Osteosarcopenia, bone-muscle interactions, and Frailty risk: A Prospective Cohort Study of Community- Dwelling Older Adults

doi:10.21203/rs.3.rs-6759229/v1

Download PDF

Research Article

Osteosarcopenia, bone-muscle interactions, and Frailty risk: A Prospective Cohort Study of Community- Dwelling Older Adults

https://doi.org/10.21203/rs.3.rs-6759229/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 10 Dec, 2025

Read the published version in Aging Clinical and Experimental Research →

You are reading this latest preprint version

Background

Osteosarcopenia, defined as the coexistence of osteopenia/osteoporosis and sarcopenia, may influence frailty risk in older adults. However, the longitudinal association between osteosarcopenia and its components with frailty remains unclear. This study aimed to address this.

Methods

Data from a prospective cohort study of community-dwelling adults in Australia. Frailty was defined by the presence of three or more components based on Fried criteria: exhaustion, slow gait speed, low grip strength, unintentional weight loss, and low physical activity. Osteosarcopenia was defined by osteopenia/osteoporosis (WHO criteria) and sarcopenia (European Working Group on Sarcopenia in Older People [EWGSOP2] and Sarcopenia Definition and Outcome consortium [SDOC]). Multivariable logistic regression models evaluated the associations between osteosarcopenia or its components with frailty.

Results

Of 301 participants enrolled, 151 (mean age: 65.1 years, 59.6% women) completed a follow-up (median: 4.5 years). Osteosarcopenia was associated with frailty risk irrespective of the definition used: EWGSOP2: OR = 8.13, 95% CI 1.61–41.10; SDOC OR = 8.0, 95% CI 1.42–45.06). Among its components, low grip strength (OR = 0.92, 95% CI 0.84–1.00) and slow gait speed (OR = 0.67, 95% CI 0.53–0.85) were independently associated with frailty; bone mineral density/lean mass were not. An interaction between bone and muscle health and frailty risk was observed (p < 0.011), whereby higher values of grip strength were protective against frailty when bone density was lower (the opposite was also true).

Conclusion

In this prospective cohort study, osteosarcopenia increased the risk of frailty. Our interaction analysis suggests therapies targeting bone density and grip strength may mitigate against frailty.

Aging

osteosarcopenia

sarcopenia

osteoporosis

frailty

Frailty is an age-related syndrome characterized by diminished physiological reserve and resilience, increasing susceptibility to stressors, and the risk of hospitalization, disability, and mortality (1). It results from senescence-related cellular and subcellular changes that disrupt key physiological systems, including the musculoskeletal system, hypothalamic-pituitary-adrenal (HPA) axis, and metabolic function. These disruptions give rise to the Fried frailty phenotype, defined when three or more of the following five criteria are present: unintentional weight loss, exhaustion, slow gait speed, reduced grip strength, and low physical activity (2). As the global population ages, frailty is becoming increasingly prevalent, with significant implications for individual health and healthcare systems (3).

Osteosarcopenia is a recently defined clinical entity that describes the co-occurrence of osteopenia or osteoporosis (low bone mineral density) and sarcopenia (low muscle mass and strength) (4). Each of these conditions individually contributes to an increased risk of falls, fractures, and physical functional decline in older adults; however, their combined impact is synergistic and accelerates musculoskeletal deterioration in older people (5). The pathophysiological basis of osteosarcopenia stems from impaired bone-muscle crosstalk and shared mechanisms such as chronic inflammation, hormonal dysregulation, and environmental factors involving both bone and muscle (6). Through its compounded impact on bone and muscle, osteosarcopenia may contribute to the development of frailty.

Several cross-sectional studies have reported an association between osteosarcopenia and frailty, suggesting a potential biological and clinical link (7–14). However, the inherent limitations of cross-sectional designs preclude conclusions about causality or temporal sequence. It remains uncertain whether osteosarcopenia precedes frailty or if both arise concurrently in response to common age-related mechanisms. To date, longitudinal data exploring osteosarcopenia as a potential predictor of frailty are limited.

To address this gap, the present study aimed to: (1) determine whether osteosarcopenia at baseline predicts the onset of frailty over time; (2) investigate whether the individual components of osteosarcopenia—specifically, low bone mass and reduced muscle strength—independently predict frailty; and (3) explore whether bone-muscle interaction contributes to frailty development. We hypothesize that osteosarcopenia, due to its combined impact on bone and muscle, is an independent and significant predictor of frailty in older adults.

Study Design

The Brimbank Ageing Well Study (BRAW) is a prospective cohort study that investigated the association between baseline osteosarcopenia and frailty among community-dwelling older adults aged ≥ 50 years in the Western suburbs of Melbourne, Australia. Participants were recruited through convenience sampling and assessed at two time points: baseline (2018–2020) and follow-up (2023–2024). Inclusion criteria were aged ≥ 50 years, living in the Western suburbs of Melbourne, and the ability to attend the study site for assessments. Exclusion criteria included inability to provide informed consent; severe cognitive impairment (dementia or dementia with Lewy body); medical conditions affecting balance or physical performance (e.g., Parkinson’s disease, progressive neurological disorders, multiple sclerosis, uncontrolled psychiatric disorders); severe arthritis, chronic lung disease requiring oxygen or renal disease requiring dialysis; history of stroke or recent major surgeries (e.g., knee replacement, spinal, or cardiac surgery) within the past 6 months; life expectancy < 12 months; or body weight > 150 kg (a contraindication for DEXA scanning). Written informed consent was obtained from all participants. The study was approved by the Human Research Ethics Committee at Melbourne Health (HREC/45024/MH-2020).

Study Assessments

Participants attended an assessment session at Sunshine Hospital that lasted approximately 2 to 2.5 hours. They were provided with a self-administered questionnaire capturing sociodemographic information (age, sex, country of birth, living arrangements, income, education, quality of life scores), lifestyle factors (smoking and alcohol consumption history), and physical activity levels using the International Physical Activity Questionnaire (IPAQ) (15). Interpreters were available for non-English speaking participants.

Subsequently, study personnel collected clinical data, including comorbidities (using the Charlson Comorbidity Index), medication use, mini nutritional assessment, and fracture history. Physical assessments included anthropometric measurements (height, weight, body mass index), handgrip strength (using a handheld dynamometer), sit-to-stand test, gait speed, Timed Up and Go (TUG) test, and the Short Physical Performance Battery (SPPB). A whole-body Dual-Energy X-ray Absorptiometry (DEXA) scan was performed to assess Bone Mineral Density (BMD), body composition, and Appendicular Lean Mass adjusted for height (ALM/height²). Finally, participants underwent blood tests to assess serum vitamin D and C-Reactive Protein (CRP) levels. Blood tests were conducted only at baseline; all other assessments were conducted at both visits.

Osteosarcopenia

Osteosarcopenia was defined by the concomitant presence of osteopenia or osteoporosis and sarcopenia (5). Osteopenia and osteoporosis were defined using the World Health Organization (WHO) criteria based on BMD measured via DEXA. Osteoporosis was defined as a BMD T-score of ≤ -2.5, while osteopenia was classified as a BMD T-score between − 1.0 and − 2.5 (16). The lowest recorded BMD at three sites, lumbar spine, femoral neck, and distal 1/3rd of the radius, was used to define osteopenia or osteoporosis. Sarcopenia was defined using two criteria: the European Working Group on Sarcopenia in Older People, revised (EWGSOP2, 2019), and the Sarcopenia Definition and Outcomes Consortium (SDOC, 2020). The EWGSOP2 criteria were employed, as the majority of our study population was Australian-born Caucasian, and these criteria have recently been endorsed for use in Australia via the Delphi convention (17). SDOC criteria are recommended in Caucasians. According to EWGSOP2, probable sarcopenia was identified by either a low handgrip strength (< 27 kg in men and < 16 kg in women) or a five-times sit-to-stand time of > 15 seconds. Confirmed sarcopenia required the presence of probable sarcopenia along with DXA evidence of low ALM/height (< 7.0 kg/m² in men and < 5.5 kg/m² in women) (18). As per SDOC, sarcopenia was defined based on a low handgrip strength (< 35.5 kg in men and < 20 kg in women) and slow gait speed (< 0.8 m/s). A second SDOC definition was used where handgrip strength was adjusted for BMI (19).

