Home-based strength and balance exercises for fall prevention among older individuals of advanced age: a randomized controlled single-blind study

Abstract

Objective

This research was to explore the effectiveness, safety, and adherence of home-based strength and balance exercises for fall prevention among the self-reliant individuals of advanced age and analyzed the beneficial components.

Method

This randomized controlled single-blind study included 124 individuals aged 80 years and over(mean age 84.4±3.2 years). The test group (n=63) performed strength and balance exercises facilitated by sports video training (≥ 3 sessions a week, ≥ 30 minutes per session), while the control group (n=61) maintained their daily routines. We conducted a comprehensive geriatric assessment (self-care ability, muscle strength, mobility, cognition, and psychological status) at baseline and 12 months later and dynamic posture mapping for balance and gait.

Results

The test group had a decreased risk of falls compared to the control group (25.4%vs.44.3%, respectively; RR = 0.747; 95% CI: 0.551–0.975; p = 0.027). There was no statistically significant difference in the fall rate between the two groups (0.48 falls per person-year vs. 0.67 falls per person-year, respectively; IRR: 0.708; 95% CI: 0.394–1.275; p = 0.251). The composite equilibrium score (SOTcom) for vestibular and integrated balance on the Sensory Organization Test (SOT) increased in the test group, while SOTcom decreased in the control group. In the test group, there was a significant improvement in the indexes pertaining to response time, movement speed, directional control, and endpoint offset in some directions. Adherence was better in the test group, with 54.0% exercised ≥ 3 times per week and 28.6% exercised 1–2 times per week on average.

Conclusion

Home-based strength and balance exercises improved balance and reduced the risk of falls among the individuals of advanced age. The video-guided, remotely monitored regimen demonstrated effectiveness, safety, and compliance, although scope for improvement remains.

1. Introduction

Falls, a common occurrence in geriatric care, are important causes of incapacitation, disabilities, loss of life, and increased medical expenditure. Falls can result in anxiety, depression, fear of falling, and social isolation, thus imposing a significant strain on care and an economic burden on families and society [1,2]. Estimates indicated that ∼50% of individuals aged 80 and over experienced at least one fall per year [3]. The United Nations projects that by 2050, 1/5th of the world’s population will be 60 years of age or older, of whom 19% will be over 80 years [4].

In China, as the population ages rapidly, the challenges of aging and solitude are likely to become increasingly prominent. In 2020, nearly 30 million were aged 80 years or older, and this is likely to reach over 110 million in 2050 [5]. The annual fall rate among older individuals in China ranged between 14.7 and 34%, and 60–75% of those who fell sustained injuries [6]. Over the past three decades, the incidence of falls among older individuals in China has increased by 79.2%, indicating a significantly rising trend, irrespective of gender, age, or region [7].

Sherrington et al. evaluated exercise interventions for preventing falls among older individuals in communities in an abridged Cochrane systematic review [8] and found that the inclusion of at least two types of exercises could decrease the risk and rate of falls and that strength and balance exercises had the highest quality assessment score. Strength and balance exercises are among the core interventions recommended by the National Institute for Health and Clinical Excellence (NICE) [9] and the American Geriatrics Society/British Geriatrics Society (AGS/BGS) [10].

However, it is also necessary to develop and further research exercises for specific subpopulations, such as older individuals at an advanced age, those with chronic diseases and dementia, and those requiring institutional care [11]. In its Recommendation Statement for interventions for falls prevention in community-dwelling older adults, the U.S. Preventive Services Task Force (USPSTF) [12] pointed out that among older individuals aged 85 and above, the effectiveness of interventions should be validated by further research. Furthermore, research should also focus on identifying the beneficial components of exercise interventions.

The effective reduction of falls among older individuals of advanced age who are self-reliant is of great significance to prolong their period of self-care and shorten their period of disability. In view of the challenges of traveling and low affordability of the study population, as well as the COVID pandemic, we designed an intervention regimen combining home-based strength and balance exercises and self-management with remote monitoring. In this research, we explored the usefulness of this intervention in decreasing the risk and rate of falls, analyzed the effective components, and evaluated its effectiveness, safety, compliance, and promotion.

2. Materials and methods

2.1. Study design

In this study, we adopted a single-center, randomized, controlled, and single-blind design with a follow-up time of 12 months. The study was done at the Geriatric Clinic of Beijing Tongren Hospital, China We used a random number table to assign participants to the test and control groups. Participants in the test group engaged in planned and supervised home-based strength and balance exercises, while those in the control group continued with their routine daily physical activities. The nurse responsible for implementing the comprehensive geriatric assessment and the technician performing the computerized dynamic posturography were blinded to the study design. The study has adhered to CONSORT guidelines.

2.1.1. Primary outcomes [8]

Risk of falls: number of people experiencing one or more falls.

Fall rate: number of falls per person-year.

2.1.2. Secondary outcome

Changes in the results of the comprehensive geriatric assessment, balance, and gait function tests before and after intervention.

2.1.3. Sample size estimation

Based on a review of literature and previous studies done by our team [3,13,14], we assumed the rate of new falls after intervention to be 25% in the test group and 50% in the control group, and when α = 0.05, β = 0.2, and the sample proportion was 1:1 in the test group and control group, we included 55 participants in each of the groups (110 in total). In view of the possibility of dropouts, we increased the sample size by 10% and planned to include 121 participants.

We followed the principles outlined in the Declaration of Helsinki for this investigation. The study protocol was approved by the Ethics Committee of the Beijing Tongren Hospital (CMU: TRECKY2020-041). It was registered in the China Clinical Trial Registry on 29 December 2021: ChiCTR2100054932.

2.2. Study participants

2.2.1. Inclusion criteria

Individuals aged 80 and above who were completely or basically self-reliant and were willing and able to cooperate in completing the related assessment, exercise plan, and follow-up required for this study.

2.2.2. Exclusion criteria

Older individuals with severe organ dysfunction, malignant tumors, progressive neurological diseases (such as Parkinson’s disease and multiple sclerosis), severe cognitive impairment, had a fracture or arthroplasty in the past 6 months, and those who were participating in any other rehabilitation projects.

Two doctors were responsible for recruiting subjects, all of whom were patients seeking treatment for chronic diseases. The study sample consisted of 124 people aged 80 and above treated at the Geriatric Clinic of Beijing Tongren Hospital, China, between December 2021 and March 2022. The respondents were 84.4 ± 3.2 (80–93) years old on average, consisting of 93 males (75.0%) and 31 (25.0%) females, and all were residents of Beijing. They were randomly assigned to the test group (n = 63) and the control group (n = 61).

2.3. Fall risk assessment

Participants arrived at the outpatient clinic at their scheduled time. A nurse accompanied them through a series of examinations according to a designed protocol, which took a total of half a day, including time for rest and meals.

2.3.1. Basic data

We collected basic information and medical history details, including age, gender, underlying diseases, history of falls in the last 12 months before enrollment, and use of medications acting on the central nervous system (such as sedative-hypnotics, antipsychotic drugs, and antidepressants). We also measured height, weight, and lying and standing blood pressure.

