Background

The aging of the global population is an international trend that must be addressed in the 21 st century. As the structure of the aged society continues to develop, issues and concerns regarding care for community-dwelling older adults are increasingly prominent. Particularly, amidst the spread and prevalence of the severe coronavirus disease 2019 (COVID-19) pandemic, restrictions on the activities and willingness of older adults have further exacerbated the frailty status of the older population in communities [1]. Progressive frailty may lead to setbacks in the ability of older adults to live independently and is closely associated with their overall health status [2]. Helping community-dwelling older adults living with frailty to address physical changes while preventing disability and promoting or restoring health has become a focal issue in current care.

Aging is one of the factors that affects physical activity (PA). Even for older adults in a frail state, engaging in strengthening physical activities can help delay the aging process. Particularly when moderate-intensity PA is substituted for prolonged periods of sitting, physical function can be improved, enabling individuals to maintain their daily activities [3]. Therefore, providing multiple physical training sessions is an important strategy to help community-dwelling older adults address their frailty and physical function. The literature has indicated that healthcare strategies for community-dwelling frail older adults should be diversified. Incorporating physical training as an intervention component is one of the most common methods and is also a consensus for preventing and delaying frailty. Additionally, many studies include nutrition as a common element of health promotion strategies [4,5,6,7]. Systematic reviews also point out that physical training intervention measures can significantly improve the physical function and status of community-dwelling older adults living with frailty [6,7,8]. Therefore, the health strategy for community-dwelling older adults living with frailty should include dynamic PA and nutritional dietary care guidance considering the individual’s health status so as to achieve substantial results in significantly improving the health status of older adults [9].

As times change, traditional health care methods are no longer sufficient to meet modern needs. The application of technology has become a major trend in reducing human resource costs. In recent empirical studies, some researchers have begun to use technology and remote technology to assist in care. This has become an emerging care strategy, especially during the COVID-19 epidemic, which has posed unprecedented challenges to community care for older adults. Therefore, developing telecare or online care service models and evaluating the implementation of health promotion programs for community-dwelling older adults provide another way to alleviate care inequalities and deficiencies [10,11,12]. Based on recent literature, it can be inferred that there is an obvious equivalence between technological healthcare models and traditional healthcare. Reducing labor costs in specific situations is an important consideration when choosing to use technology-assisted health care measures [1, 13,14,15,16]. Although the current evidence for the use of tele-interventions for older people in the community is still limited, this lack of support does not signify that these measures are ineffective. The remote technology care model can supplement traditional care, especially when considering costs or other uncontrollable factors, and can provide specific intervention strategies for community-dwelling older adults living with frailty.

Crafting practical and effective interventions for older adults with frailty is an exceedingly challenging endeavor. Frailty symptoms encompass a range of indicators, such as loss of muscle strength and mass, decreased mobility, diminished physical function, and fatigue. These manifestations can be mitigated through PA training and nutritional and dietary care guidance. Hence, this research aims to investigate the efficacy of the Fitness and Nutrition Program for Seniors (the FANS), developed specifically for community-dwelling older adults experiencing frailty. The program adopts a hybrid learning approach and adheres to the Transtheoretical Model (TTM), integrating in-person PA training, nutritional diet care guidance, group discussions, and remote group care [17]. Building upon the literature mentioned above, the hypothesis of this study was that, compared with the control group (CG), participants in the FANS program would demonstrate improvements in physical activity levels. These improvements were assessed with the Short Physical Performance Battery (SPPB), which evaluates standing balance, timed 4-meter walk speed, and the Five Times Sit to Stand test (FTSST). Additionally, grip strength, the Timed Up and Go (TUG) test, and evaluations of PA levels were included to assess physical activity. It was also anticipated that kinanthropometric measurements, including body mass index (BMI), mid-upper arm circumference (MUAC), and calf circumference (CC), would exhibit significant differences. Simultaneously, it monitored in-person course attendance, home practice implementation rates, and adverse events.

Methods

Study design

This research employed a non-randomized experimental control group design with pre-test and post-test assessments. The 6-month FANS intervention was implemented among community-dwelling older adults living with frailty, and its effectiveness was monitored and evaluated. It was impossible to blind the participants to the intervention because it involved the FANS. However, the evaluator was blinded, as the evaluator was a different person from the interventionist.