Frailty

The validated criterion for frailty, the Fried criteria, was used, which classified individuals as frail if they met at least three out of five criteria, prefrail if they met two or one, and non-frail if they met none. The five criteria included:

1) Unintentional weight loss of > 4.5 kg in the past year or a BMI < 18 kg/m², modified as in previous studies (20–22)

2) Self-reported exhaustion (assessed using two questions from the CES-D depression scale, if the participants provided a positive answer to any of following two questions, i) I felt that everything I did was an effort, for ≥ 3 days in a week, ii) I could not get going, for ≥ 3 days in a week)

3) Low physical activity, defined based on the IPAQ score, categorized as low physical activity, using the gender-specific lowest quintile for the study population.

4) Slowness (gait speed < 1.0 m/s)

5) Weakness (grip strength in the lowest 20% by gender and BMI) (23)

Co-variates

Covariates were selected based on previous evidence and studies suggesting a potential influence on the exposure (osteosarcopenia) and outcome (frailty) variables (24). These included demographic factors (age, sex, living status), socioeconomic factors (education level, income status), and lifestyle behaviors (alcohol consumption, smoking status). Annual income was classified as binary, categorizing those earning less than AUD 40,000 and those earning more, based on an Australian couple’s full age pension of approximately AUD 39,640 per year (25). Clinical variables such as comorbidities, current medications, serum levels of vitamin D, C-reactive protein (CRP), and nutritional status were also considered.

Statistical Analysis

Descriptive statistics were used to summarize the baseline characteristics of study participants. Categorical variables (frequencies and percentages) were compared using chi-square tests, and continuous variables (means ± SD or medians ± IQR for skewed data) using one-way ANOVA. Participants were classified into two groups (non-frail and frail), and frailty transition was reported as a percentage. At baseline, participants were categorized into four mutually exclusive groups: (1) no condition (neither osteopenia/osteoporosis nor sarcopenia), (2) sarcopenia only, (3) osteopenia or osteoporosis only, and (4) osteosarcopenia.

Logistic regression was used to estimate the odds of developing frailty at follow-up, comparing each condition group to the no-condition group. Due to the limited number of frailty cases for multivariable analyses, participants were stratified into two groups: those with osteosarcopenia and those without. To minimize overfitting, each confounder was adjusted individually in separate models, adhering to the recommended guideline of at least 10 events per variable in logistic regression analyses (26). Confounders were selected based on significant associations with frailty in univariate analyses and included demographic (age, sex), lifestyle (smoking, alcohol use, physical activity), clinical (polypharmacy, vitamin D, CRP), and socioeconomic (education, income) factors.

Muscle parameters (grip strength, gait speed, five-times sit-to-stand test, SPPB, ALM, ALM/height²) and bone measures (BMD and T-scores at various sites) were analyzed as continuous variables using logistic regression to assess associations with frailty at follow-up. Models were adjusted for sex, physical activity, and income, based on initial findings.

Interaction analysis models were employed to assess whether muscle parameters significantly interacted with bone measures in predicting frailty. If significant interactions were identified, indicating effect modification by value level, no further analysis was conducted. Otherwise, mutually adjusted models were used to assess the independent effects of muscle and bone measures. Results are presented as odds ratios (OR) with 95% confidence intervals (CI). A p-value < 0.05 was considered statistically significant, except for interactions where p < 0.06 was used, aligning with exploratory research practices that recognize the value of marginal trends (27). All analyses were conducted using Stata version 18 (StataCorp, College Station, TX, USA).

Participants Characteristics

Out of 300 participants at baseline, 151 completed follow-ups (reasons detailed in Supplementary Figure S1). The characteristics of participants who attended and those who did not attend are compared in Supplementary Table 1. Of those who completed follow-ups, 143 were non-frail and eight were frail at baseline assessment. Over the study follow-up, out of the non-frail, 13 (9.1%) became frail; of the frail, 4 (50%) improved to non-frail. The overall frailty incidence in this cohort was 6.6%.

Table 1 shows the baseline characteristics of 151 participants categorized as robust, prefrail, or frail. Eight frail participants at baseline were excluded from follow-up analysis, leaving 143 non-frail individuals for the final analysis.