2.3.2. Comprehensive geriatric assessment

A specially trained nurse practitioner completed the comprehensive assessment, which included the following components. This procedure took ∼1 h.

1. Assessment of activities of daily living using the Barthel Index[ADL (Barthel)]: We used the ADL (Barthel) for assessing activities of daily living, and a score of ≥75 indicated the ability to manage basic self-care independently [15].

2. Assessment of upper limb muscle strength using the grip strength measurement method: The grip strength of the participants was checked using a dynamometer. Each hand was tested twice, and the highest value was taken. A grip strength of <28 kg (for males) and <18 kg (for females) indicated strength impairment [16].

3. Assessment of lower limb muscle strength using the Five Times Sit to Stand Test (FTSST): Each participant was asked to cross his/her arms in front of the chest, sit on a chair of normal height (48 cm high), and sit and stand as fast as possible for five times. The nurse recorded the time taken from the beginning to sitting on the chair for the fifth time. A duration of >10 s or <5 sets of sitting and standing indicated a high risk for falls [17].

4. Assessment of muscle functions using the usual gait speed (UGS): Each participant was asked to walk forward for 6 m from the starting point at their normal pace (usual gait speed). The timing was started when the toe of the participant was moved forward from the starting point until reaching the end point (6 m away), and the walking speed (m/s) was calculated. A gait speed of <0.8 m/s indicated a decrease in walking ability [18].

5. Assessment of dynamic balance function Timed Up and Go (TUG) test: Each participant was asked to stand up from a chair of normal height with armrests (the seat of the chair was 48 cm high and the armrest was 68 cm high), walk at their normal walking speed for 3 m, then return and sit again. The time was recorded from the start to sitting down, and if TUG was >12 s, it indicated a high fall risk [19].

6. Assessment of cognitive functions using the Montreal Cognitive Assessment (MoCA): The total score in this tool is 30. Scores were interpreted as follows: Normal: ≥26; mild cognitive impairment: 18–25; moderate cognitive dysfunction: 11–17; severe cognitive dysfunction: ≤10 [20].

7. All participants filled out the Zung Self-Rating Anxiety Scale (SAS), the Zung Self-Rating Depression Scale (SDS), the FRAIL (Fatigue, Resistance, Ambulation, Illnesses, and Loss of Weight) Scale, and the Falls Efficacy Scale International (FES-I) to assess their anxiety, depression, frailty, and fear of falling, respectively [21–23].

2.3.3. Computerized dynamic posturography (CDP)

We checked the balancing function using the EquiTest Dynamic Balancing Bench Tester and the Balancing Test Board (NeuroCom) and then processed the data with the supported computer information analysis and processing system to complete the calculation [24]. This procedure took ∼40 min.

Sensory Organization Test (SOT): This test is used to detect the individual’s ability to effectively use the three sensory systems, namely, the vestibular sense, visual sense, and proprioceptive sense, to maintain balance.

The participants completed six testing conditions of SOT and were tested three times for each condition (20 s per test). They were required to maintain their balance without moving their feet from the ground or shifting their position; otherwise, it was considered a fall. Condition 1: Stand upright with eyes open. Condition 2: Stand upright with eyes closed. Condition 3: Stand upright with eyes open, accompanied by sway-referenced vision. Condition 4: Stand upright with eyes open, accompanied by sway-referenced platform. Condition 5: Stand upright with eyes closed, accompanied by sway-referenced platform. Condition 6: Stand upright with eyes open, accompanied by sway-referenced vision and platform. After testing, the system calculated the composite equilibrium score (SOTcom) and the weights of somatosensory (SOM), vision (VIS), vestibular (VEST), and visual preference (PREF) based on the scores of each condition.

Limits of Stability Test (LOS): This test can effectively assess the ability of the participants to maintain balance during voluntary movement. Each participant was asked to stand barefoot on the support board, look forward, and stabilize the center of gravity in the center of the testing area. When the corresponding signal was shown on the screen, the participant was required to move the center of gravity towards the target area as soon as possible, keep the center of gravity stable, maintain the position for 10 s, and then move the center of gravity back to the center of the testing area again. While moving, the participant had to keep his/her feet firmly on the ground, body straight, and move to the farthest distance that he/she could reach. Eight tests were performed in the forward, forward/right, right, backward/right, backward, backward/left, left, and forward/left directions. The corresponding test data was calculated by the system, and we obtained the following values for the forward, backward, left, and right directions:

• Reaction time (RT): The time from the appearance of the signal to the start of movement by the participant (unit: second, s);

• Movement velocity (MVL): The angle at which the center of gravity moves per second (unit: °/s);

• Endpoint excursion (EPE): The distance that the center of gravity moves when the participant tries to reach the target highlighted on the screen (unit: %);

• Maximum excursion (MXE): The furthest distance that the center of gravity of the participant reaches during the test (unit: %);

• Directional control (DCL): The amount of movement in the intended direction (towards the target) subtracted from the amount of external movement (off-axis) (unit: %).

Tandem Walk (TW): for testing the ability of the participants to maintain balance while walking on a narrow supporting surface. The participant was asked to stand at the starting point of the balance board in a toe-to-heel manner, and when the signal was given by the instrument, he/she had to immediately walk forward along the central line of the board in a toe-to-heel manner with a distance of no more than 1 cm between the two feet; when the stop signal was flashed, the participant had to immediately stop walking and maintain balance while standing. The stride, gait speed, and endpoint sway were calculated.

2.4. Laboratory investigations

Participants arrived fasting at the geriatric clinic for blood collection. Ten ml of fasting blood was collected from the median vein of the elbow of each participant between 8 and 10 am, and the samples were sent to the laboratory department. Blood urea nitrogen (BUN), creatinine (Cr), albumin (ALB), estradiol (E2), total testosterone (T), and 25 hydroxyvitamin D (25(OH)D) were detected. The estimated glomerular filtration rate (eGFR) was calculated using the Modification of Diet in Renal Disease (MDRD) formula.

2.5. Bone density examination

This procedure took ∼10 min. We used the Lunar Prodigy Dual Energy X-Ray Absorptiometry (DXA) (manufactured by GE) to evaluate bone mineral density (BMD) and calculate the T score of lumbar vertebrae 1–4, bilateral femoral necks, and the total hip. The interpretation was as follows: Normal bone mass: T value ≥ −1; low bone mass: T value > −2.5 but < −1; osteoporosis: T value ≤ −2.5 [25].

2.6. Interventions

Physiotherapists designed and recorded four levels of instructional videos, consisting of visual function exercises, body position change exercises, reduced support surface exercises, standing/walking exercises, strength exercises, and flexibility exercises (Table 1). In the test group, physiotherapists and geriatricians recommended the appropriate level of exercises based on the baseline assessment of the participants and ACSM’s guidelines for exercise testing and prescription [26], and sent the instructional videos to the participants via WeChat. The participants were advised to exercise at home ≥ three times per week (for ≥30 min each session). The participants could communicate about the home-based exercises with the geriatricians and physiotherapists via WeChat, receive guidance, and modify their exercise levels as needed. Participants in the control group could also communicate with geriatricians and physiotherapists about home-based exercises via WeChat but continued their daily activities.

