As a clinically relevant syndrome, frailty is defined by Fried et al. as a series of syndromes that result in an increased vulnerability to stressors, related to the deterioration in multiple physiological systems, typically presenting with slow step velocity, decreased activity, weight loss, fatigue, and muscle weakness (Fried et al., 2001). This could increase the risk of adverse events in old adults, including functional decline, falls, and high hospitalization rates (Brummel et al., 2017). Meanwhile, frailty was independently associated with disability and death (Fried et al., 2001; Wewerka et al., 2015; Sáez de Asteasu et al., 2019). And a growing number of studies have revealed that gait speed is one of the strongest indexes to predict adverse outcomes and the most useful indicator for the identification of frailty (Montero-Odasso et al., 2005; Rothman et al., 2008; Binotto et al., 2018).

Recent studies revealed that there were a large number of sarcopenic patients caused by age-related decline in muscle mass and strength in the frail older population (Roberts et al., 2021). Sarcopenia is defined as loss of skeletal muscle mass and strength (Cruz-Jentoft et al., 2010). Similar to frailty, sarcopenia is related to mobility disorders, falls and fractures, and severe reductions in their independence in activities of daily living (Liu et al., 2021). Frailty and sarcopenia share high symptomatic similarity and diagnostic overlap and thus cannot easily be distinguished (Fried et al., 2001; Cederholm, 2015; Mijnarends et al., 2015). As life spans lengthen, frailty and sarcopenia of aging are becoming more common, significantly affecting function and quality of life of elderly (Inoue et al., 2018). There has been a concomitant increase in frailty with sarcopenia, which can be more harmful to elderly’s health and quality of life. This condition requires early detection and careful treatment (Carmeli, 2017).

When sarcopenia occurs in the frail elderly population, the risks of falls, disability, and death will be substantially elevated in these individuals (Pfortmueller et al., 2014; Marques and Queirós, 2018; Thompson et al., 2021). A meta-analysis study revealed that there was a high prevalence of overlap between frailty and sarcopenia in hospitalized older adults, the combination of these conditions probably synergistically impairs outcome, leading to its adverse prognosis, including mobility disorders, falls, and fractures (Ligthart-Melis et al., 2020). A latest 10-year follow-up study demonstrated that a combination model of frailty and sarcopenia was more predictive of mortality than either condition alone. Patients with both frailty and sarcopenia were at nearly double risk of mortality than those with just one of these two diseases, and were over three times at risk of mortality than those neither frail nor sarcopenic. Individuals identified as frail would benefit from screening and assessment for sarcopenia (Thompson et al., 2021). Moreover, individuals with sarcopenia and frailty showed the strongest associations with cardiovascular disease and respiratory disease incidence, and mortality for all-cause and respiratory disease and cancer, and those with both sarcopenia and frailty had the highest mortality rate (Petermann-Rocha et al., 2021). Consequently, early recognition of sarcopenia in the frail elderly would be of particular significance, the aim is to start prevention as early as possible through exercise, nutrition, supplementation, and medical intervention. This approach could effectively decrease fall rates, disability rates, and mortality.

Stride speed has played an important role in the diagnosis of frailty and sarcopenia (Cruz-Jentoft et al., 2010). However, Gait encompasses a broad array of quantifiable parameters, such as cadence, gait speed, stride length, gait variability, and so on, and they reflect various aspects of gait (Verlinden et al., 2013). Studies focusing on the comprehensive gait assessment of the frail population and sarcopenic patients are still few. The gait characteristic of sarcopenic patients in the frail elderly has not been covered in literature so far. Due to the difficulty in the measurement of skeletal muscle, sarcopenia in the frail elderly is often not detected on time, sarcopenia comorbid with frailty will always result in a much worse prognosis, significant loss of quality of life, and can even lead to mortality (Petermann-Rocha et al., 2021). We can see that most of the older adults move slowly and show a significantly decline in daily physical activity. Recent work in this field suggested that there was a considerable overlap between frailty and sarcopenia in older adults (Ligthart-Melis et al., 2020; Faxén-Irving et al., 2021; AlMohaisen et al., 2022). However, diagnosis of sarcopenia has received little attention in these frail older adults and good and timely intervention therefore cannot be guaranteed. This special population needs much weighted attention, as they are worse and more hidden.