Table 1

Summary of Participants’ Characteristics at Baseline, Categorized by Frailty (Three Groups)
	Fried Frailty categories
	Robust	Pre-Frail	Frail	Total	p-value
Numbers	97 (64.2%)	46 (30.5%)	8 (5.3%)	151 (100.0%)
Age at first visit	64.5 (5.9)	66.3 (7.1)	65.7 (6.7)	65.1 (6.3)	0.398
Women	56 (57.7%)	30 (65.2%)	4 (50.0%)	90 (59.6%)	0.592
Lives alone	19 (19.6%)	12 (26.1%)	3 (37.5%)	34 (22.5%)	0.398
Born in Australia	68 (70.1%)	30 (65.2%)	3 (37.5%)	101 (66.9%)	0.163
ATSI	0 (0.0%)	1 (2.2%)	0 (0.0%)	1 (0.7%)	0.317
Speak English at home	94 (96.9%)	44 (95.7%)	7 (87.5%)	145 (96.0%)	0.419
Education
Post-secondary/tertiary qualification	82 (84.5%)	35 (76.1%)	5 (62.5%)	122 (80.8%)	0.196
High school or less	15 (15.5%)	11 (23.9%)	3 (37.5%)	29 (19.2%)
Income
Annual income > 40K AUD	77 (79.4%)	30 (65.2%)	2 (25.0%)	109 (72.2%)	0.002
Has private health insurance	79 (81.4%)	23 (50.0%)	2 (25.0%)	104 (68.9%)	< 0.001
Health card holder	24 (24.7%)	21 (45.7%)	6 (75.0%)	51 (33.8%)	0.002
Smoking
Smoking (past or present)	42 (43.3%)	21 (45.7%)	3 (37.5%)	66 (43.7%)	0.904
Smoking (current)
No, not at all	41 (91.1%)	20 (95.2%)	3 (75.0%)	64 (91.4%)	0.614
Yes, but not daily	3 (6.7%)	1 (4.8%)	1 (25.0%)	5 (7.1%)
Yes, daily	1 (2.2%)	0 (0.0%)	0 (0.0%)	1 (1.4%)
Duration smoking (years)	[3] 23.0 (12.8)	[3] 23.3 (20.8)	[1] 55.0 (.)	[7] 27.7 (18.5)	0.319
Smoking (past)	38 (80.9%)	21 (91.3%)	2 (50.0%)	61 (82.4%)	0.120
Alcohol
Alcohol (past or present)	94 (96.9%)	43 (93.5%)	6 (75.0%)	143 (94.7%)	0.026
Alcohol (last month)	79 (83.2%)	32 (74.4%)	5 (71.4%)	116 (80.0%)	0.417
Alcohol amount last year
No days	10 (10.4%)	6 (14.0%)	1 (14.3%)	17 (11.6%)	0.567
Less than once a month	11 (11.5%)	9 (20.9%)	2 (28.6%)	22 (15.1%)
One to three days per month	23 (24.0%)	12 (27.9%)	2 (28.6%)	37 (25.3%)
One to four days per week	31 (32.3%)	12 (27.9%)	1 (14.3%)	44 (30.1%)
Five or more days per week	21 (21.9%)	4 (9.3%)	1 (14.3%)	26 (17.8%)
Alcohol (average) over 12 months	[90] 2.2 (3.2)	[40] 1.7 (1.0)	[6] 2.8 (3.5)	[136] 2.1 (2.7)	0.437
Comorbidities
Any comorbidity	70 (72.2%)	38 (82.6%)	7 (87.5%)	115 (76.2%)	0.290
Arthritis	33 (34.0%)	17 (37.0%)	3 (37.5%)	53 (35.1%)	0.933
Chronic Back pain	26 (26.8%)	18 (39.1%)	2 (25.0%)	46 (30.5%)	0.308
Cardiovascular Disease	9 (9.3%)	12 (26.1%)	1 (12.5%)	22 (14.6%)	0.029
Chronic Lung disease	5 (5.2%)	7 (15.2%)	2 (25.0%)	14 (9.3%)	0.044
Cancer	4 (4.1%)	4 (8.7%)	2 (25.0%)	10 (6.6%)	0.059
Depression or Anxiety	7 (7.2%)	8 (17.4%)	2 (25.0%)	17 (11.3%)	0.089
Diabetes	4 (4.1%)	6 (13.0%)	1 (12.5%)	11 (7.3%)	0.134
Cerebrovascular disease	1 (1.0%)	2 (4.3%)	1 (12.5%)	4 (2.6%)	0.105
Other comorbidities	16 (16.5%)	9 (19.6%)	1 (12.5%)	26 (17.2%)	0.844
Previous fragility fractures
Any fracture in last 5 years	13 (13.8%)	8 (17.8%)	2 (25.0%)	23 (15.6%)	0.631
Vertebral fracture	2 (2.1%)	2 (4.3%)	0 (0.0%)	4 (2.7%)	0.655
Non-vertebral fractures (exclude hip)	49 (50.5%)	23 (50.0%)	6 (75.0%)	78 (51.7%)	0.397
Hip fractures	2 (2.1%)	4 (8.7%)	0 (0.0%)	6 (4.0%)	0.143
Physical Activity Scale (IPAQ)
Low	0 (0.0%)	22 (47.8%)	8 (100.0%)	30 (19.9%)	< 0.001
Moderate	19 (19.6%)	5 (10.9%)	0 (0.0%)	24 (15.9%)
High	78 (80.4%)	19 (41.3%)	0 (0.0%)	97 (64.2%)
Nutrition (MNA)
At risk of malnutrition	2 (8.0%)	4 (28.6%)	0 (0.0%)	6 (15.0%)	0.206
Medications
Any medication (exclude nutritional supplements)	65 (67.0%)	31 (67.4%)	8 (100.0%)	104 (68.9%)	0.148
Polypharmacy ( > = 5 medications)	10 (10.3%)	9 (19.6%)	2 (25.0%)	21 (13.9%)	0.212
Psychological Distress Scale (Kessler K6+)
No/low distress	95 (97.9%)	36 (80.0%)	4 (66.7%)	135 (91.2%)	< 0.001
Moderate distress	2 (2.1%)	6 (13.3%)	2 (33.3%)	10 (6.8%)
High distress	0 (0.0%)	3 (6.7%)	0 (0.0%)	3 (2.0%)
Clinical & anthropometric data
Systolic BP (mmHg)	135.9 (18.7)	133.5 (18.7)	136.1 (14.6)	135.2 (18.4)	0.792
Diastolic BP (mmHg)	85.3 (10.4)	84.5 (11.3)	91.9 (10.9)	85.4 (10.7)	0.211
Body Mass Index (Kg/m²)	27.1 (4.1)	28.4 (6.6)	33.8 (6.3)	27.8 (5.3)	0.011
Hip circumference (cm)	104.7 (8.9)	102.9 (16.9)	120.6 (17.3)	105.0 (12.8)	0.020
Waist circumference (cm)	91.7 (14.6)	95.3 (18.9)	114.8 (17.3)	94.0 (16.8)	0.002
Pathology
Vitamin D levels	67.9 (20.3)	59.6 (22.4)	83.1 (37.3)	66.2 (22.6)	0.127
Hs-CRP	1.5 (2.0)	3.0 (3.7)	6.0 (6.3)	2.1 (3.1)	0.004
Physical measurements
Hand Grip Strength Max (Kg)	37.8 (8.1)	29.7 (10.0)	24.4 (8.3)	34.6 (9.8)	< 0.001
Hand Grip Strength/BMI	1.4 (0.3)	1.1 (0.4)	0.8 (0.3)	1.3 (0.4)	< 0.001
5x Sit to Stand (sec)	10.8 (7.8)	14.8 (13.0)	17.3 (5.1)	12.3 (9.8)	< 0.001
Short Physical Performance battery	11.5 (0.9)	10.6 (2.0)	8.8 (2.0)	11.1 (1.5)	< 0.001
Gait speed best (m/s)	1.4 (0.2)	1.2 (0.3)	0.7 (0.2)	1.3 (0.3)	< 0.001
Muscle measurement on DXA
ALM	20.9 (5.1)	19.5 (5.7)	21.0 (6.1)	20.5 (5.4)	0.181
ALM/Height	7.3 (1.2)	7.0 (1.4)	7.8 (1.7)	7.3 (1.3)	0.247
Subtotal bodyfat, %age	34.4 (8.3)	37.1 (9.0)	43.7 (11.9)	35.7 (9.0)	0.031
Whole body fat, %age	33.8 (7.8)	36.3 (8.6)	42.7 (11.5)	35.0 (8.5)	0.032
Whole body fat, kg	26.8 (8.7)	29.4 (11.2)	41.5 (16.7)	28.3 (10.5)	0.012
Bone measurement on DXA
Lowest site BMD (g/cm2)	[97] 0.5 (0.1)	[46] 0.5 (0.1)	[8] 0.5 (0.1)	[151] 0.5 (0.1)	0.892
Lumbar spine BMD (g/cm2)	[97] 1.0 (0.2)	[46] 1.0 (0.2)	[8] 1.1 (0.2)	[151] 1.0 (0.2)	0.366
Femoral Neck BMD (g/cm2)	[96] 0.8 (0.1)	[46] 0.7 (0.1)	[8] 0.8 (0.1)	[150] 0.7 (0.1)	0.497
Hip total BMD (g/cm2)	[96] 0.9 (0.1)	[46] 0.9 (0.2)	[8] 0.9 (0.1)	[150] 0.9 (0.2)	0.687
Distal 1/3 radius BMD (g/cm2)	[97] 0.7 (0.1)	[45] 0.7 (0.1)	[8] 0.7 (0.1)	[150] 0.7 (0.1)	0.295
Lowest site T-score	[97] -1.3 (1.0)	[46] -1.5 (1.2)	[8] -1.2 (1.0)	[151] -1.3 (1.0)	0.589
Lumbar spine (T-score)	[97] -0.6 (1.4)	[46] -0.7 (1.7)	[8] 0.1 (1.6)	[151] -0.6 (1.5)	0.477
Femoral neck (T-score)	[95] -1.1 (0.9)	[46] -1.2 (1.0)	[8] -1.1 (1.1)	[149] -1.1 (0.9)	0.687
Distal 1/3 radius (T-score)	[96] -0.6 (1.2)	[46] -1.0 (1.3)	[6] -1.3 (1.2)	[148] -0.8 (1.2)	0.173
Fried criteria – individual measurements
Exhaustion	0 (0.0%)	4 (8.7%)	2 (25.0%)	6 (4.0%)	< 0.001
Weight loss	0 (0.0%)	7 (15.2%)	0 (0.0%)	7 (4.6%)	< 0.001
Low physical activity	0 (0.0%)	22 (47.8%)	8 (100.0%)	30 (19.9%)	< 0.001
Slowness	0 (0.0%)	12 (26.1%)	8 (100.0%)	20 (13.2%)	< 0.001
Weakness	0 (0.0%)	17 (37.0%)	6 (75.0%)	23 (15.2%)	< 0.001
Osteoporosis (WHO criteria)
Normal	33 (34.0%)	17 (37.0%)	4 (50.0%)	54 (35.8%)	0.753
Osteopenia	54 (55.7%)	22 (47.8%)	3 (37.5%)	79 (52.3%)
Osteoporosis	10 (10.3%)	7 (15.2%)	1 (12.5%)	18 (11.9%)
Sarcopenia categorization
Sarcopenia EWGSOP	8 (8.2%)	13 (28.3%)	7 (87.5%)	28 (18.5%)	< 0.001
Sarcopenia EWGSOP (severe)	2 (2.1%)	7 (15.2%)	4 (50.0%)	13 (8.6%)	< 0.001
Sarcopenia SDOC 1	0 (0.0%)	18 (39.1%)	8 (100.0%)	26 (17.2%)	< 0.001
Sarcopenia SDOC 2	4 (4.1%)	17 (37.0%)	6 (75.0%)	27 (17.9%)	< 0.001
Osteosarcopenia categorization
Osteosarcopenia (EWGSOP2) mutually exclusive groups
Healthy	28 (28.9%)	14 (30.4%)	0 (0.0%)	42 (27.8%)	< 0.001
Osteopenia/osteoporosis	61 (62.9%)	19 (41.3%)	1 (12.5%)	81 (53.6%)
Sarcopenia	5 (5.2%)	3 (6.5%)	4 (50.0%)	12 (7.9%)
Osteosarcopenia	3 (3.1%)	10 (21.7%)	3 (37.5%)	16 (10.6%)
Osteosarcopenia (SDOC 1) mutually exclusive groups
Healthy	33 (34.0%)	11 (23.9%)	0 (0.0%)	44 (29.1%)	< 0.001
Osteopenia/osteoporosis	64 (66.0%)	17 (37.0%)	0 (0.0%)	81 (53.6%)
Sarcopenia	0 (0.0%)	6 (13.0%)	4 (50.0%)	10 (6.6%)
Osteosarcopenia	0 (0.0%)	12 (26.1%)	4 (50.0%)	16 (10.6%)
Osteosarcopenia (SDOC 2) mutually exclusive groups
Healthy	30 (30.9%)	9 (19.6%)	1 (12.5%)	40 (26.5%)	< 0.001
Osteopenia/osteoporosis	63 (64.9%)	20 (43.5%)	1 (12.5%)	84 (55.6%)
Sarcopenia	3 (3.1%)	8 (17.4%)	3 (37.5%)	14 (9.3%)
Osteosarcopenia	1 (1.0%)	9 (19.6%)	3 (37.5%)	13 (8.6%)
Unless otherwise specified, mean (SD) for continuous variables and number (%age) for categorical variables. [] before reading shows number of observation. ALM Appendicular lean mass, ALM/height Appendicular lean mass adjusted for height, ATSI Aboriginal and Torres Strait Island, AUD Australian Dollar, BMD Bone Muscle Density, BP blood pressure, DXA Dual energy X-ray absorptiometry, EWGSOP2 European Working Group of Sarcopenia in Older People 2, Hs CRP High sensitivity C-Reactive Protein, IPAQ International Physical Activity Questionnaire, MNA Mini Nutritional Assessment, SDOC 1 Sarcopenia Definition and Outcome Consortium, SDOC 2 SDOC adjusted for BMI, WHO World Health Organization.