Table 1.

Strength and balance training protocol for the test group.

Phase	Training purpose	Training contents	Training conditions	Selection of training actions	Time
Phase 1	Activate body balance, restore physical flexibility and muscle strength	Visual function exercises, body position change exercises, standing/walking exercises, strength exercises and flexibility exercises	Sitting posture	Fix eyes and turn head from side to side; fix eyes and look up and nod; rotate eyeballs clockwise/counterclockwise and perform right- and left-handed ball throwing. March on the spot, raise both arms, move forward and backward with hands laced behind head, and rotate the trunk; hook toes while sitting, heel raising while sitting, practice sitting and standing, stand on half front and back feet, and stand on one foot; perform shoulder press with weight, perform bicep curl with weight, abduce with weight, raise heel in place, perform one-legged knee extension, and deep squat with chair support; walk around the room (protected); perform shoulder stretch while sitting, calf stretch while standing (protected), and quadriceps stretch while standing (protected).	20–30 s per action, 3 sets, with an interval of 30 s. 3 times/week, 30–40 min per session.
Phase 2	Improve the balance ability, and restore lower limb stability	Visual function exercises, body position change exercises, reduced support surface exercises, standing/walking exercises, strength exercises and flexibility exercises	Standing posture, partial actions while standing (holding a chair)	Fix eyes and turn head from side to side, fix eyes and look up and nod, rotate eyeballs clockwise/counterclockwise, turn head from side to side with eyes open, look up and nod with eyes open, perform lateral flexion of head with eyes open, perform right- and left-handed ball throwing; act like the fortune cat, move forward and backward with hands laced behind head, and rotate the trunk; shift weight from side to side with bow stance, raise heel in place, stand on half front and back feet, stand on front and back feet, stand on one foot, stand on one foot-roll on ball; practice sitting and standing, perform side step walks and tiptoe; perform shoulder press with a resistance band, perform bicep curl with a resistance band, abduce with a resistance band, perform hip abduction with a resistance band while sitting, perform leg lifts and turns while sitting, clamp a ball with thighs while sitting, knee extension/flexion/hip flexion/extension with weight on ankles, perform shoulder stretch, calf stretch while standing, and quadriceps femoris stretch while standing.	20–30 s or 12–15 times per action, 3 sets, with an interval of 30 s. 3 times/week, 30–40 min per session.
Phase 3	Improve body stability control ability and physical coordination	Visual function exercises, body position change exercises, reduced support surface exercises, standing/walking exercises, strength exercises and flexibility exercises	Standing posture, hold the chair while standing during actions with eyes closed	Fix eyes and turn head from side to side, fix eyes and look up and nod, rotate eyeballs clockwise/counterclockwise, turn head from side to side with eyes closed, look up and nod with eyes closed, perform lateral flexion of head with eyes closed, perform right- and left-handed ball throwing; act like the fortune cat, move forward and backward with hands laced behind head, and rotate the trunk; shift weight from side to side with bow stance, raise heel in place, stand on half front and back feet, stand on front and back feet, stand on one foot, stand on one foot-roll on ball; practice sitting and standing, perform side step walks; perform shoulder press with a resistance band, perform bicep curl with a resistance band, abduce with a resistance band, perform hip abduction with a resistance band while sitting, perform leg lifts and turns while sitting, clamp a ball with thighs while sitting, knee extension/flexion/hip flexion/extension with weight on ankles, perform shoulder stretch, calf stretch while standing, and quadriceps femoris stretch while standing.	30–40 s or 8–12 times per action, 3 sets, with an interval of 30 s. 3 times/week, 40–50 min per session.
Phase 4	Strengthen systemic stability control ability	Body position change exercises, reduced support surface exercises, unstable surface exercises, standing/walking exercises, strength exercises and flexibility exercises	Standing posture	Move head to write the character ‘mi’ with eyes closed, turn head from side to side with eyes closed, look up and nod with eyes closed, perform lateral flexion of head with eyes closed; perform right- and left-handed ball throwing, act like the fortune cat, move forward and backward with hands laced behind head, and rotate the trunk; stand with a single leg-swing leg back and forward, stand on half front and back feet, shift weight from side to side with bow stance on the balance pad, stand on one foot-roll on ball, practice sitting and standing on the balance pad, jump over obstacles forward and backward, jump over obstacles left and right; perform shoulder press with a resistance band, perform bicep curl with a resistance band, abduce with a resistance band, perform hip abduction with a resistance band while sitting, knee extension/flexion/hip flexion/extension with weight on ankles, perform shoulder stretch, calf stretch while standing, and quadriceps femoris stretch while standing	30–40 s or 8–12 times per action, 3 sets, with an interval of 30 s; 3 times/week, 60 min per session.

2.7. Evaluation of falls and adherence

A calendar was made. In the test group, participants recorded their exercises (time and duration of exercises and associated symptoms of discomfort) and falls (time, location, and injuries) in the calendars. In the control group, only the details of falls (time, place, and injury) needed to be recorded in the calendars. Each participant received one follow-up visit per month at the clinic, via WeChat, or by telephone, with an observation period of 12 months.

2.8. Statistical analysis

Statistical analysis was performed based on intention to treat (ITT) using SPSS 19.0 software. Counting data were expressed using percentages. The rates of new falls were analyzed using the chi-square test and compared using the negative binomial regression analysis. After the normality test, measurement data conforming to a normal distribution were expressed using mean ± standard deviation, and those of a non-normal distribution were shifted to a normal distribution by log and described as mean (95% confidence interval); the means were compared using the t-test based on two independent samples, and self-comparisons pre- and post-intervention were conducted using the paired t-test. Data not in a normal distribution were described as median (interquartile range), and a non-parametric test was conducted. A p-value of <0.05 indicated that the difference was statistically significant.

3. Results

In the test group (n = 63), all participants completed their follow-up visits regarding falls and exercises, while in the control group (n = 61), one participant was lost to follow-up and the remaining participants completed the follow-up (one died due to biliary duct cancer in the 12th month of follow-up). The functional assessment after 12 months was completed for 60 participants in the test group and for 56 participants in the control group (Figure 1).

Figure 1.

Flow chart of study participants.

3.1. Description of the sample and comparison of baseline data between the test and control groups

There were no statistically significant differences between the groups in terms of age, gender ratio, body mass index (BMI), major underlying diseases, use of medication affecting the central nervous system, history of falls in the past 12 months, estradiol, testosterone, 25(OH)D, or BMD (p > 0.05 for all variables). In the comprehensive geriatric assessment, participants in the test group had a significantly higher MoCA score than those in the control group (p = 0.028), and the grip strength was significantly lower than that in the control group (p = 0.006). There were no statistically significant differences between the two groups in terms of ADL, FRAIL scores, SAS scores, SDS scores, FES-I scores, gait speed, TUG, or FTSST values (p > 0.05 for all indicators), as shown in Tables 2 and 3.