Participants

The sample size required for this study’s objectives was determined with the G Power 3.1.9.4 program. The effect size (f) was set to 0.25, the significance level (α) was set to 0.05, and the statistical power was set to 0.8. A baseline attrition rate of 14%, consistent with domestic interventional studies involving community-dwelling older adults living with frailty, was considered [18]. For sampling locations, this research selected community care centers located in various areas within the same city. Research briefing sessions were held at two community care centers. Subsequently, face-to-face explanations were provided to community-dwelling older adults who met the admission criteria, and they were provided assistance in signing consent forms and completing the sampling process. The inclusion criteria were as follows: (i) age of 65 years of age or older; (ii) diagnosis of pre-frailty or frailty criteria based on the frailty phenotype outlined in the Cardiovascular Health Study (CHS), with cutoff points and questions used to classify frailty in accordance with the FANS protocol [17]; (iii) ability to express awareness of and consent to cooperate with the implementation of the research intervention measures and data collection; (iv) for participants in the experimental group (EG), possession of and ability to independently operate communication software equipment. The exclusion criteria were as follows: (i) diagnosis of acute or chronic diseases unsuitable for PA, or history of joint or lower extremity surgery within the past 6 months; (ii) severe visual impairment; (iii) residency in long-term care facilities; (iv) previous participation in PA or nutrition-related interventional research within the past six months. During the recruitment period, a total of 83 community-dwelling older adults participated in the research screening. One individual was found ineligible for inclusion, and one declined to accept the research intervention plan. Consequently, a total of 81 community-dwelling older adults living with frailty met the inclusion and exclusion criteria. They were then allocated into the EG (comprising 43 individuals) and the CG (comprising 38 individuals) using convenience sampling (Fig. 1).

Fig. 1
figure 1

CONSORT intervention flowchart of FANS

Full size image

Ethical

This research received approval from the Fu Jen Catholic University Institutional Review Board (FJU-IRB) in 2022, with FJU-IRB NO: C110071. Furthermore, it is registered with ClinicalTrials.gov under the identifier Clinical Trials.gov ID: NCT05242549 (16/02/2022). In addition, all methods were conducted in accordance with relevant ethical standards. The study procedure was thoroughly explained to all participants. Those identified as eligible were provided with written informed consent forms and asked to sign them. Participant information was kept strictly confidential, and participants were informed that they could withdraw from the study at any time.

Intervention

In this research, the FANS adhered to the TTM as a comprehensive framework. First, in-person group courses were developed, drawing on “ACSM’s Guidelines for Exercise Testing and Prescription - Chinese Edition, 10th Edition” and the “Handbook of Nutrition in the Old Age Stage.” These courses encompassed PA training, nursing guidance for nutritional diets, and group discussions. Next, the “Line” mobile app was utilized to create an official account for offering personalized home practice sessions online. Subsequently, the “one-to-one” function of the Line official account and the “one-to-many” function of the Line group were employed to facilitate communication tailored to individual differences. Both the exercise load and nutritional intake were also adjusted through the Line official account and the Line group. These adjustments were made progressively over the six months of physical training. Lastly, the Line app facilitated remote consultation and coaching, supporting remote technology-assisted care. The aim of the FANS is to integrate home PA practices and dietary nutrition into daily routines, rather than relying on nutritional supplements to enhance effectiveness. Finally, the program content can be adjusted to suit individual needs, inspiring research subjects to develop a strong desire for health and thereby achieve a truly healthy older life [17].

The FANS program was developed through the collaborative guidance of experts in geriatric care, long-term care, physical therapy, sports coaching, nutrition, and nursing. The content validity index (CVI) value for the Handbook of FANS was 0.995. Additionally, to ensure the effectiveness of the FANS in remote technology-assisted care, the System Usability Scale (SUS) was used to assess the perceived usability of the official account and group by older adults. The average score on the scale was 70.50 ± 9.04, indicating that the official account and FANS group in Line achieved a level of usability that met the standard.

