms of additional CF hypothesized effects in the SEM, significant results were as expected: older age causing a reduction in VO2max[16] (SE = -0.27, p = 0.008), BMR[17] (SE = -0.32, p = 0.001), and handgrip strength[18] (SE = -0.21, p = 0.021); reduced food intake leading to weight loss (SE = -0.63, p < 0.001) and worse MNA®-SF status (SE = -0.52, p < 0.001); higher CIRS-G score prolonging 5CST time (SE = 0.20, p = 0.011); and sarcopenia impairing 5CST time (SE = 0.45, p < 0.001) and physical activity (SE = -0.32, p = 0.005). The direct covariance between age and CIRS-G score (SE = 0.30, p = 0.004) was also as expected; and the covariance of error terms between maximum handgrip strength and RMR highlights that both these measures are influenced by anthropometric factors such as weight and/or height, age, and sex[17,18].

The proximity to statistical significance (p < 0.100) of certain anticipated effects might indicate statistical underpower within the SEM. This scenario is likely applicable to hypothesized effects, such as higher CIRS-G score leading to decreased physical activity (SE = -0.19, p = 0.061) and VO2max (SE = -0.17, p = 0.090), along with a correlation with reduced food intake (SE = -0.13, p = 0.065). Similarly, it may pertain to worse nutrition status causing a decrease in VO2max (SE = -0.18, p = 0.053). Notably, the exclusion of adults aged 50 to 64 from the sample, considering that the original FP was validated in adults aged 65 or over[3], would have further compromised the power of the SEM.

In terms of the remaining SEM effects (p ≥ 0.100), statistical underpower may have also been implicated in the non-significant inverse association between age and food intake[19] (SE = -0.13, p = 0.197). Additionally, limitations inherent in self-reporting, particularly in subjective weight loss, may have contributed to the lack of significance in hypothesized effects with sarcopenia and MNA®-SF status.

Further limitations include that our calculation of VO2max was based on age-predicted maximal heart rate, which can be less accurate in non-healthy adults[9]; furthermore, while resting heart rate serves as a valid biomarker of cardiorespiratory fitness at the population level, the relationship between resting heart rate and fitness may be weakened by factors such as physical activity and adiposity[20]. Indeed, even though sarcopenia status → VO2max was significant (SE = 0.31, p = 0.004), the direction of the association (sarcopenia increasing VO2max) was counterintuitive[21], even if VO2max in older subjects may be less influenced by muscle mass than in younger subjects[22]. Uth and colleagues observed that the ratio of maximal heart rate to resting heart rate could serve as a potential indicator for estimating VO2max in well-trained men, but applicability to other groups was less certain[8]. Interestingly, no other standardized estimates in the SEM displayed an unexpected direction. Finally, the standard activity factors derived from the Harris-Benedict Equation, which were utilized to estimate Total Energy Expenditure (TEE) in our sample, may not accurately reflect the energy requirements of this clinical sample[23].

In conclusion, this study’s primary strength lies in offering some empirical support for the CF hypothesis through SEM analysis of a clinical dataset comprising adult patients from a falls clinic, representing a novel methodological contribution to the literature. Nonetheless, limitations include the relatively small sample size, utilization of surrogate CF measures, and the lack of an independent sample for external validation of the model. Further research addressing these limitations is encouraged.

Ethics approval

The study received ethical approval from the Tallaght University Hospital (TUH)/ St. James’s Hospital (SJH) Joint Research Ethics Committee (Project ID: 0221; approval date: 4 May 2021). Approval was also granted by St. James’s Hospital Research & Innovation Office (Reference: 6567, approval date: 26 July 2021). The study adhered to the World Medical Association Declaration of Helsinki on ethical principles for medical research involving human subjects.

Consent to participate

All participants provided written informed consent for inclusion in the study.

Funding

This research was funded by Science Foundation Ireland (SFI), grant number 18/FRL/6188.