Table 2.

Description of the sample and comparison between the test and control groups.

Variable	Whole sample (n = 124)	Test group (n = 63)	Control group (n = 61)	t, χ2, z-value	p-Value
Age (in years)	84.4 ± 3.2	84.1 ± 3.1	84.6 ± 3.3	−1.009*	0.315
Male [number (%)]	93 (75.0%)	46 (73.0%)	47 (77.0%)	0.269#	0.604
BMI (kg/m2)	23.3 ± 3.2	23.2 ± 3.2	23.4 ± 3.3	−0.237*	0.813
Coronary heart disease [number (%)]	49 (39.5%)	23 (36.5%)	26 (42.6%)	0.485#	0.486
Hypertension [number (%)]	85 (68.5%)	41 (65.1%)	44 (72.1%)	0.715#	0.398
Diabetes mellitus [number (%)]	38 (30.6%)	18 (28.6%)	20 (32.8%)	0.259#	0.611
Cerebrovascular disease [number (%)]	35 (28.2%)	16 (25.4%)	19 (31.1%)	0.506#	0.477
Stage 3 CKD [number (%)]	21 (16.9%)	9 (14.3%)	12 (19.7%)	0.944#	0.624
Osteoporosis [number (%)]	27 (21.8%)	15 (23.8%)	12 (19.7%)	0.260#	0.610
Postural hypotension [number (%)]	11 (8.9%)	5 (7.9%)	6 (9.8%)	0.138#	0.710
Use of medications that impact the central nervous system [number (%)]	27 (21.8%)	16 (25.4%)	11 (18.0%)	0.987#	0.321
History of falls [number (%)]	45 (36.3%)	21 (33.3%)	24 (39.3%)	0.484#	0.486
E2 (pg/ml)	28.0 ± 17.2	25.3 ± 14.3	30.7 ± 19.5	−1.727*	0.087
T (ng/ml)	3.3 (0.5, 4.4)	3.4 (0.5, 4.3)	3.0 (0.6, 4.4)	−0.539@	0.590
25OHD (ng/ml)	25.1 (23.3–26.9)	25.7 (23.5–28.3)	24.4 (21.8–27.1)	0.962*	0.338
eGFR (ml/min)	74.7 ± 16.3	76.9 ± 17.0	72.6 ± 15.5	1.450*	0.150
L1–4BMD (g/cm2)	1.21 ± 0.26	1.20 ± 0.26	1.22 ± 0.26	−0.307*	0.759
LtBMD (g/cm2)	0.90 ± 0.17	0.89 ± 0.16	0.91 ± 0.18	−0.417*	0.677
RtBMD (g/cm2)	0.89 ± 0.14	0.90 ± 0.16	0.88 ± 0.13	0.497*	0.620
ADL (score)	100 (100, 100)	100 (100, 100)	100 (100, 100)	−0.329@	0.742
FRAIL (score)	0 (0, 1)	0 (0, 1)	0 (0, 1)	−1.209@	0.227
MoCA (score)	25.3 ± 3.3	26.0 ± 3.2	24.6 ± 3.6	2.228*	0.028
SDS (score)	30.0 (26.3, 37.8)	30.0 (26.3, 40.0)	30.0 (27.5, 33.7)	−0.350@	0.726
SAS (score)	30.0 (27.5, 35.0)	30.0 (27.5, 35.0)	30.0 (28.0, 35.0)	−0.095@	0.924
FES-I (score)	35.0 ± 12.3	36.8 ± 12.8	33.2 ± 11.5	1.619*	0.108
Grip strength (kg)	23.7 ± 5.6	22.4 ± 5.0	25.1 ± 6.0	−2.777*	0.006
UGS (m/s)	0.83 ± 0.24	0.83 ± 0.21	0.82 ± 0.26	0.319*	0.750
TUG (s)	14.0 (13.0–15.0)	14.6 (13.0–16.2)	13.5 (12.2–14.8)	0.973*	0.332
FTSST (s)	14.6 ± 5.1	14.3 ± 5.0	14.9 ± 5.3	0.636*	0.526

BMI: body mass index; Stage 3 of CKD: stage 3 of chronic kidney disease; history of falls: had falls in 12 months prior to enrollment; E2: estradiol; T: testosterone; 25OHD: 25 hydroxyvitamin D; eGFR: estimated glomerular filtration rate; L1–4 BMD: mean bone density of lumbar vertebrae 1–4; LtBMD: bone mineral density of the left total hip; RtBMD: bone mineral density of the right total hip; ADL: Activity of Daily Living Scale (Barthel Index); FRAIL: the FRAIL Scale; MoCA: Montreal Cognitive Assessment; SDS: Zung Self-Rating Depression Scale; SAS: Zung Self-Rating Anxiety Scale; FES-I: Falls Efficacy Scale International; UGS: usual gait speed; TUG: Timed Up and Go Test; FTSST: Five Times Sit to Stand Test.

*Two-sample t-tests were used for statistical comparison.

^@Non-parametric tests were used for statistical comparison.

^#Chi-square tests were used for statistical comparison.

Table 3.

Baseline comparison of balance and gait function between the test and control groups.