Fitness and Nutrition Program for Seniors (FANS)

The experimental group received the FANS in community care centers. The FANS program comprised a total of 9 in-person sessions, with each session lasting for 90 min. The content of each session included a 50-minute PA training session, a 20-minute nursing guidance session for nutritional diets, and a 20-minute group face-to-face discussion. These sessions were conducted alternately by fitness trainers and nurses. In the first month, four sessions of in-person group training were scheduled, followed by one session of in-person group training each subsequent month.

The in-person PA training sessions were conducted in a group format, beginning with joint mobility exercises that incorporated stretching and mobility of major joints, along with muscle strength and balance training. The primary focus was on weight and resistance training for the upper and lower limbs to enhance overall muscle strength. Balance training emphasized coordinated movements of core muscle groups and dynamic/static balance functions, requiring coordination and cognitive processing abilities. Prior to the conclusion of each session, simple cooling-down exercises were integrated to lower body temperature, stimulate blood flow, boost metabolism, and maintain muscle length and elasticity to facilitate body recovery. In-person nutritional dietary sessions were conducted in a group nursing guidance format. The main content covered the nutritional requirements of older individuals in terms of calories, protein, fat, and carbohydrates, along with dietary recommendations for various food groups. Key points of nutritional nursing guidance for the older population and the impact of PA and nutrition on frailty status were also addressed. At the same time, we also conducted face-to-face sessions to assess participants’ nutritional intake status. These sessions aimed to enhance older participants’ understanding of overall nutrition, further motivating them and promoting behavioral change.

In addition to in-person sessions, the program utilized the official account on the Line mobile app for remote technology-assisted home practices, with the aim of meeting the needs of community-dwelling older adults living with frailty and assessing their condition. Home practices were structured according to the individual’s current stages of change in the TTM and ten distinct processes of change. They were divided into three groups of home practice programs, each tailored to different TTM stages of change. The first group was designed for individuals in the precontemplation and contemplation stages, referred to as the “Senior FANS” group. The second group targeted those in the preparation stage, known as the “Leader FANS” group. The third group was intended for individuals in the action and maintenance stages, labeled as the “Veteran FANS” group. Adjustments to the groups were made during the practice period based on the individual stages of change each month. The three FANS groups (Senior FANS, Leader FANS, and Veteran FANS) were categorized based on their current stages of change and were guided through cognitive and behavioral strategies to support behavior transformation. Each group received tailored interventions aligned with their stages to help participants gradually achieve sustainable behavior change. Consultation and guidance were provided through the LINE official account and LINE group. Additionally, the official account was used to monitor the rates of home practice implementation and adverse events (Appendix 1) [17].

Control group

The CG participants continued receiving health promotion courses organized by their respective community care centers. They needed only to complete the pre-intervention assessment alongside the EG before implementation. After the intervention, follow-up measurement data were collected simultaneously with those of the EG at the third and sixth months. This data collection occurred three times in total over a period of 6 months.

Outcome measures

Demographic characteristics

The basic information included date of birth, biological gender, level of education, marital status, living situation, smoking history, alcohol consumption history, perceived economic status, number of chronic diseases, and frailty status, totaling 10 questions.

Physical activity

SPPB

Including the standing balance, timed 4-meter walk speed, and FTSST, this assessment evaluates muscle strength, muscular endurance, agility, sense of balance, and walking ability. Furthermore, scoring is based on the measurement methods and standards established by the National Health Administration of the Ministry of Health and Welfare, where a score of 10 or above indicates normal mobility, while a score of 9 or below indicates impaired mobility [19].

Grip strength

The average grip strength (measured in kilograms) was assessed using a handgrip dynamometer (CAMRY EH101 dynamometer, Guangdong Sensun Weighing Apparatus Group Ltd, Guangdong, China) to indicate muscle strength. The instrumentation had a measurement accuracy of 0.01 kg, and it was calibrated periodically throughout the study.

TUG

Participants were timed as they stood up from a chair, walked straight to a cone placed 2.44 m away, and returned to sit down. The time taken to complete this task was recorded in seconds.

PA level

The Taiwanese short version of the International Physical Activity Questionnaire (IPAQ) was used to calculate the PA levels of community-dwelling older adults living with frailty over the course of one week. This assessment allowed for the determination of the average metabolic equivalent (MET) of PA per hour for the participants. The test–retest reliability of this version is 0.67, and the content validity is as high as 0.9 [20, 21].