Edited by: George Lyritis
==== Refs
References

1 Fried LP Walston J Hazzard WR Blass JP Ettinger WHJ Halter JB Ouslander J Frailty and failure to thrive Principles of Geriatric Medicine and Gerontology 1998 4 ed New York McGraw-Hill 1387 402
2 Buta BJ Walston JD Godino JG Frailty assessment instruments:Systematic characterization of the uses and contexts of highly-cited instruments Ageing Res Rev 2016 26 53 61 26674984
3 Fried LP Tangen CM Walston J Frailty in older adults:evidence for a phenotype J Gerontol A Biol Sci Med Sci 2001 56 3 M146 56 11253156
4 Lionetti E Duggan E Romero-Ortuno R The SHARE Frailty Instrument for Primary Care was Associated with Sarcopenia, as Measured by Bioelectrical Impedance, in Falls Clinic Attendees J Frailty Sarcopenia Falls 2024 9 1 10 5 38444542
5 Duggan E Knight SP Romero-Ortuno R Relationship between sarcopenia and orthostatic blood pressure recovery in older falls clinic attendees Eur Geriatr Med 2023 14 3 439 46 37029293
6 Miller MD Paradis CF Houck PR Rating chronic medical illness burden in geropsychiatric practice and research:application of the Cumulative Illness Rating Scale Psychiatry Res 1992 41 3 237 48 1594710
7 Kaiser MJ Bauer JM Ramsch C Validation of the Mini Nutritional Assessment short-form (MNA-SF):a practical tool for identification of nutritional status J Nutr Health Aging 2009 13 9 782 8 19812868
8 Uth N Sorensen H Overgaard K Pedersen PK Estimation of VO2max from the ratio between HRmax and HRrest-the Heart Rate Ratio Method Eur J Appl Physiol 2004 91 1 111 5 14624296
9 Tanaka H Monahan KD Seals DR Age-predicted maximal heart rate revisited J Am Coll Cardiol 2001 37 1 153 6 11153730
10 Romero-Ortuno R Walsh CD Lawlor BA Kenny RA A frailty instrument for primary care:findings from the Survey of Health, Ageing and Retirement in Europe (SHARE) BMC Geriatr 2010 10 57 20731877
11 Calculators.org BMR Calculator Accessed 28 March 2024 Available online on: https://www.calculators.org/health/bmr.php.2024
12 Cho YJ Cho MH Han B Park M Bak S Park M The Association between the Ratio of Energy Intake to Basal Metabolic Rate and Physical Activity to Sarcopenia:Using the Korea National Health and Nutrition Examination Surveys (|y2008-2011) Korean J Fam Med 2020 41 3 167 74 32456384
13 Soysal P Ates Bulut E Yavuz I Isik AT Decreased Basal Metabolic Rate Can Be an Objective Marker for Sarcopenia and Frailty in Older Males J Am Med Dir Assoc 2019 20 1 58 63 30122323
14 Nishimura T Arima K Okabe T Usefulness of chair stand time as a surrogate of gait speed in diagnosing sarcopenia Geriatr Gerontol Int 2017 17 4 659 61 27120799
15 Theou O Cann L Blodgett J Wallace LM Brothers TD Rockwood K Modifications to the frailty phenotype criteria:Systematic review of the current literature and investigation of 262 frailty phenotypes in the Survey of Health, Ageing, and Retirement in Europe Ageing Res Rev 2015 21 78 94 25846451
16 Betik AC Hepple RT Determinants of VO2 max decline with aging:an integrated perspective Appl Physiol Nutr Metab 2008 33 1 130 40 18347663
17 Pavlidou E Papadopoulou SK Seroglou K Giaginis C Revised Harris-Benedict Equation:New Human Resting Metabolic Rate Equation Metabolites 2023 13 2 189 36837808
18 Kenny RA Coen RF Frewen J Donoghue OA Cronin H Savva GM Normative values of cognitive and physical function in older adults:findings from the Irish Longitudinal Study on Ageing J Am Geriatr Soc 2013 61 Suppl 2 S279 90 23662720
19 Drewnowski A Shultz JM Impact of aging on eating behaviors, food choices, nutrition, and health status J Nutr Health Aging 2001 5 2 75 9 11426286
20 Gonzales TI Jeon JY Lindsay T Resting heart rate is a population-level biomarker of cardiorespiratory fitness:The Fenland Study PLoS One 2023 18 5 e0285272 37167327
21 Prior SJ Ryan AS Blumenthal JB Watson JM Katzel LI Goldberg AP Sarcopenia Is Associated With Lower Skeletal Muscle Capillarization and Exercise Capacity in Older Adults J Gerontol A Biol Sci Med Sci 2016 71 8 1096 101 26888434
22 Kim CH Wheatley CM Behnia M Johnson BD The Effect of Aging on Relationships between Lean Body Mass and VO2max in Rowers PLoS One 2016 11 8 e0160275 27479009
23 Gaillard C Alix E Salle A Berrut G Ritz P Energy requirements in frail elderly people:a review of the literature Clin Nutr 2007 26 1 16 24 17034905