The mean age of participants was 65.1 years (SD 6.3), with 59.6% female. No age or sex differences were found between frailty groups. Most participants lived with others (78%), were Australian-born (66.9%), English-speaking (96%), and had completed secondary school education or higher (> 80.8%). Socioeconomic differences were noted across frailty groups when classified by an annual income > 40K AUD (p = 0.002), private health insurance (p < 0.001), and health care card ownership (p = 0.002). Differences were also noted between groups in lifestyle factors; alcohol use (p = 0.026), physical activity (IPAQ, p < 0.001), and inflammatory markers, hs-CRP levels (p = 0.004). No differences were found for smoking, comorbidities, fractures, medications, polypharmacy, or vitamin D. While muscle function measures (grip strength, gait speed) differed between groups, muscle and bone mass (ALM, ALM/height², BMD, T-scores) did not. Supplementary Figure S2 shows the overlap of osteopenia/osteoporosis and sarcopenia at baseline. Each condition (osteosarcopenia, sarcopenia and osteopenia/osteoporosis) was analyzed as a distinct group to assess its association with frailty at follow-up.

Association Between Osteosarcopenia and Frailty

Using the EWGSOP2 criteria, osteosarcopenia was significantly associated with increased odds of developing frailty compared to individuals without osteopenia/osteoporosis or sarcopenia (OR = 8.13, 95% CI: 1.61–41.10, p = 0.011). Sarcopenia alone at baseline was associated with higher odds of frailty (OR = 1.86, 95% CI: 0.17–20.51, p = 0.613), though this was not statistically significant. Interestingly, baseline osteopenia or osteoporosis was associated with lower odds of frailty (OR = 0.68, 95% CI: 0.15–3.21, p = 0.63), although this result was also not significant. Similar patterns were observed using the SDOC definition, with osteosarcopenia again showing a significant association with frailty (OR = 8.00, 95% CI: 1.42–45.06, p = 0.018), while associations for sarcopenia and osteopenia/osteoporosis remained non-significant (Table 2).

Table 2

Association of Baseline Osteosarcopenia (and its components) with Frailty Risk in a Longitudinal Cohort.
Frailty vs non-frailty at follow-up	Total number at follow up	Non-Frail at follow up	Frail at follow up	Unadjusted – univariable analysis
	n = 143	n = 13	n = 13	OR	95% CI	p-value
European Working Group of Sarcopenia in older People – Revised (EWGSOP2) Definition
Non-osteopenic/non-sarcopenic (control group)	42	39	3	Ref
Osteopenia or osteoporosis only	80	76	4	0.68	[0.15,3.21]	0.63
Sarcopenia only	8	7	1	1.86	[0.17,20.51]	0.613
Osteosarcopenia	13	8	5	8.13	[1.61,41.10]	0.011
Sarcopenia Definition and Outcome Consortium (SDOC) Definition
Non-osteopenic/non-sarcopenic (control group)	39	36	3	Ref
Osteopenia or osteoporosis only	83	78	5	0.77	[0.17,3.40]	0.729
Sarcopenia only	11	10	1	1.2	[0.11,12.83]	0.88
Osteosarcopenia	10	6	4	8	[1.42,45.06]	0.018

In a multivariable analysis comparing two groups, osteosarcopenia to non-osteosarcopenia, osteosarcopenia (EWGSOP2 definition) was significantly associated with higher odds of frailty (OR = 9.53, 95% CI: 2.53–35.92, p = 0.001). Models were individually adjusted for age, sex, polypharmacy, smoking, alcohol use, education, income, physical activity, and CRP levels. Osteosarcopenia remained a consistent predictor of frailty across all models. Among the covariates, only an income greater than AUD 40,000/year was protective; other variables showed no significant associations (Table 3). Using the SDOC criteria (adjusted for BMI), osteosarcopenia was again significantly associated with increased odds of frailty (OR = 9.19, 95% CI: 2.19–38.56, p = 0.002). This association remained robust in multivariable models. Similar to the EWGSOP2 findings, an income above AUD 40,000/year was the only covariate significantly associated with a reduced risk of frailty (Table 3).

Table 3

Association of Baseline Osteosarcopenia with Frailty Risk - Multivariable Analysis
Osteosarcopenia vs non-osteosarcopenia					Frail vs non-frail (dependent variable)
Osteosarcopenia vs non-osteosarcopenia	Total n	Non-Frail n	Frail n					Multivariable analysis adjusted for each variable
	143	130	13		OR	95% CI	p-value		OR	95% CI	p-value
Osteosarcopenia (by EWGSOP2)	13	8	5	Unadjusted	9.53	[2.53,35.92]	0.001
				Adjusted	8.28	[2.08,32.91]	0.003	Age at first visit	1.04	[0.94,1.14]	0.481
	57	53	4		9.3	[2.46,35.20]	0.001	Sex - Men	0.72	[0.20,2.60]	0.612
	19	16	3		9.16	[2.27,36.98]	0.002	Polypharmacy	1.16	[0.24,5.64]	0.858
	63	60	3		8.87	[2.30,34.14]	0.002	Smoking	0.39	[0.10,1.56]	0.183
	137	125	12		9.56	[2.45,37.31]	0.001	Alcohol	1.03	[0.08,12.40]	0.984
	117	107	10		9.81	[2.49,38.70]	0.001	Post-secondary qualification	1.13	[0.25,5.19]	0.871
	107	101	6		12.68	[2.91,55.15]	0.001	Income AUD > 40K/yr	0.19	[0.05,0.69]	0.012
	143	93	4		11.91	[2.57,55.09]	0.002	High Physical activity	0.29	[0.06,1.49]	0.139
	138				10.95	[2.63,45.50]	0.001	Hs-CRP	1.16	[0.98,1.37]	0.093
Osteosarcopenia (by SDOC2)	10	6	4	Unadjusted	9.19	[2.19,38.56]	0.002
				Adjusted	8.52	[1.97,36.84]	0.004	Age at first visit	1.06	[0.96,1.16]	0.244
	57	43	4		8.8	[2.06,37.65]	0.003	Sex - Men	0.8	[0.22,2.90]	0.738
	19	16	3		8.92	[2.10,37.88]	0.003	Polypharmacy	1.98	[0.45,8.68]	0.367
	63	60	3		8.52	[1.98,36.69]	0.004	Smoking	0.38	[0.10,1.52]	0.172
	137	125	12		9.92	[2.32,42.43]	0.002	Alcohol	0.34	[0.03,3.23]	0.345
	117	107	10		9.3	[2.12,40.80]	0.003	Post-secondary qualification	1.05	[0.23,4.74]	0.946
	107	101	6		10.14	[2.20,46.86]	0.003	Income AUD > 40K/yr	0.23	[0.07,0.79]	0.020
	143	93	4		11.45	[2.29,57.32]	0.003	High Physical activity	0.22	[0.04,1.07]	0.060
	138				7.79	[1.55,39.05]	0.013	Hs-CRP	1.10	[0.92,1.31]	0.303