Kinanthropometric

Nutritional status changes were assessed in terms of BMI, MUAC, and CC [22].

Data collection

The research data were collected three times during in-person visits at two care centers: before the intervention, at the 3rd month after the implementation of the FANS, and at the 6th month for follow-up measurements. This study commenced in July 2022 and concluded in January 2023, lasting approximately six months. To ensure intervention fidelity, a special intervention coach, a lecturer, and a staff training workshop were organized to maintain consistency among all participating staff members. The measurement staff all participated in and completed the certification training course for the Senior Fitness Test to ensure the credibility and consistency of the study.

Statistical analysis

In the statistical analysis section, an intention-to-treat analysis (ITT analysis) was employed, wherein all participants were included for analysis. The statistical analysis in this study was conducted in SPSS for Windows 22.0. Descriptive statistics for categorical variables were calculated as frequencies and percentages, while continuous variables were analyzed using means and standard deviations. To compare the differences between the experimental and control groups in demographic characteristics, PA, and kinanthropometric measurements, the Chi-Squared Test was used for categorical variables, while the Independent Samples t-test was employed for continuous variables to examine the homogeneity between the two groups before the intervention. The research aimed to ascertain whether significant improvements occurred over time in PA and kinanthropometric measurements among the EG of community-dwelling older adults living with frailty after receiving the FANS, in comparison to the CG. Therefore, generalized estimating equations (GEE) were employed to analyze the outcome variables to infer differences between the groups [23, 24]. We examined the data collected at 3 and 6 months post-intervention to determine whether there was an interaction between the groups regarding PA and kinanthropometric-related data.

Results

Homogeneity test

The homogeneity testing results showed no statistically significant differences between the two groups in demographic characteristics, including date of birth, biological sex, level of education, marital status, living situation, smoking history, alcohol consumption history, and perceived economic status (p > 0.05). Additionally, the average numbers of chronic diseases were 0.93 in the experimental group and 0.97 in the control group. Regarding frailty status, 27.90% of the experimental group were in the frailty stage, while 72.10% were in the pre-frailty stage. In the control group, 18.40% were in the frailty stage, and 81.60% were in the pre-frailty stage. There were no statistically significant differences between the two groups in terms of chronic diseases or frailty status distribution (p > 0.05). However, significant statistical differences were observed in the analysis of PA homogeneity between the two groups, particularly in walking speed, the TUG, and PA level (p < 0.05). Furthermore, in the analysis of the homogeneity of the kinanthropometric measurements between the two groups, a significant statistical difference was found in CC (p < 0.05) (Table 1).

Table 1 Homogeneity tests were conducted for demographic characteristics, PA, and kinanthropometric measurements
Full size table

Analysis of PA

After participating in the FANS intervention for 6 months, the EG exhibited statistically significant improvements in SPPB, timed 4-meter walk speed, FTSST, TUG, and PA level, regardless of intra-group or inter-group comparisons with the CG. While the FANS intervention also resulted in improvements in grip strength in the EG, these enhancements did not reach statistical significance when compared within the group or with the CG (Fig. 2).

Fig. 2
figure 2

Comparison of PA between the experimental and control groups. Significant within group: * p < 0.05, ** p < 0.01, *** p < 0.001, Significant between groups: † p < 0.05, †† p < 0.01, ††† p < 0.001. T1: Data collection was conducted before the intervention. T2: The second data collection was carried out in the third month of the intervention. T3: The third data collection was conducted after the completion of the intervention in the sixth month. SPPB, Short Physical Performance Battery. FTSST, Five Times Sit to Stand test. TUG, Timed Up and Go test. PA, physical activity

Full size image

After adjustments for covariates including timed 4-meter walk speed, TUG, PA level, and CC, there were no statistically significant differences between the experimental and control groups in the SPPB, timed 4-meter walk speed, or grip strength at the mid-test or at the post-test compared to the baseline (T2 vs. T1; T3 vs. T1). In the FTSST, the EG exhibited a significant increase of 1.196 s at the mid-test (T2 vs. T1) compared to the CG (Wald χ2 = 4.401, p < 0.05). Similarly, in the TUG, the EG showed a significant reduction of 1.304 s at the post-test (T3 vs. T1) compared to the CG (Wald χ2 = 6.856, p < 0.01). Additionally, in terms of physical activity (PA) level, the EG demonstrated a significant increase of 3.589 MET/h at the mid-test (T2 vs. T1) (Wald χ2 = 4.080, p < 0.05) and a significant increase of 9.995 MET/h at the post-test (T3 vs. T1) compared to the CG (Wald χ2 = 34.888, p < 0.001) (Table 2).