Few studies have focused on dual-task (DT) gait testing (walking while simultaneously performing a cognitive task) as a means to predict the onset of frailty or sarcopenia (Verghese et al., 2012; Thiede et al., 2016). Past research showed that gait changes would become more marked in the dual-task testing, the measurement of step speed under dual-task testing would enable researchers to more effectively distinguish between older adults with and without frailty (Guedes et al., 2014; Montero-Odasso et al., 2015). However, the researches of gait in dual-task conditions specific to frail older adults or sarcopenic patients are still inadequate. Combining Single- and dual-task gait analysis can reveal the feature of gait more comprehensively. Thus, we introduce the gait parameters within the dual-task trials for a deeper and more comprehensive understanding of gait characteristics in the frail elderly with sarcopenia.

No other study has taken a gait measurement technique to early analysis of sarcopenia in frail older adults to the best of our knowledge. Therefore, this study aims to examine the gait parameters of frail elderly population and its association with sarcopenia, using a gait assessment protocol including single- and dual-task gait testing to develop the model of gait characteristics for the early screening of frail older adults with sarcopenia.

To our knowledge, this is the first time that gait analysis has been used for the identification of sarcopenia in the frail population. Compared with the non-sarcopenic participants, frail older adults with sarcopenia showed significantly worse gait performance. The most important finding of this study was the role of gait performance under dual-task conditions in identification of sarcopenia among frail elders, not only single-task gait performance. Our data revealed a novel and surprising role of turn duration under dual-task walking, which was that the prediction performance of turn duration (DT) was good alone or in combination. The combined model of gait speed (DT)-turn duration (DT) yields relatively higher accuracy, while balancing both convenience and ease of access. In the stage of frailty with sarcopenia, reduction of gait speed and increase of turn duration under dual-task conditions could be the more salient model of gait for sarcopenia in frail elderly populations.

In this study, the sociodemographic and medical characteristics of subjects from the two groups showed that frail patients with sarcopenia were leaner and presented with a higher proportion of male, lower gait speed, lower muscle mass, a higher rate of osteoporosis, and a higher degree of frailty, and these findings were in accordance with previous studies (Reiss et al., 2019; Wu et al., 2021). This category of individuals could be the group with a higher risk of adverse events, especially falls. For this reason, early screening or identification is of great significance. However, most studies were specific to frailty or sarcopenia alone, studies on frail patients with sarcopenia remain very scarce. Thus, we performed this pioneering research among the frail elderly with sarcopenia, and we hope this study may promote the development of a new screening method and a new insight into the diagnosis in such individuals.

The dual-task gait testing can reveal the subtleties about the role of cognitive control in participants’ gait, and patients with cognitive impairment often suffer from gait disorders (Studenski et al., 2003; Montero-Odasso et al., 2009, 2012a). In this study, we found that there was no significant difference between the two groups in MoCA scores of patients with frailty and frailty combined with sarcopenia (18.157 ± 4.483 vs. 17.295 ± 6.021, p > 0.05), we can thus exclude the impact of differences in cognitive function on outcome. In this regard, we demonstrated that these two patient groups were comparable. However, the cognitive level of the two groups was generally low, which even reached the level of mild cognitive impairment (MCI; Allali and Verghese, 2017). The correlation between cognitive impairment and frailty has been previously demonstrated in many studies (Buchman et al., 2008; Mulero et al., 2011). Our results coincided with previous studies, indicating that both frail elderly with and without sarcopenia would develop some degree of decline in cognitive capacity. However, frail older adults with sarcopenia did not present more severe cognitive impairment.