Association Between Bone and Muscle Measures and Frailty

Clinical measures of muscle strength, handgrip strength (both unadjusted and BMI-adjusted), gait speed, and the SPPB were significantly associated with frailty at follow-up. In contrast, DEXA-derived measures of muscle (ALM and ALM/height²) and bone (BMD and T-scores) showed no significant associations. The models were adjusted for sex, income, and physical activity, based on significant differences observed in the initial analysis (Table 4).

Table 4

Association of Baseline Bone and Muscle Measures with Frailty Risk in a longitudinal cohort.
	Frailty vs non-frailty at follow-up (13 vs 130)		Model 1			Model 2			Model 3
		n	OR	95% CI	p-value	OR	95% CI	p-value	OR	95% CI	p-value
Muscle	Hand Grip Strength Max (Kg)	143	0.92	[0.84,1.00]	0.038	0.93	[0.87,1.00]	0.057	0.94	[0.87,1.01]	0.085
	Hand Grip Strength/BMI	143	0.15	[0.03,0.90]	0.037	0.2	[0.04,0.97]	0.046	0.24	[0.04,1.28]	0.095
	Gait speed best(m/s) [0.1]	143	0.67	[0.53,0.85]	0.001	0.69	[0.54,0.88]	0.003	0.64	[0.49,0.84]	0.001
	5x Sit to Stand (sec)	143	1.04	[1.00,1.08]	0.077	1.04	[1.00,1.09]	0.043	1.03	[0.99,1.07]	0.152
	Short Physical Performance Battery	142	0.58	[0.42,0.81]	0.001	0.59	[0.43,0.82]	0.001	0.6	[0.43,0.83]	0.002
	ALM (kg)	143	0.94	[0.78,1.14]	0.545	0.94	[0.84,1.05]	0.301	0.95	[0.84,1.07]	0.361
	ALM/Height (kg/m2)	143	0.75	[0.38,1.47]	0.396	0.75	[0.46,1.20]	0.23	0.75	[0.45,1.25]	0.268
	Subtotal body fat, %age	143	1.05	[0.96,1.14]	0.308	1.05	[0.98,1.13]	0.134	1.02	[0.96,1.10]	0.489
Bone	Lowest site BMD (g/cm2) [0.5]	143	1.22	[0.04,33.96]	0.906	0.48	[0.02,10.93]	0.648	0.47	[0.02,11.80]	0.645
	Lumbar spine BMD (g/cm2) [0.5]	143	0.92	[0.15,5.56]	0.924	0.73	[0.14,3.93]	0.719	0.56	[0.09,3.30]	0.521
	Femoral Neck BMD (g/cm2) [0.5]	142	0.14	[0.01,2.75]	0.195	0.11	[0.05,1.81]	0.122	0.08	[0.00,1.75]	0.108
	Hip total BMD (g/cm2) [0.5]	142	0.31	[0.03,3.13]	0.319	0.27	[0.04,2.04]	0.206	0.14	[0.12,1.48]	0.102
	Forearm distal 1/3rd BMD (g/cm2) [0.5]	142	10.83	[0.20,589.19]	0.243	1.01	[0.07,14.46]	0.995	1.59	[0.10,25.0]	0.742
	Lowest site (T-score)	143	0.65	[0.35,1.21]	0.177	0.64	[0.35,1.16]	0.14	0.53	[0.27,1.04]	0.064
	Lumbar Spine (T-score)	143	1.05	[0.71,1.55]	0.796	1.01	[0.69,1.46]	0.976	0.94	[0.63,1.40]	0.767
	Femoral neck (T-score)	141	0.6	[0.29,1.24]	0.168	0.56	[0.27,1.16]	0.121	0.51	[0.23,1.11]	0.09
	Distal 1/3rd forearm (T-score)	142	1.13	[0.69,1.83]	0.626	1.06	[0.64,1.76]	0.814	1.07	[0.65,1.75]	0.791
Model 1 Adjusted for sex
Model 2 Adjusted for income
Model 3 Adjusted for high physical activity score
All analysis represents an increase by one unit unless specified in square brackets []. Gait speed analysis for an increase by 0.1m/s and BMD increase by 0.5g/cm².

A 1 kg increase in handgrip strength was associated with an 8% reduction in frailty odds (OR = 0.92, 95% CI: 0.84–1.00, p = 0.038). When adjusted for BMI, a 1-unit increase in handgrip strength corresponded to an 85% lower risk of frailty (OR = 0.15, 95% CI: 0.03–0.90, p = 0.037). This association remained significant after adjusting for income but was attenuated after further adjustment for physical activity. For gait speed, each 0.1 m/s increase was linked to a 33% reduction in frailty odds (OR = 0.67, 95% CI: 0.53–0.85, p = 0.001), with the association persisting across all models. Similarly, a 1-point improvement in SPPB score was associated with 42% lower odds of frailty (OR = 0.58, 95% CI: 0.42–0.81, p = 0.001), consistent across adjusted models (Table 4).

Bone-Muscle Interaction and Frailty

Bone–muscle interaction analyses were conducted to assess whether the effects of muscle function (handgrip strength, gait speed, sit-to-stand time) on frailty vary by bone strength (Table 5). Handgrip strength (adjusted for sex) showed significant interactions with lumbar spine BMD (p = 0.061), lumbar spine T-scores (p = 0.051), and the lowest site T-score (p = 0.011), the latter being clinically relevant for defining osteopenia and osteoporosis (WHO criteria). These findings suggest that the protective effect of handgrip strength on frailty is more pronounced at lower T-scores and BMD values (Fig. 1a and 1b). In contrast, gait speed showed no significant interaction with bone measures, indicating its association with frailty is independent of bone strength.

Table 5

Interaction Analysis of Baseline Bone and Muscle Measures with Frailty Risk in a Longitudinal Cohort
Muscle Parameter	Bone Parameter	P-value (Interaction)	Main Effect (Muscle) OR	95% CI	P-value	Main Effect (Bone) OR	95% CI	P-value
Hand Grip Strength Max (Kg) adjusted for sex	Hip total BMD (g/cm2)	0.101
	Lumbar spine BMD (g/cm2)	0.061
	Lowest site (T-score)	0.051
	Femoral neck (T-score)	0.592
	Lumbar spine (T-score)	0.011
Gait Speed (m/sec)	Hip total BMD (g/cm2)	0.844	0.02	[0.00,0.20]	0.001	0.06	[0.00,5.78]	0.232
	Lumbar spine BMD (g/cm2)	0.253	0.02	[0.00,0.19]	0.001	0.33	[0.01,12.27]	0.551
	Lowest site (T-score)	0.406	0.02	[0.00,0.20]	0.001	0.6	[0.30,1.18]	0.141
	Femoral neck (T-score)	0.109	0.02	[0.00,0.21]	0.001	0.59	[0.28,1.26]	0.174
	Lumbar spine (T-score)	0.168	0.02	[0.00,0.19]	0.001	0.95	[0.64,1.42]	0.811
STS (sec)	Hip total BMD (g/cm2)	0.983	1.04	[1.00,1.08]	0.051	0.06	[0.00,4.47]	0.202
	Lumbar spine BMD (g/cm2)	0.243	1.04	[1.00,1.08]	0.051	0.39	[0.01,12.67]	0.598
	Lowest site (T-score)	0.838	1.04	[1.00,1.09]	0.032	0.58	[0.31,1.08]	0.088
	Femoral neck (T-score)	0.961	1.04	[1.00,1.09]	0.044	0.59	[0.29,1.24]	0.165
	Lumbar spine (T-score)	0.243	1.04	[1.00,1.08]	0.055	0.97	[0.65,1.42]	0.859
ALM/height (Kg/m2) adjusted for sex	Hip total BMD (g/cm2)	0.819	0.84	[0.40,1.80]	0.661	0.16	[0.00,27.71]	0.485
	Lumbar spine BMD (g/cm2)	0.245	0.73	[0.36,1.48]	0.381	1.52	[0.04,61.62]	0.826
	Lowest site (T-score)	0.500	0.91	[0.42,1.95]	0.809	0.68	[0.34,1.36]	0.275
	Femoral neck (T-score)	0.982	0.88	[0.43,1.81]	0.729	0.63	[0.29,1.35]	0.235
	Lumbar spine (T-score)	0.079	0.7	[0.34,1.42]	0.322	1.12	[0.76,1.67]	0.561