Table 2 GEE analysis was conducted to assess the effectiveness of the intervention on PA
Full size table

Analysis of kinanthropometric measures

After participating in the FANS intervention for 6 months, the EG exhibited statistically significant improvements in MUAC, regardless of intra-group or inter-group comparisons with the CG. Additionally, the EG demonstrated statistically significant difference in CC compared to the CG. However, the improvements in BMI for the EG did not reach statistical significance when compared within the group or with the CG (Fig. 3).

Fig. 3
figure 3

Comparison of kinanthropometric measures between the experimental and control groups. Significant within group: * p < 0.05, ** p < 0.01, *** p < 0.001, Significant between groups: † p < 0.05, †† p < 0.01,††† p < 0.001. T1: Data collection was conducted before the intervention. T2: The second data collection was carried out in the third month of the intervention. T3: The third data collection was conducted after the completion of the intervention in the sixth month. BMI, body mass index. MUAC, mid-upper arm circumference. CC, calf circumference

Full size image

After adjusting for covariates including timed 4-meter walk speed, TUG, PA level, and CC, the BMI of the EG showed no statistically significant differences compared to the CG at the mid-term and post-term assessments (T2 vs. T1; T3 vs. T1). In MUAC, the EG exhibited a significant increase of 2.194 cm at the post-test (T3 vs. T1) compared to the CG (Wald χ2 = 19.695, p < 0.001). Similarly, in CC, the EG showed a significant increase of 2.500 cm at the post-test (T3 vs. T1) compared to the CG (Wald χ2 = 10.181, p < 0.01) (Table 3).

Table 3 GEE analysis was conducted to assess the effectiveness of the intervention on kinanthropometric measures
Full size table

In-person course attendance rate, home practice implementation rates, and adverse events

The EG had an average in-person course attendance rate of 67.18% in the FANS. The average implementation rate of home practice per month was 65.21%. During the intervention period, no adverse events were observed among any of the research participants. Occasional absences from in-person sessions were attributed to unrelated acute short-term illnesses, such as regular clinic visits, fear of contracting COVID-19, or ongoing COVID-19 infection.

Discussion

During the initial two months preceding the commencement of this research project, the severe impact of the COVID-19 pandemic led to the cessation of activities at community care centers for a duration of two months. This posed a considerable challenge for the older population, who relied on these centers for their daily activities, resulting in notable declines in their PA levels during this period. These declines could potentially accelerate the deterioration of their physical function [25]. Furthermore, the pandemic-induced restrictions significantly impacted the daily lives of community-dwelling older adults, constraining their range of activities and potentially causing reductions in muscle mass in their upper arms and calves. Within the context of aging, frailty, and the COVID-19 pandemic, the health status of these community-dwelling older adults living with frailty could significantly deteriorate [26]. The findings of this study highlight that participation in the FANS led to more substantial improvements in PA among frail older adults residing in the community compared to non-participants. Particularly noteworthy were the significant differences observed in assessments related to lower limb muscle strength, including the SPPB, timed 4-meter walk speed, FTSST, and TUG. Additionally, significant disparities were noted in overall PA levels.

Therefore, we conducted a further analysis to examine the disparities in intervention approach design compared to previous research. While most studies had higher doses and frequencies of in-person sessions, our research not only provided in-person sessions but also utilized remote group care to offer immediate guidance and assistance. This approach addressed the participants’ needs and supplemented the limitations of in-person sessions, thereby enhancing the sense of security, reassurance, and independence among older adults [27]. Past studies provided nutritional care through written instructions [28], dietary advice [29], or high-protein nutritional shakes [30]. However, this research assisted the participants in assessing the dietary status of each one during in-person sessions, incorporated remote nutritional counseling to enhance the self-efficacy of community-dwelling older adults living with frailty, and was centered on community-dwelling older adults. The use of remote care provided post-session nutritional information and immediate nutritional consultation, all of which contributed to reinforcing participants’ knowledge, achieving balanced diets, and promoting sustained behavior. Additionally, integrating the TTM into the intervention program aided the older adults engaged in home practices in overcoming barriers, in managing emotional conflicts, and by providing support. This, in turn, encouraged them to maintain adherence to healthy behaviors [17].