Variable	Test group (n = 63)	Control group (n = 61)	T, z-value	p-Value
SOM (score)	99.0 (97.0, 101.0)	99.0 (96.0, 101.0)	−0.294@	0.769
VIS (score)	76.0 (69.0, 85.0)	76.0 (64.0, 83.0)	−0.415@	0.678
VEST (score)	62.0 (20.0, 72.0)	59.0 (34.0, 68.0)	−0.446@	0.656
PREF (score)	96.0 ± 14.1	98.9 ± 14.6	−1.131*	0.260
SOTcom (score)	68.0 (53.0, 77.0)	67.0 (53.0, 76.0)	−0.088@	0.930
RT-F (s)	1.02 (0.94–1.08)	0.99 (0.90–1.09)	0.816*	0.416
RT-B (s)	0.81 (0.73–0.90)	0.96 (0.81–1.13)	−2.291*	0.024
RT-R (s)	1.04 ± 0.36	0.98 ± 0.34	0.950*	0.344
RT-L (s)	1.01 ± 0.38	1.03 ± 0.38	−0.329*	0.743
RT-C (s)	0.95 (0.83, 1.10)	0.97 (0.78, 1.25)	−0.912@	0.362
MVL-F (°/s)	2.69 ± 0.89	2.47 ± 0.80	1.352*	0.179
MVL-B (°/s)	1.91 ± 0.62	1.77 ± 0.84	1.068*	0.288
MVL-R (°/s)	3.08 ± 0.96	2.99 ± 0.86	0.544*	0.588
MVL-L (°/s)	3.21 ± 1.30	2.82 ± 1.09	1.733*	0.086
MVL-C (°/s)	2.74 ± 0.65	2.50 ± 0.71	1.905*	0.059
DCL-F (%)	73.08 ± 10.45	73.07 ± 12.69	0.004*	0.997
DCL-B (%)	47.27 ± 19.13	43.33 ± 25.28	0.943*	0.348
DCL-R (%)	69.21 ± 8.69	66.29 ± 12.29	1.466*	0.146
DCL-L (%)	71.61 ± 9.16	67.15 ± 15.85	1.836*	0.070
DCL-C (%)	64.41 (61.23–67.35)	59.10 (52.41–64.69)	−0.134*	0.894
EPE-F (%)	46.85 ± 12.28	43.78 ± 13.39	1.295*	0.198
EPE-B (%)	35.74 ± 11.29	33.00 ± 16.33	1.043*	0.299
EPE-R (%)	59.40 ± 17.43	57.95 ± 15.25	0.479*	0.633
EPE-L (%)	55.98 ± 16.49	53.31 ± 17.37	0.854*	0.395
EPE-C (%)	49.66 ± 8.92	46.44 ± 12.78	1.564*	0.121
MEX-F (%)	65.92 ± 13.11	64.35 ± 18.51	0.525*	0.601
MEX-B (%)	49.44 ± 17.59	42.80 ± 21.43	1.838*	0.069
MEX-R (%)	75.87 ± 16.22	73.98 ± 17.42	0.607*	0.545
MEX-L (%)	71.63 ± 17.88	66.45 ± 18.24	1.548*	0.124
MEX-C (%)	65.74 ± 9.78	61.04 ± 15.50	1.936*	0.056
Step width (cm)	15.54 (13.92–17.48)	15.05 (13.21–17.12)	0.669*	0.505
Speed (cm/s)	17.49 ± 8.92	16.28 ± 7.27	0.766*	0.446
End sway (°/s)	8.12 (7.24–9.05)	8.18 (7.05–9.51)	0.326*	0.745

SOM: somatosensory; VIS: vision; VEST: vestibular; PREF: visual preference; SOTcom: Composite Equilibrium Score, the weighted average score of sensory integration test under six test conditions; RT: reaction time; MVL: movement velocity; DCL: directional control; EPE: endpoint excursion; MXE: maximum excursion; F: forward; B: backward; R: right; L: left; C: comprehensive.

*Two-sample t-tests were used for statistical comparison.

^@Non-parametric tests were used for statistical comparison.

CDP detection of balance and gait functions showed that in the LOS test, participants in the test group had significantly lower backward/right (RT-B) scores than those in the control group (p = 0.024). However, there were no statistically significant differences in the results of the other items of LOS, SOT, and TW (p > 0.05 for all variables).

3.2. Primary outcomes

As per the intention-to-treat principle, the participant lost to follow-up in the control group was considered ‘no fall’ in the statistical analysis.

3.2.1. Risk of new falls

In the test group (n = 63), there were 16 new falls, accounting for 25.4% (16/63 cases), while in the control group (n = 61), there were 27 new falls, accounting for 44.3% (27/61 cases). χ² = 4.869; p = 0.027; RR = 0.747; and 95% CI: 0.551–0.975, indicating that the difference between the groups in terms of new falls was statistically significant.

3.2.2. Rate of new falls

In the test group (n = 63), there were 30 falls in total (0.48 falls per person-year), while in the control group (n = 61), there were 41 falls in total (0.67 falls per person-year). IRR: 0.708; 95% CI: 0.394–1.275; and p = 0.251, indicating that there was no statistically significant difference between the groups in terms of the rate of new falls.

3.2.3. Injuries resulting from falls

There were a total of nine falls resulting in fractures in both groups (two cases in the test group and seven cases in the control group).

3.3. Secondary outcomes

3.3.1. Comparison of comprehensive geriatric assessments before and after the intervention

Post-intervention, there was a significant increase in the grip strength in the test group (p = 0.036) while it decreased in the control group (p = 0.000) when compared with the situation before intervention; pre and post-intervention, the gait speed slowed down in the test and control groups (p = 0.000, 0.017) and the SAS score increased (p = 0.004, 0.002); TUG duration increased in the control group (p = 0.005), and FES-I scores, FRAIL scores, and SDS scores increased (p = 0.000, 0.000, 0.000, respectively), while ADL and MoCA scores decreased (p = 0.001, 0.045, respectively). There were no statistically significant differences in FTSST before and after intervention in both groups, nor in TUG, scores of FES-I, ADL, FRAIL, MoCA, or SDS in the test group (p > 0.05 for all) (Table 4).

Table 4.

Comparison of comprehensive geriatric assessment in the two groups before and after intervention.

Item	Intervention phase	Test group (n = 63)			Control group (n = 61)			Inter-group Comparison
		Value	T, z-value	p-Value	Value	T, z-value	p-Value	T, z-value	p-Value
Grip strength (kg)	Before	22.3 ± 5.1	−2.150*	0.036	25.5 ± 5.9	4.537*	0.000	−3.082*	0.003
	After	23.5 ± 6.5			23.0 ± 5.1			0.530*	0.597
UGS (m/s)	Before	0.85 ± 0.20	4.396*	0.000	0.84 ± 0.26	2.464*	0.017	0.217*	0.829
	After	0.77 ± 0.21			0.78 ± 0.22			−0.170*	0.866
TUG (s)	Before	14.1 (12.6–15.6)	−1.853*	0.069	13.1 (11.9–14.5)	−2.938*	0.005	0.724*	0.470
	After	14.6 (12.3–16.1)			14.4 (12.9–15.9)			0.469*	0.640
FTSST (s)	Before	14.1 ± 4.9	0.671*	0.505	14.7 ± 5.3	−0.952*	0.345	−0.470*	0.639
	After	13.8 ± 5.3			15.3 ± 5.2			−1.541*	0.126
FES-I (score)	Before	37.2 ± 13.0	−0.993*	0.325	33.7 ± 11.6	−4.849*	0.000	1.549*	0.127
	After	39.3 ± 11.7			40.8 ± 10.6			−0.759*	0.450
ADL (score)	Before	100 (100, 100)	−0.587@	0.557	100 (100, 100)	−3.461@	0.001	−0.178@	0.859
	After	100 (100, 100)			100 (91.3, 100)			−2.666@	0.008
FRAIL (score)	Before	0 (0, 1)	−0.577@	0.564	0 (0, 1)	−3.491@	0.000	−0.976@	0.329
	After	0 (0, 1)			1 (0, 1)			−3.002@	0.003
MoCA (score)	Before	25.9 ± 2.9	0.065*	0.949	24.7 ± 3.3	2.049*	0.045	2.036*	0.044
	After	25.9 ± 2.7			24.2 ± 3.7			2.872*	0.005
SDS (score)	Before	30.0 (26.3, 40.0)	−1.850@	0.064	30.0 (26.3, 33.8)	−3.988@	0.000	−0.453@	0.651
	After	33.6 (30.0, 40.0)			35.0 (28.8, 41.5)			−0.465@	0.642
SAS (score)	Before	30.0 (27.5, 35.0)	−2.908@	0.004	30.0 (28.0, 35.0)	−3.105@	0.002	−0.239@	0.811
	After	33.8 (27.5, 38.4)			32.5 (28.8, 36.3)			−0.205@	0.837

UGS: usual gait speed; TUG: Timed Up and Go Test; FTSST: Five Times Sit to Stand Test; FES-I: Falls Efficacy Scale International; ADL: Activity of Daily Living Scale (Barthel Index); FRAIL: The FRAIL Scale; MoCA: Montreal Cognitive Assessment; SDS: Zung Self-Rating Depression Scale; SAS: Zung Self-Rating Anxiety Scale.