Multivariable logistic regression analyses were performed to assess whether bone and muscle measures independently predicted frailty. In models adjusted for both, gait speed remained significantly associated with frailty, independent of bone strength. Similarly, the sit-to-stand test retained a significant association after adjusting for bone parameters. In contrast, DEXA-derived muscle mass measures (ALM and ALM/height²), adjusted for sex and bone measurements (BMD and T-scores), were not significantly associated with frailty.

In this longitudinal cohort study of older adults followed for an average of 4.5 years, we identified an association between osteosarcopenia and an increased risk of frailty. Participants with osteosarcopenia were compared to those without, and in a separate analysis, each condition (osteosarcopenia, sarcopenia, osteopenia/osteoporosis) was compared to individuals without any condition at baseline. In both analyses, osteosarcopenia was consistently linked to frailty risk. Although sarcopenia showed a trend of increased risk, this did not reach statistical significance, likely due to sample size limitations. Muscle strength measures, such as handgrip strength and gait speed, were stronger predictors of frailty than DXA-derived bone (BMD, T-score) and muscle parameters (ALM and ALM/Height²), which did not predict frailty. Furthermore, an interaction between bone and muscle health and frailty risk was observed, whereby higher grip strength values were protective against frailty when bone density was lower (the opposite was also true). These findings support our hypothesis that osteosarcopenia contributes to frailty risk and emphasize the importance of addressing both bone and muscle health in frailty prevention.

The temporal relationship between osteosarcopenia and frailty remains insufficiently understood due to limited longitudinal data. While cross-sectional studies have consistently demonstrated an association, longitudinal cohort studies are better suited to assess its predictive value over time. We previously reported findings from the I-Lan Longitudinal Study on Aging (ILAS) in Taiwan, where osteosarcopenia (osteopenia/osteoporosis by WHO definition and sarcopenia by AWGS 2019 criteria) was not associated with frailty (Fried criteria) among 998 participants followed over eight years (28). Two additional cohort studies have examined this relationship. A retrospective analysis of the prospective ROAD cohort in Japan found a significant association between osteosarcopenia (sarcopenia by AWGS) and frailty (Fried criteria) over four years (OR 5.80, 95% CI 1.38–24.4, p = 0.017) in community-dwelling adults aged 60 and above (29). In contrast, the Canadian Longitudinal Study on Aging (CLSA) reported no association between osteosarcopenia (sarcopenia by SDOC) and frailty (Rockwood deficit accumulation index) over three years in Caucasian adults aged 65 and older. However, osteosarcopenia was associated with an increased risk of falls, fractures, reduced quality of life, and impaired activities of daily living (30). Our findings match the ROAD study in supporting an association between osteosarcopenia and frailty risk, but differ from the CLSA and ILAS results. These inconsistencies likely reflect variations in cohort characteristics, geographic settings, study follow-up duration, and the definitions of frailty and sarcopenia employed.

Cohort characteristics are critical to study outcomes, with variations in lifestyle factors (e.g., alcohol use, smoking, physical activity, health literacy, socioeconomic status, education) influencing bone and muscle health through their impact on diet, exercise behaviors, and lifestyle choices (31, 32). Geographic differences in comorbidities, medication use, and genetic factors, such as regional variation in insulin resistance and metabolic syndrome, may further impact musculoskeletal outcomes uniquely (33). Inconsistencies in findings across studies may also stem from differing definitions of frailty and sarcopenia. The CLSA employed the Rockwood Frailty Index, based on a deficit accumulation model, while our study, along with the ROAD and ILAS cohorts, used the Fried Frailty Phenotype. Sarcopenia definitions likewise varied: the CLSA used the SDOC criteria, we applied both SDOC and EWGSOP2, and the ROAD and ILAS studies used the AWGS classification. These methodological differences may partly explain the divergent results.

In addition to the differences in cohort characteristics and definitions, frailty and osteosarcopenia are complex, dynamic, and interact bidirectionally with multiple feedback loops, challenging the unidirectional assumptions of longitudinal cohort analyses. Long-term studies with multiple assessment points are needed to clarify this relationship.

We propose that the biological basis of our results may be due to the shared pathways underlying bone–muscle interactions and frailty. Bone–muscle crosstalk, mediated by factors such as insulin-like growth factor and growth hormone implicated in frailty development (6, 34), have been suggested. Additionally, nutrition, inflammation, lifestyle, and socioeconomic factors collectively influence musculoskeletal health and frailty risk (1, 35). These interconnected mechanisms support a multifactorial basis for the association of osteosarcopenia and frailty risk.

The association between osteosarcopenia and frailty risk in our cohort implies that osteosarcopenia may represent an early, pre-frail stage in the frailty continuum. Frailty is known to develop progressively, often through an intermediate ‘pre-frail phase’, and is associated with increased vulnerability and adverse outcomes (36, 37). Osteosarcopenia is linked to adverse outcomes, such as falls, fractures, impaired ADLs, reduced quality of life, and increased mortality, outcomes that are also strongly linked to frailty (30, 38, 39). However, defining pre-frailty remains challenging when using the same criteria due to its clinical overlap with frailty. Unlike the pre-frailty ‘conceptual’ definition, osteosarcopenia offers objective measures of bone and muscle deficits, making it a promising marker for early risk identification. Recognizing it as a precursor could enable earlier, targeted interventions to prevent or delay frailty progression in both clinical and research contexts.

Our study observed a non-significant trend toward a positive association between sarcopenia and frailty, and an inverse trend for osteoporosis. Although limited by sample size, a possible explanation is that osteoporosis patients were more likely to receive pharmacotherapy to improve BMD, which may enhance muscle strength, potentially mitigating frailty risk.

Our secondary analysis examining the relationship between bone and muscle parameters and frailty demonstrated that muscle parameters, specifically handgrip strength, gait speed, the five-times sit-to-stand test, and the SPPB, were associated with frailty. In contrast, bone parameters, represented by BMD and T-scores obtained via DXA, did not show significant associations. One limitation of these analyses is that handgrip strength and gait speed are integral components of the Fried frailty phenotype, and their inclusion in the frailty definition may partly influence the observed associations. Moreover, the SPPB is a composite measure incorporating gait speed, handgrip strength, and the five-times sit-to-stand test, which complicates the interpretation of its independent association with frailty. Nevertheless, by analyzing these parameters as continuous variables, rather than dichotomizing them, as included in sarcopenia classification, we were able to more precisely assess the influence of handgrip strength and gait speed on frailty risk.