Indeed, the literature highlights that providing videos can aid in learning new behaviors and facilitate changes in health behavior [31]. Specifically for older adults, videos can enhance cognitive abilities [32]. This research provided video courses and textual instructions remotely, allowing older adults in the community to better understand the content of training and engage in PA and nutrition practice at home. Literature suggests that offering various consultation methods and different timings can more effectively motivate older adults [33]. Therefore, this research employed various consultation methods, including in-person group discussions, remote one-on-one consultations, and online group discussions, to motivate the older adults to enhance their adherence to home practice [33]. These methods can provide immediate feedback, aid in building trust, and enhance satisfaction [34].

Although the FANS did not significantly affect upper limb muscle strength, as measured by grip strength, this finding aligns with a systematic review and meta-analysis that reported no significant improvement in grip strength [7]. Therefore, the lack of significant improvements in grip strength in this study was not entirely unexpected. The lack of significant improvements in handgrip strength may have been due to the focus of the FANS program on functional activities rather than targeted grip exercises. Insufficient emphasis on forearm muscles or the intensity and duration of strength training could also have been contributing factors. While lower limb muscle strength is considered crucial for the functional abilities of older adults in daily life, grip strength does not directly impact walking activities. However, it is worth noting that many geriatric assessments rely on grip strength to evaluate the functional capacity of older adults [35]. Therefore, this result underscores the potential importance of community-dwelling older adults living with frailty to engage in resistance training for the upper limbs [36] or increase their protein intake [37] to prevent grip strength loss and maintain their ability to perform daily activities. These recommendations, cited from the literature, can serve as a useful reference for future interventions aimed at more effectively improving upper limb muscle strength.

In this research, the design of the FANS program included in-person sessions for PA training, nursing guidance sessions for nutritional diets, and group face-to-face discussions, alongside a blended course intervention approach that integrated remote technology-assisted group care. The research findings suggest that, despite the challenges posed by the pandemic, these interventions proved effective in maintaining the BMI, MUAC, and CC of community-dwelling frail older adults. Changes in body composition are closely linked to aging. These changes include decreased muscle mass, diminished subcutaneous fat mass, and increased total fat mass. Typically, visceral fat, liver fat, and muscle fat infiltration escalate with age, making alterations in body weight reflective of older adults’ health status [38]. The literature advises maintaining a BMI within the range of 18.5 to 29.9 kg/m2 to mitigate frailty risk [39]. Consequently, integrating BMI assessment and monitoring is essential in evaluating the health status of older adults. Moreover, advancing age not only heightens the likelihood of reduced muscle mass but may also impact the functioning of the oral cavity and digestive systems, elevating malnutrition prevalence [40]. Nutritional status closely correlates with muscle mass, strength, and function in older adults [41]. Concurrently, pandemic-induced declines in PA and the adoption of unhealthy dietary patterns among older adults can exacerbate the overall health threat [42]. In such contexts, measurements like upper arm and CC emerge as crucial indicators for evaluating malnutrition in older adults [22]. Especially when they are in frail states, older adults may experience concurrent sarcopenia [43].

Prior to the intervention, the MUAC and CC of both groups were close to or below the established sarcopenia thresholds (below 28.6 cm for men and below 27.5 cm for women for MUAC [44]; below 34 cm for men and below 33 cm for women for CC [45]), indicating a susceptibility to frailty and sarcopenia among the participants. In consideration of this susceptibility, behavior change techniques based on the TTM were employed to encourage older adults at different stages of change to recognize their self-efficacy and boost their confidence in participating in the FANS [17]. Despite the EG showing consistent or minimal decreases in limb circumference at each measurement after the intervention, the CG demonstrated significant decreases. This difference underscores a notable discrepancy in limb circumferences between the two groups post-intervention. Despite the lack of sufficient evidence regarding the duration needed for physical training or nutritional interventions to augment limb circumferences in older adults, this study indirectly suggests their pivotal role, particularly during the pandemic, in preserving muscle regeneration capacity. Interventions focusing on PA and nutrition may effectively curb or mitigate disease progression during early sarcopenic stages and potentially attenuate frailty risks during the pandemic.