*Two-sample t-tests were used for statistical comparison (two-sample t-test and paired t-test).

^@Non-parametric tests were used for statistical comparison.

3.3.2. Comparison of balance and gait functions before and after intervention

Compared with the situation before intervention, SOTcom and VEST increased in the test group (p = 0.000, 0.003, respectively), while in the control group, SOTcom decreased (p = 0.022). There were no statistically significant differences in SOM, VIS, and PREF in the test group and SOM, VIS, VEST, and PREF in the control group before and after intervention (p > 0.05 for all).

In LOS, reaction time-comprehensive (RT-C) and reaction time-right (RT-R) reduced in the test group (p = 0.006, 0.006, respectively); movement velocity-comprehensive (MVL-C), movement velocity-backward (MVL-B), and movement velocity-right (MVL-R) accelerated (p = 0.011, 0.010, 0.038, respectively); directional control-comprehensive (DCL-C), directional control-forward (DCL-F), directional control-backward (DCL-B), and directional control-right (DCL-R) improved (p = 0.004, 0.010, 0.041, 0.019, respectively); and EFE-B increased (p = 0.018); while in the control group, DCL-F decreased (p = 0.013). There were no statistically significant differences in other items before and after intervention (p > 0.05, all).

There were no statistically significant differences in stride, gait speed, or endpoint sway values in the gait function test before and after intervention (p > 0.05 for all). The comparison of balance and gait function before and after intervention is shown in Table 5.

Table 5.

Changes in balance and gait function in the test group and control group before and after intervention.

Item	Intervention phase	Test group (n = 60)			Control group (n = 56)			Inter-group comparison
		Value	T, z-value	p-Value	Value	T, z-value	p-Value	T, z-value	p-Value
SOTcom (score)	Before	68.0 (50.8, 77.0)	−3.718@	0.000	66.5 (53.0, 76.0)	−2.289@	0.022	−0.105@	0.916
	After	78.0 (60.8, 82.0)			61.5 (45.5, 74.8)			−3.822@	0.000
SOM (score)	Before	99.0 (97.0, 101.0)	−0.640@	0.522	99.0 (97.0, 101.0)	−1.001@	0.317	−0.003@	0.998
	After	98.0 (96.0, 100.0)			98.0 (96.0, 100.0)			−0.618@	0.537
VIS (score)	Before	75.0 (69.0, 85.0)	−1.484@	0.138	76.0 (63.0, 83.8)	−0.080@	0.937	−0.420@	0.674
	After	79.5 (70.0, 85.8)			74.0 (60.5, 85.0)			−1.819@	0.069
VEST (score)	Before	60.5 (19.3, 70.8)	−2.942@	0.003	57.5 (34.0, 67.8)	−1.646@	0.100	−0.410@	0.682
	After	69.0 (49.8, 83.0)			50.0 (14.3, 67.8)			−3.359@	0.001
PREF (score)	Before	95.53 ± 14.20	−1.675*	0.099	98.66 ± 15.23	0.929*	0.357	−1.145*	0.255
	After	99.42 ± 14.13			96.19 ± 14.22			1.189*	0.237
RT-C (s)	Before	0.95 (0.83, 1.10)	−2.737@	0.006	0.97 (0.78, 1.25)	−1.231@	0.218	−0.813@	0.416
	After	0.84 (0.75, 1.00)			0.92 (0.80, 1.11)			−2.349@	0.019
RT-F (s)	Before	1.03 (0.95–1.10)	1.989*	0.051	0.99 (0.90–1.10)	−0.130*	0.897	0.950*	0.344
	After	0.94 (0.87–1.01)			1.01 (0.92–1.11)			−0.621*	0.536
RT-B (s)	Before	0.82 (0.72–0.93)	0.517*	0.607	0.98 (0.81–1.19)	0.541*	0.591	−2.138*	0.035
	After	0.78 (0.69–0.90)			1.04 (0.88–1.21)			−2.352*	0.020
RT-R (s)	Before	1.06 ± 0.36	2.872*	0.006	1.00 ± 0.33	0.533*	0.596	0.889*	0.376
	After	0.90 ± 0.26			0.97 ± 0.36			−1.275*	0.205
RT-L (s)	Before	1.02 ± 0.38	1.620*	0.111	1.04 ± 0.38	0.432*	0.668	−0.256*	0.798
	After	0.91 ± 0.40			1.00 ± 0.36			−1.288*	0.201
MVL-C (°/s)	Before	2.73 ± 0.66	−2.615*	0.011	2.49 ± 0.72	−0.081*	0.936	1.843*	0.068
	After	2.99 ± 0.64			2.52 ± 0.87			3.287*	0.001
MVL-F (°/s)	Before	2.64 ± 0.89	−1.437*	0.156	2.47 ± 0.81	−0.512*	0.611	1.040*	0.301
	After	2.83 ± 0.74			2.61 ± 0.87			1.438*	0.153
MVL-B (°/s)	Before	1.91 ± 0.59	−2.650*	0.010	1.77 ± 0.86	−0.840*	0.405	0.945*	0.347
	After	2.21 ± 0.77			1.89 ± 0.93			2.027*	0.045
MVL-R (°/s)	Before	3.07 ± 0.98	−2.125*	0.038	2.96 ± 0.85	0.327*	0.745	0.619*	0.537
	After	3.39 ± 1.06			2.94 ± 0.90			2.392*	0.018
MVL-L (°/s)	Before	3.26 ± 1.31	−0.475*	0.637	2.83 ± 1.10	−0.729*	0.469	1.831*	0.070
	After	3.38 ± 1.02			3.00 ± 1.12			1.870*	0.064
DCL-C (%)	Before	64.52 (61.10–67.50)	−3.037*	0.004	58.80 (51.16–65.24)	0.106*	0.916	−0.125*	0.901
	After	67.64 (63.26–71.36)			60.67 (54.18–66.31)			2.255*	0.026
DCL-F (%)	Before	76.00 (66.00, 81.25)	−2.538@	0.010	75.00 (68.00, 82.00)	−2.483@	0.013	−0.376@	0.707
	After	79.00 (74.00, 83.00)			73.00 (61.00, 80.75)			−3.147@	0.002
DCL-B (%)	Before	47.32 ± 19.54	−2.085*	0.041	43.15 ± 25.80	−0.853*	0.398	0.950*	0.345
	After	53.50 ± 20.04			45.92 ± 20.16			1.790*	0.076
DCL-R (%)	Before	69.56 ± 8.63	−2.411*	0.019	66.15 ± 12.52	−0.633*	0.530	1.646*	0.103
	After	72.35 ± 8.80			67.42 ± 14.82			2.098*	0.039
DCL-L (%)	Before	71.76 ± 9.22	−0.170*	0.866	67.02 ± 16.17	−0.648*	0.520	1.865*	0.066
	After	71.88 ± 14.58			68.62 ± 14.68			1.179*	0.241
EPE-C (%)	Before	49.75 ± 9.11	−1.728*	0.089	46.33 ± 13.09	0.342*	0.734	1.577*	0.118
	After	52.30 ± 9.47			45.62 ± 13.93			3.002*	0.003
EPE-F (%)	Before	46.61 ± 12.50	−0.554*	0.582	43.85 ± 13.77	0.453*	0.653	1.109*	0.270
	After	47.50 ± 10.73			43.33 ± 12.31			1.917*	0.058
EPE-B (%)	Before	35.73 ± 11.53	−2.440*	0.018	33.02 ± 16.80	−0.790*	0.433	0.978*	0.331
	After	41.03 ± 12.75			34.19 ± 15.42			2.570*	0.012
EPE-R (%)	Before	59.19 ± 17.79	−0.592*	0.556	57.33 ± 15.44	0.547*	0.587	0.584*	0.560
	After	61.13 ± 16.34			56.00 ± 15.74			1.687*	0.094
EPE-L (%)	Before	56.78 ± 15.99	−0.945*	0.348	53.54 ± 17.38	0.007*	0.995	1.023*	0.308
	After	59.18 ± 16.13			54.02 ± 15.26			1.732*	0.086
MEX-C (%)	Before	66.34 ± 9.58	−0.516*	0.608	61.35 ± 15.84	0.445*	0.658	1.977*	0.051
	After	67.07 ± 10.79			60.56 ± 18.27			2.330*	0.022
MEX-F (%)	Before	66.27 ± 13.28	0.666*	0.508	65.00 ± 18.66	1.067*	0.291	0.408*	0.684
	After	64.90 ± 12.96			63.19 ± 17.29			0.596*	0.552
MEX-B (%)	Before	49.98 ± 17.81	−1.206*	0.233	43.02 ± 22.01	−0.946*	0.349	1.817*	0.072
	After	53.50 ± 14.71			45.15 ± 19.46			2.580*	0.011
MEX-R (%)	Before	76.20 ± 16.54	−0.347*	0.730	73.98 ± 17.90	0.647*	0.520	0.680*	0.498
	After	77.12 ± 15.31			72.87 ± 16.31			1.421*	0.158
MEX-L (%)	Before	72.80 ± 16.79	0.658*	0.513	66.98 ± 18.40	−0.470*	0.641	1.741*	0.085
	After	71.48 ± 16.56			68.87 ± 17.91			0.803*	0.423
Step width (cm)	Before	15.65 (13.93–17.71)	0.159*	0.874	15.18 (13.06–17.57)	−0.722*	0.474	0.716*	0.476
	After	15.21 (13.95–16.61)			15.64 (13.82–18.10)			−0.037*	0.971
Speed (cm/s)	Before	17.49 ± 8.93	1.802*	0.077	16.34 ± 7.49	0.275*	0.785	0.696*	0.488
	After	15.82 ± 7.56			15.88 ± 8.55			−0.040*	0.968
End sway (°/s)	Before	8.14 (7.27–9.11)	1.640*	0.106	8.32 (6.98–9.79)	−0.290*	0.773	0.154*	0.878
	After	7.23 (6.64–7.95)			8.41 (7.27–9.67)			−0.842*	0.402