These findings align with previous research, such as the Geelong Osteoporosis Study (GOS), which found that muscle parameters (lower limb strength and lean mass index) were better predictors of frailty than bone parameters (40). A previous 10-year longitudinal study (Osteoporosis Risk Assessment, OPRA) reported an association between low BMD and frailty. The discrepancy with our results is likely attributable to the shorter duration of follow-up in our study, as clinically significant changes in bone density typically occur over longer timeframes and would therefore take longer to impact frailty onset (41). Our findings also support the hypothesis that muscle function precedes and influences bone adaptation through mechanical loading and muscle–bone crosstalk, potentially initiating changes that contribute to frailty development. Another plausible explanation for the superior predictive value of muscle measures lies in the inherently dynamic and adaptable nature of muscle tissue, which better reflects early functional decline. In contrast, DXA-derived bone and muscle mass metrics represent more static structural measures that evolve gradually. Therefore, physical performance tests for muscle were better associated with frailty risk.

The analysis of bone–muscle interaction in frailty risk revealed a noteworthy finding. This suggests a synergistic relationship between bone and muscle, wherein the combined decline in both parameters increases the likelihood of frailty. Interestingly, the interaction analysis demonstrated that higher handgrip strength attenuates the risk of frailty even in individuals with low BMD, and conversely, higher BMD mitigates frailty risk in those with reduced handgrip strength. These findings may emphasize the interdependence of bone and muscle in the pathophysiology of frailty and highlight the potential of targeting either organ to reduce frailty risk. As illustrated in Figs. 1a and 1b, the co-occurrence of low BMD and low handgrip strength markedly elevates frailty risk, whereas improvement in either parameter is associated with a meaningful reduction in frailty probability.

A limitation of this interaction analysis is that it did not account for other contributing factors that may impact the relationship. For instance, individuals with low BMD may have experienced fragility fractures, which are independently associated with increased frailty risk. Likewise, participants with low grip strength may have been more vulnerable to falls, potentially leading to functional decline and frailty. These unmeasured clinical events may have influenced the observed associations and should be considered in future analyses. Despite these limitations, the key implication of our findings is that improving either bone or muscle strength may reduce the risk of frailty. This underscores the importance of integrated interventions that target both bone and muscle to preserve physical function and delay the onset of frailty. Our group has previously reported a synergistic effect of hip BMD and gait speed on fracture risk, reinforcing the notion that the interplay between bone and muscle health has critical implications for adverse functional outcomes (42).

Strengths and Limitations

One of the key strengths of our research is its longitudinal cohort design, which allows for the assessment of changes over time and supports causal inference. In addition to assessing the impact of osteosarcopenia on frailty, we conducted a detailed analysis of the individual contributions of bone and muscle parameters and their interaction, offering a comprehensive evaluation of bone–muscle interplay in predicting frailty risk. Validated tools and well-recognized criteria were used; Sarcopenia was defined using validated criteria from EWGSOP2 and SDOC, while bone and muscle mass were measured via DEXA, and frailty was measured using the Fried phenotype. Another strength of our study was the comprehensive data collection (demographic, socioeconomic, lifestyle, and clinical variables), allowing for robust adjustments for covariates, thus enhancing the reliability of our findings.

However, our study has several limitations. First, although the sample size was adequately powered based on the expected transition to frailty, a higher-than-anticipated loss to follow-up (50% vs. 30%) may have reduced the number of transitions to frailty, thereby limiting the model’s ability to adjust for potential confounders. Second, although the follow-up period was adequate to assess a single transition to frailty, the complex and dynamic nature of frailty would be better captured through extended follow-up with repeated assessments. Third, the study population consisted of community-dwelling adults aged 50 and older, which may limit generalizability to broader older populations. Fourth, while the cohort included individuals from diverse backgrounds, with 70% born in Australia and 30% overseas, these demographic characteristics should be considered when interpreting and generalizing the findings. Fifth, as with many cohort studies involving older adults, there may be unmeasured bias due to the likely inclusion of a relatively health-conscious and motivated subgroup, potentially limiting external validity. Lastly, the use of the Fried frailty phenotype introduces overlap with sarcopenia through the inclusion of gait speed and handgrip strength, which may confound associations. Although alternative models, such as the Rockwood frailty index, could offer a broader perspective, the physical domain remains central to frailty assessment, particularly when investigating musculoskeletal contributions. We used validated definitions to align with research in this field.

This longitudinal study demonstrated that osteosarcopenia is associated with frailty risk. Clinical measures of muscle function—specifically handgrip strength and gait speed—were more strongly associated with frailty risk than DXA-derived bone measures. Notably, a synergistic interaction between bone density and muscle strength was observed, suggesting that improvements in either may mitigate frailty risk. Future research should incorporate larger study sizes, more extended follow-up periods, repeated assessments, and broader frailty definitions (particularly Rockwood definition) to explore this relationship further.

The authors have no competing interests to declare relevant to this article's content.

Funding

Mizhgan Fatima acknowledges support from the Australian Government Research Training Program (RTP) Scholarship for the conduct of this research work.

Author Contributions

Mizhgan Fatima – Conceptualization, Methodology, Ethics approval, Project administration, Data curation, Formal analysis, Resources, Visualization, Writing – original draft, Writing – review & editing.

Ben Kirk – Conceptualization, Resources, Supervision, Writing – review & editing.

Sara Vogrin - Formal analysis, Writing – review & editing.

Jason Talevski- Data curation, Writing – review & editing.

Sharon Brennan-Olsen- Conceptualization, Methodology, Resources, Supervision, Writing – review & editing.

Gustavo Duque – Conceptualization, Resources, Supervision, Writing – review & editing.

Ethics Declarations

Ethics Approval: All procedures performed involving human participants were in accordance with the ethical standards of the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Written informed consent was obtained from all study participants. The study protocol was approved by the Melbourne Human Research Ethics Committee (HREC/17/MH/381).