Stride speed can be used as a simple and time-saving screening tool for frailty and sarcopenia. Because it is one of the major diagnostic criteria both for frailty and sarcopenia, which is more related to muscular dystrophy, functional independence, vitality, and weakness (Fried et al., 2001; Tanimoto et al., 2012; Middleton et al., 2015). Thus, our result is not surprising that gait speed is a good predictor of sarcopenia in frail elderly. Because of the convenience of pace measurement, we ignore other gait parameters and the specificity of patient’s gait changes in other gait modes. In our research, it was found that the features of dual-task gait in cognitive intervention were more obvious than those of single-task gait. At the same time, we found that gait speed exhibited higher predictive performance under the dual-task walking. To our surprise, turn duration in dual-task testing had a stronger predictive effect than gait speed on identifying frailty with sarcopenia, this result remained significant even after controlling for potential confounds. At the same time, the turning assessment has been confirmed to be a potential way to identify older adults at fall risk, older fallers took significantly longer turning time than non-fallers (Yeh et al., 2022). In addition, the latest research suggests that turn duration also plays an important role in distinguishing the Parkinson’s disease patients from normal, and predicting the future risk of fall in poststroke hemiplegic individuals (Vitorio et al., 2021; Zou et al., 2021), these categories of individuals at a higher risk of falls could be potential frail patients with sarcopenia. Our result also validated that turn duration in the dual-task condition was more relevant to muscle mass, muscle function, and severity of frailty (SMI, FREID scores, and grip strength), compared with other dual-task gait parameters.

Using low-cost and convenient gait analysis techniques to determine sarcopenia is a promising prediction method in frail elderly populations (Mitsiopoulos et al., 1998; Levine et al., 2000; Beaudart et al., 2016). In this study, compared with other gait parameters, gait speed (DT) and turn duration (DT) displayed stronger predictive power. Compared to other indicators, these two indicators are not dependent on sophisticated gait assessment devices and trained professionals. These two fundamental measures can be easily accessed by a smartphone device with inertial measurement unit. Moreover, validity and reliability of mobile phone-based gait assessments for the elderly were demonstrated in recent studies (Serra-Añó et al., 2020; Silsupadol et al., 2020; Rashid et al., 2021). The addition of turn duration (DT) can increase the power to identify sarcopenia in frail elderly over the use of gait speed (DT) alone (increasing AUC from 0.68 to 0.76). Therefore, we believe that the combination of gait speed and turn duration under dual-task conditions could be a convenient and effective digital biomarker of sarcopenia in the frail older adults and will be suitable for application in primary health institutions.

A major strength in this study is the first detailed assessment of gait for sarcopenic patients in frail elderly, using the currently popular wearable platform (APDM’s Mobility Lab). The main benefit of using wearable sensors is that we can easily analyze detailed gait metrics (Mancini et al., 2011). Gait parameters were assessed not only in their normal walking pattern, but also under dual-task conditions. Additionally, this study assessed the cognitive functions of patients to understand changes across cognitive in the frailty with sarcopenia. Although the number of patients included in each group was modest, we did substantial work to screen residents in care homes for frailty. The disease may be often overlooked and requires a series of detailed assessments including grip strength, walking speed, and cognitive function, among others. Thus, we believe that these data are of high quality.

The major limitation of this study is its cross-sectional design. Although it is obvious that associations exist between gait changes and sarcopenia in patients with frailty, we cannot ascertain the temporal order. Although global cognitive function of the two groups did not hold obvious differences in this study, the tight relationship between brain network function and changes in gait has been conclusively demonstrated (Montero-Odasso et al., 2012b, 2015; Lo et al., 2017). It will be important for future studies to investigate further to understand how neural networks take part in the regulation process of gait changes, which would help to provide a better understanding of the mechanism of worse gait performance in frail and sarcopenic patients respectively, for a wider application of the gait-derived digital metrics.

This study shows that turn duration under dual-task is a better predictive marker for sarcopenia in frail older people compared to other gait parameters, and the dual-task walking is more effective than single-task walking in identifying frailty with sarcopenia. In addition to the gait speed, adding the gait index of turn duration can improve the ability to identify sarcopenia. We believe that our research results can provide more convenient and effective ideas for the future technical design of biomarkers for screening sarcopenia in the frail elderly, encourage future research to use gait indicators to detect the incidence of frailty with sarcopenia, and identify individuals at high risk of falls. Future studies are also recommended to use of gait indicators to promote timely intervention and evaluation of treatment results.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by the First Affiliated Hospital of Soochow University. The patients/participants provided their written informed consent to participate in this study. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

YW and HT designed this study. ZZ, YH, JW, and HS collected the patients’ data. ZZ, YH, JW, HS, HT, and YW participated in writing the article and gave final agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work were appropriately investigated and resolved. All authors contributed to the article and approved the submitted version.

This work was partially supported by the Jiangsu Province Cadre Health Project (grant no. BJ19008). The leader project of clinical technology application research of Jiangsu elderly health research project (Project LR2021011).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.