Current research suggests that the personalities and experiences of coaches and lecturers are pivotal factors influencing the attendance and participation of older adults in community programs [45]. Furthermore, fostering close connections, positive interactions, and motivation, as well as sharing a sense of achievement with older adults, can enhance their participation [46]. Therefore, this study enlisted experienced and fully qualified sports coaches and nursing guidance lecturers to deliver customized and easily accessible courses. Comprehensive health assessments conducted prior to the courses ensured that these courses met the specific needs of older adults. Additionally, the inclusion of practical aids such as the stretch strap and dowel rod enhanced the appeal of the sessions. Held in familiar community settings, all sessions facilitated group interaction, effectively boosting motivation to participate. Consequently, these methods and strategies led to higher attendance rates in this study than in similar studies. Many studies have failed to comprehensively track older adults’ adherence to PA and nutrition at home. It is known that under supervised conditions, older adults typically exhibit higher adherence rates [47]. This research adopted a novel approach and strategy compared to previous research, utilizing the TTM as the foundation for behavior change in frail older adults. It implemented a hybrid program that integrated both in-person and remote support care to encourage sustained participation. Moreover, remote technology was employed to aid the participants, enabling them to report their home-based PA and nutritional statuses through official accounts and groups at their convenience. This approach enhanced comprehension of and adherence to home practices.

Finally, the results of this research confirm that the FANS is safe and well-tolerated, with no adverse events observed. This reflects the relative safety of the intervention and aids researchers and participants in assessing the acceptability of the intervention and balancing its effectiveness with implementation risks. In our study, we chose a quasi-experimental design over a randomized controlled trial (RCT) to balance internal and external validity. While RCTs can prevent human bias, impede the deliberate manipulation of results, and eliminate selection bias by randomization, they often do not reflect the variability of real-world interventions. Our goal was to understand the intervention’s effectiveness in real-life settings, where variability in delivery is common. Practical constraints, including participant recruitment challenges and the impact of the COVID-19 pandemic, made an RCT unfeasible. By adopting a quasi-experimental design with a control group, we aimed to enhance the internal validity and more confidently attribute observed outcomes to the intervention. Additionally, we prioritized feasibility by requiring participants to independently operate communication devices for remote data collection. However, we acknowledge that this may have introduced selection bias, potentially excluding individuals without access to or proficiency in digital technology.

Future research should focus on eliminating potential confounding factors by adopting randomized allocation and blinding methods to minimize bias and ensure the integrity of the study results. Additionally, incorporating qualitative research planning to gain detailed insights from the perspective of the community-dwelling older adult population living with frailty is crucial. This approach will facilitate a comprehensive understanding and consideration of all information related to the intervention. Moreover, government support for such health promotion initiatives could gradually increase awareness among older adults about frailty prevention, further encouraging their ongoing participation in these activities.

Conclusion

The findings of this research offer robust evidence supporting the beneficial impact of the FANS program on frail older adults residing in the community during the COVID-19 pandemic. In the short term, the intervention effectively enhanced participants’ lower limb muscle strength and significantly improved their performance across various metrics, including the SPPB, timed 4-meter walk speed, FTSST, TUG, and PA levels. Additionally, it positively influenced anthropometric measurements, ensuring the maintenance of BMI, MUAC, and CC. Notably, no adverse events were observed, despite consistent attendance rates in the courses. However, it is worth noting that the FANS intervention did not lead to a significant improvement in grip strength, potentially due to inadequate upper limb resistance training and protein intake. In terms of long-term implications, the overall findings of this research provide substantial evidence supporting the effectiveness of the FANS program. They emphasize the importance of promoting suitable exercise and nutrition initiatives, especially during a pandemic, to meet the health needs of frail older adults in the community. Furthermore, these findings offer valuable insights for gerontological care and lay the groundwork for future research in this field.