Kim DH, Rockwood K (2024) Frailty in older adults. N Engl J Med 391(6):538–548
Fried LP, Cohen AA, Xue Q-L, Walston J, Bandeen-Roche K, Varadhan R (2021) The physical frailty syndrome as a transition from homeostatic symphony to cacophony. Nat aging 1(1):36–46
Hoogendijk EO, Afilalo J, Ensrud KE, Kowal P, Onder G, Fried LP (2019) Frailty: implications for clinical practice and public health. Lancet 394(10206):1365–1375
Duque G (2021) Osteosarcopenia: a geriatric giant of the XXI century. Springer, pp 716–719
Kirk B, Zanker J, Duque G (2020) Osteosarcopenia: epidemiology, diagnosis, and treatment—facts and numbers. Wiley Online Library, pp 609–618
Kirk B, Al Saedi A, Duque G (2019) Osteosarcopenia: A case of geroscience. Aging Med 2(3):147–156
Laskou F, Fuggle NR, Patel HP, Jameson K, Cooper C, Dennison E (2022) Associations of osteoporosis and sarcopenia with frailty and multimorbidity among participants of the Hertfordshire Cohort Study. J cachexia sarcopenia muscle 13(1):220–229
Saeki C, Kanai T, Nakano M, Oikawa T, Torisu Y, Abo M et al (2020) Relationship between osteosarcopenia and frailty in patients with chronic liver disease. J Clin Med 9(8):2381
Davenport A, Frailty (2022) appendicular lean mass, osteoporosis and osteosarcopenia in peritoneal dialysis patients. J Nephrol 35(9):2333–2340
Tembo MC, Mohebbi M, Holloway-Kew KL, Gaston J, Sui SX, Brennan-Olsen SL et al (2021) The contribution of musculoskeletal factors to physical frailty: a cross-sectional study. BMC Musculoskelet Disord 22(1):921
Yoshimura N, Muraki S, Iidaka T, Oka H, Horii C, Kawaguchi H et al (2019) Prevalence and co-existence of locomotive syndrome, sarcopenia, and frailty: the third survey of Research on Osteoarthritis/Osteoporosis Against Disability (ROAD) study. J Bone Min Metab 37(6):1058–1066
Heng MWY, Chan AW, Man RE, Fenwick EK, Chew ST, Tay L et al (2023) Individual and combined associations of sarcopenia, osteoporosis and obesity with frailty in a multi-ethnic asian older adult population. BMC Geriatr 23(1):802
Wang Y-J, Wang Y, Zhan J-K, Tang Z-Y, He J-Y, Tan P et al (2015) Sarco-osteoporosis: prevalence and association with frailty in Chinese community-dwelling older adults. Int J Endocrinol 2015:482940
Szlejf C, Parra-Rodríguez L, Rosas-Carrasco O (2017) Osteosarcopenic Obesity: Prevalence and Relation With Frailty and Physical Performance in Middle-Aged and Older Women. J Am Med Dir Assoc 18(8):733. .e1-.e5
Hagströmer M, Oja P, Sjöström M (2006) The International Physical Activity Questionnaire (IPAQ): a study of concurrent and construct validity. Public Health Nutr 9(6):755–762
Kanis JA, Melton LJ 3rd, Christiansen C, Johnston CC, Khaltaev N (1994) The diagnosis of osteoporosis. J bone mineral research: official J Am Soc Bone Mineral Res 9(8):1137–1141
Zanker J, Scott D, Reijnierse E, Brennan-Olsen S, Daly R, Girgis C et al (2018) Establishing an Operational Definition of Sarcopenia in Australia and New Zealand: Delphi Method Based Consensus Statement. The journal of nutrition, health & aging. :1–6
Cruz-Jentoft AJ, Bahat G, Bauer J, Boirie Y, Bruyère O, Cederholm T et al (2019) Sarcopenia: Revised European consensus on definition and diagnosis. Age Ageing 48(1):16–31
Bhasin S, Travison TG, Manini TM, Patel S, Pencina KM, Fielding RA et al (2020) Sarcopenia definition: the position statements of the sarcopenia definition and outcomes consortium. J Am Geriatr Soc 68(7):1410–1418
Bandeen-Roche K, Xue Q-L, Ferrucci L, Walston J, Guralnik JM, Chaves P et al (2006) Phenotype of frailty: characterization in the women's health and aging studies. Journals Gerontol Ser A: Biol Sci Med Sci 61(3):262–266
Woods NF, LaCroix AZ, Gray SL, Aragaki A, Cochrane BB, Brunner RL et al (2005) Frailty: emergence and consequences in women aged 65 and older in the Women's Health Initiative Observational Study. J Am Geriatr Soc 53(8):1321–1330
Espinoza SE, Woods RL, Ekram AS, Ernst ME, Polekhina G, Wolfe R et al (2022) The effect of low-dose aspirin on frailty phenotype and frailty index in community-dwelling older adults in the ASPirin in Reducing Events in the Elderly study. Journals Gerontology: Ser A 77(10):2007–2014
Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J et al (2001) Frailty in older adults: evidence for a phenotype. The journals of gerontology Series A, Biological sciences and medical sciences. 56(3):M146–M156
Brennan-Olsen SL, Vogrin S, Balogun S, Wu F, Scott D, Jones G et al (2020) Education, occupation and operational measures of sarcopenia: Six years of Australian data. Australas J Ageing 39(4):e498–e505
ServicesAustralia [Available from: https://www.servicesaustralia.gov.au/how-much-carer-payment-you-can-get?context=21816
Vittinghoff E, McCulloch CE (2007) Relaxing the rule of ten events per variable in logistic and Cox regression. Am J Epidemiol 165(6):710–718
Goodman S (ed) (2008) editor A dirty dozen: twelve p-value misconceptions. Seminars in hematology. Elsevier
Fatima M, Kirk B, Vogrin S, Lee W-J, Peng L-N, Chen L-K et al (2025) Osteosarcopenia and frailty risk in community-dwelling older adults: A follow-up of the I-Lan Longitudinal Aging Study. Arch Gerontol Geriatr 136:105888
Yoshimura N, Muraki S, Oka H, Iidaka T, Kodama R, Horii C et al (2018) Do sarcopenia and/or osteoporosis increase the risk of frailty? A 4-year observation of the second and third ROAD study surveys. Osteoporosis international: a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA. 29(10):2181–2190
Lee A, McArthur C, Ioannidis G, Duque G, Adachi JD, Griffith LE et al (2024) Associations between Osteosarcopenia and Falls, Fractures, and Frailty in Older Adults: Results From the Canadian Longitudinal Study on Aging (CLSA). J Am Med Dir Assoc 25(1):167–76e6
Stolz E, Mayerl H, Waxenegger A, Rásky é, Freidl W (2016) Impact of socioeconomic position on frailty trajectories in 10 European countries: Evidence from the Survey of Health, Ageing and Retirement in Europe (2004–2013). J Epidemiol Commun Health 71(1):73–80
Brennan SL, Winzenberg TM, Pasco JA, Wluka AE, Dobbins AG, Jones G (2013) Social disadvantage, bone mineral density and vertebral wedge deformities in the Tasmanian Older Adult Cohort. Osteoporosis international: a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA. 24(6):1909–1916
Gurka MJ, Filipp SL, DeBoer MD (2018) Geographical variation in the prevalence of obesity, metabolic syndrome, and diabetes among US adults. Nutr diabetes 8(1):14
Greco EA, Pietschmann P, Migliaccio S (2019) Osteoporosis and sarcopenia increase frailty syndrome in the elderly. Front Endocrinol 10:255
Kirk B, Miller S, Zanker J, Duque G (2020) A clinical guide to the pathophysiology, diagnosis and treatment of osteosarcopenia. Maturitas 140:27–33
Jenkins ND, Welstead M, Stirland L, Hoogendijk EO, Armstrong JJ, Robitaille A et al (2023) Frailty trajectories and associated factors in the years prior to death: evidence from 14 countries in the Survey of Health, Aging and Retirement in Europe. BMC Geriatr 23(1):49
Hoogendijk EO, Dent E (2022) Trajectories, transitions, and trends in frailty among older adults: a review. Annals geriatric Med Res 26(4):289
Paulin TK, Malmgren L, McGuigan FE, Akesson KE (2024) Osteosarcopenia: prevalence and 10-year fracture and mortality risk–a longitudinal, population-based study of 75-year-old women. Calcif Tissue Int 114(4):315–325
Balogun S, Winzenberg T, Wills K, Scott D, Callisaya M, Cicuttini F et al (2019) Prospective associations of osteosarcopenia and osteodynapenia with incident fracture and mortality over 10 years in community-dwelling older adults. Arch Gerontol Geriatr 82:67–73
Tembo MC, Mohebbi M, Holloway-Kew KL, Gaston J, Brennan-Olsen SL, Williams LJ et al (2021) The Predictability of Frailty Associated with Musculoskeletal Deficits: A Longitudinal Study. Calcif Tissue Int 109(5):525–533
Bartosch P, McGuigan FE, Akesson KE (2018) Progression of frailty and prevalence of osteoporosis in a community cohort of older women-a 10-year longitudinal study. Osteoporosis international: a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA. 29(10):2191–2199
Kirk B, Zhang S, Vogrin S, Harijanto C, Sales M, Duque G (2023) Comparing the fracture profile of osteosarcopenic older adults with osteopenia/osteoporosis alone. Calcif Tissue Int 112(3):297–307

No competing interests reported.

SupplementaryFigureandTable.docx

Download PDF

Journal Publication

published 10 Dec, 2025

Read the published version in Aging Clinical and Experimental Research →

Editorial decision: Revision requested
06 Aug, 2025
Reviews received at journal
05 Aug, 2025
Reviewers agreed at journal
16 Jul, 2025
Reviews received at journal
12 Jul, 2025
Reviewers agreed at journal
11 Jul, 2025
Reviewers agreed at journal
02 Jun, 2025
Reviews received at journal
02 Jun, 2025
Reviewers agreed at journal
02 Jun, 2025
Reviewers invited by journal
02 Jun, 2025
Editor assigned by journal
29 May, 2025
Submission checks completed at journal
27 May, 2025
First submitted to journal
27 May, 2025

You are reading this latest preprint version

Osteosarcopenia, bone-muscle interactions, and Frailty risk: A Prospective Cohort Study of Community- Dwelling Older Adults

Status:

Journal Publication

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

Introduction

Materials and Methods

Study Design

Study Assessments

Osteosarcopenia

Frailty

Co-variates

Statistical Analysis

Results

Participants Characteristics

Association Between Osteosarcopenia and Frailty

Association Between Bone and Muscle Measures and Frailty

Bone-Muscle Interaction and Frailty

Discussion

Strengths and Limitations

Conclusion

Declarations

Funding

Author Contributions

Ethics Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1