SOM: somatosensory; VIS: vision; VEST: vestibular; PREF: visual preference; SOTcom: Composite Equilibrium Score, the weighted average scores for 6 test conditions of the sensory integration test; RT: reaction time; MVL: movement velocity; DCL: directional control; EPE: endpoint excursion; MXE: maximum excursion; F: forward; B: backward; R: right; L: left; C: comprehensive.

*Two-sample t-tests were used for statistical comparison (two-sample t-test and paired t-test).

^@Non-parametric tests were used for statistical comparison.

3.4. Adherence and adverse effects

In the test group (n = 63), a total of 34 participants (54.0%) exercised ≥ three times per week on average, nine (14.3%) exercised twice per week on average, nine (14.3%) exercised once per week on average, and 11 (17.5%) exercised less than once a week on average.

Five (7.9%) participants reported mild muscle soreness; one (1.6%) reported a wrist ganglion cyst but continued exercising; and one (1.6%) reported the recurrence of periarthritis of the shoulder and stopped exercising. No one reported falls while exercising.

4. Discussion

The WHO Guidelines on Physical Activity and Sedentary Behavior [27] recommend that older individuals engage in functional balance and strength exercises for three or more days per week to enhance functional capacity and prevent falls. In this study, we designed and implemented a home-based strength and balance exercise regimen based on the characteristics and needs of self-reliant older individuals of advanced age. We found that after the 12-month intervention, the risk of new falls significantly decreased by 25.3% in the test group (RR = 0.747; 95% CI: 0.551–0.975; p = 0.027), but there was no statistically significant difference in the rate of new falls between the two groups (0.48 falls per person-year vs. 0.67 falls per person-year; IRR = 0.708;p = 0.251). Wang et al. [28] showed in their review that single falls were more common in older individuals in mainland China; it is essential to effectively decrease the risk of falls.

Through teaching based on exercise videos, self-management, and remote monitoring, 54.0% of the participants in the test group exercised ≥ three times per week on average, and 28.6% exercised once or twice per week on average, showing a certain degree of compliance. Compliance was identified as an important determinant of whether the participants could reach the required amount of exercise, and it had an impact on the effectiveness of the intervention [29]. In their systematic evaluation and meta-analysis on the prevention of falls in older individuals living in the community using exercise, Hughes et al. [30] found that the compliance rate was 27–97%. Individual factors (socioeconomic status, health status, motivation, physical functioning, and psychology) and project factors (mode of dissemination and place of exercise) may influence the inability of older individuals to persist with exercising. They also found that digital interventions, including TV courses, telephone counseling, and mobile applications, could help improve exercise compliance.

Feedback from the participants in this study for further improving our exercise regimen included suggestions, such as reducing and simplifying the movements, including cues to make it easier for older individuals to remember, adding subtitles in the teaching videos to help users with impaired hearing, bringing in human-computer interactive programs to correct movements, making the content more interesting, and increasing motivation for self-monitoring. In terms of safety, 7.9% of the participants in the test group reported mild muscle soreness, 1.6% reported new wrist ganglia and recurrence of periarthritis of the shoulder, and no one reported falls while exercising, suggesting that we should include more detailed instructions in our tutorial videos to avoid injuries from incorrect exercising.

Problems relating to balance and gait are the most important causes of falls in older individuals [10]. In a survey of older individuals in China [31], it was found that compared with the conditions of those aged 60, the static and dynamic balancing capacity significantly decreased from 70, and after 80, the overall balance and performance of specific tasks started declining significantly.

One of the indicators of balance is the stability of postural control. Postural control is defined as the act of maintaining, achieving, or restoring balance in any posture or activity. It is associated with three types of human activities: (1) maintenance of a specific posture, such as sitting or standing; (2) voluntary movements, such as postural transition; (3) response to external disturbances, such as tripping, slipping, or being pushed [32]. The first category belongs to static balance, and the other two categories belong to dynamic balance.

In this study, we used computerized dynamic posturography, and participants underwent the SOT, LOS, and TW tests. These evaluations help in objectively and quantitatively refining the evaluation parameters of balance function so as to further understand the effective components of strength and balance exercises. SOT results showed that SOTcom and VEST scores increased after intervention in the test group (p < 0.01 for all), and SOTcom scores decreased in the control group (p < 0.05). This suggests that balance exercises could improve vestibular function and comprehensive balance sensory input function while improving the ability to respond to external disturbances. Additionally, among the older individuals who did not receive the intervention, the comprehensive balance sensory input function was further weakened.

The human body perceives spatial information and relative motion through vestibular, visual, and proprioceptive senses and sends the information to the central nervous system. Then the central nervous system integrates the information and sends regulatory commands to direct the motor systems, such as muscles, bones, and joints, to adjust posture and movements, thus maintaining balance [33]. Vestibular sense was identified as a major component of balance maintenance, and the role of the vestibular system was found to be more crucial, especially when the supporting surface is unstable [34]. Physiological deterioration of the vestibular system, being a gradual process, seldom manifests as significant vertigo and can be easily overlooked by older individuals. However, the cumulative impairment of balance is characterized by movement intolerance, instability, and an unsteady gait. Particularly in the case of making a sudden turn, older individuals are more likely to fall due to a failure to correctly process motion information [35].

However, the vestibular system also has great plasticity and balance exercises include vestibular rehabilitation that can guide and enhance vestibular compensation to reshape balance with the aid of exercise-based physical therapy. Vestibular compensation refers to the process of remodeling and adaptation of the conflicting sensory afferent signals in the central nervous system due to vestibular lesions through the restoration of overall functions and behaviors [36].

The LOS results in our study showed that, in the test group, the indexes, such as response time, movement speed, directional control, and endpoint offset improved in some directions, while in the control group, the directional control capability weakened, suggesting that the stability of voluntary postural transition increased in the test group through strength and balance exercises while partial functions weakened in the control group. In addition to vestibular compensation, strength and balance exercises can also improve postural control through exercises related to proprioception, muscle strength, and muscle function. Impairments in proprioception pertaining to ankle tactile sensitivity, knee vibration sense, and knee position sense were identified as important and independent risk factors for falls in older individuals [37].

Post-intervention, grip strength increased in the test group and decreased in the control group (p < 0.05, p < 0.01, respectively). After intervention, there were no statistically significant differences in FTSST between the two groups compared to before intervention, suggesting that methods for training lower limb strength need improvement. Sarcopenia is a risk factor for falls in community-dwelling older adults [37]. Skeletal muscle mass and function decline with age; by 75–80 years old, muscle mass in older adults is half that of younger adults, with the greatest decline observed in the lower limbs. Between ages 20 and 80, the cross-sectional area of the quadriceps muscle can decrease by up to 40% [38]. Muscle strength declines more significantly than muscle mass; gradual decreases in strength occur from ages 30 to 50, with a 15% decline by age 60 and up to a 30% decline by age 80 [39]. Core muscle strength was identified as an important determinant of balance in older individuals [40]. Aagaard et al. [41] showed that strength training could induce effective responses even among older individuals of advanced age (>80), resulting in muscle hypertrophy and substantial changes in neuromuscular functions. Notably, training-induced changes in muscle mass and neurological function could improve functional ability in daily activities.

In our study, with regard to the results of the comprehensive geriatric assessment after 12 months, participants in the control group had increased scores of FES-I, FRAIL, and SDS while ADL and MoCA scores decreased (p < 0.05, p < 0.01); however, there was no significant change in the scores of participants in the test group, suggesting that strength and balance exercises could also postpone debility, depression, fear of falling, and cognitive and self-care decline, thus reducing the risk of falls.

5. Limitations

In this study, the sample was not completely representative, as 75% of the participants were male. Compared with grip strength, lower limb strength is more relevant to postural stability; however, we found that there was no significant difference in FTSST values in the two groups before and after intervention. On the one hand, FTSST, as an indirect method to evaluate lower limb strength, could impact the evaluation results. On the other hand, the participants expressed that lower limb strength exercises were more strenuous than upper limb strength exercises, and they could have reduced such exercises in the absence of external supervision; additionally, when they found it hard to complete partial actions, they might have chosen not to do such exercises, which could have affected the evaluation of the outcomes. The MoCA and grip strength at baseline show statistically significant differences between the test and control groups, suggesting that in subsequent studies, these assessments should be stratified in randomization to avoid bias in the results.

6. Conclusion

In this study, we found that even among older individuals of advanced age, strength, and balance exercises could activate physiological compensatory mechanisms, improve balance functions, and effectively decrease the risk of falls. The home-based exercise regimen combining video teaching with remote monitoring showed satisfactory effectiveness, safety, and compliance, giving it potential for promotion. However, there is still scope for improving the regimen, and we plan to continue efforts to optimize the exercise program and promote it further.

Funding Statement

This work was supported by Central Health Care Research Project (2020YB48).

Ethical approval

This study was conducted with approval from the Ethics Committee of Beijing Tongren Hospital, China Capital Medical University (TRECKY2020-041) and registered in the Chinese Clinical Trial Register website (www.chictr.org.cn, ChiCTR2100054932, 29 December 2021), and have adhered to CONSORT guidelines. This study was conducted in accordance with the declaration of Helsinki. Written informed consent was obtained from all participants.

Consent for publication

Not applicable.

Authors contributions

Conception and design of the research: Jian Zhou, Bo Liu, Jian-fang Xu, and Fu-bai-hui Wang. Acquisition of data: Hui Ye, Jin-Ping Duan, and Xin-wen Cui. Analysis and interpretation of the data: Jian Zhou, Bo Liu, Hui Ye, Jian-fang Xu, Fu-bai-hui Wang, and Jin-Ping Duan. Statistical analysis: Jian Zhou. Obtaining financing: Jian Zhou and Bo Liu. Writing of the manuscript: Jian Zhou and Xin-wen Cui. Critical revision of the manuscript for intellectual content: Bo Liu and Jian-fang Xu. All authors read and approved the final draft.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.