Effectiveness of telerehabilitation in postoperative outcomes in patients on hip fracture surgery: a meta-analysis of randomized controlled trials

Abstract

Objective

To comprehensively assess the rehabilitation efficacy of telerehabilitation compared with other traditional rehabilitation therapies in postoperative hip fracture patients.

Data sources

Seven electronic databases: PubMed, Embase, The Cochrane Library, Web of Science, CNKI, Wan Fang, and VIP Databases, were searched from inception until October 2023.

Study selection

Two independent reviewers selected randomized controlled trials (RCTs) that assessed the efficacy of telerehabilitation intervention approach to postoperative hip fracture rehabilitation. An outcome measure related to hip function, functional independence, anxiety levels, walking ability, quality of life, and treatment adherence were eligible.

Data extraction

Two reviewers independently used the Cochrane Risk of Bias 2 (RoB 2) tool for risk of bias and data extraction. RevMan 5.4 and Stata 15.1 were used for statistical analysis.

Data synthesis

Seventeen RCTs (n = 1577) met the inclusion criteria. Compared to the usual care group, the telerehabilitation group demonstrated a noteworthy enhancement in hip function, as evidenced by the Harris Hip Score (SMD = 1.05, 95% CI (0.64, 1.45)). Significant improvements in functional independence (Functional Independence Measure: SMD = 1.38, 95% CI (1.08, 1.68)), adherence to rehabilitation treatment (Medical Compliance Behavior Scale: SMD = 1.23, 95% CI (0.71, 1.76)), and quality of life (SMD = 1.04, 95% CI (0.42, 1.65)) were also observed in the telerehabilitation group. However, no statistically significant distinction was observed in anxiety improvement (as assessed by the Self-Rating Anxiety Scale: SMD = -0.67, 95%CI (-1.65, 0.31)) or in terms of walkability (Timed Up and Go Test: SMD = -0.06, 95% CI (-0.32, 0.20)) when compared to the usual care group. This may be related to patient participation, differences in telerehabilitation interventions, and inconsistent follow-up durations among different studies.

Conclusions

Current evidence suggests that telerehabilitation may help improve hip function, increase functional independence, and improve treatment adherence in patients after hip fracture surgery. Nonetheless, it does not demonstrate a significant impact on reducing patients’ anxiety or improving their walking ability.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13102-025-01169-4.

Introduction

Hip fractures are anatomically categorized into intracapsular and extracapsular fractures, including femoral neck fractures, fractures in the rotator region and within 5 cm below the lesser trochanter [1]. These fractures predominantly afflict individuals aged 65 years and older. They represent a prevalent orthopedic injury that threatens human health, and the incidence has been rising annually [2, 3]. It is estimated that by 2050, approximately 52% (3.3 million cases) of global hip fractures will manifest in Asian countries, notably in China [4, 5]. Hip fractures not only lead to high treatment costs [6–8], but also cause adverse consequences such as diminished quality of life [9], disability [10], and even death [11, 12]. With the aging of the demographic structure, the medical and economic burden associated with hip fractures will become a major challenge for society and families. Therefore, it is of utmost importance to enhance the prognosis and daily life functionality of hip fracture patients.

Previous studies have shown the positive effects of extended exercise rehabilitation [13], such as Lower-Limb Progressive Resistance Exercise [14], and balance training [15] in postoperative hip fracture rehabilitation. However, these therapies require a large number of time, space, equipment, and skilled instructors. Furthermore, due to the substantial costs associated with these treatments and the challenges posed by remote access, particularly during the COVID-19 pandemic, many patients struggle to complete the entire treatment regimen, leading to reduced treatment adherence that can potentially hinder hip function recovery. Consequently, the concept of telerehabilitation, which involves delivering rehabilitation services through internet-based technologies like videoconferencing or non-internet-based methods such as telephone consultations [16–18], has garnered increasing attention from researchers.

In recent years, telerehabilitation has been widely used in rehabilitation after cardiac surgery [19, 20], and postoperative rehabilitation for patients with hip fractures because of its cost-effectiveness, accessibility, and free of time and space limitations. Some studies have shown the positive effects of telerehabilitation in improving compliance with rehabilitation treatment, ameliorating quality of life, enhancing self-care ability, and promoting functional recovery of the hip joint in patients on hip fracture surgery. However, there are conflicting results in existing studies regarding the clinical effectiveness of telerehabilitation for postoperative hip fracture patients. For example, K.C. Cheng’s [21] randomized controlled trial showed that the use of a mobile app improved patients’ adherence to exercise but did not improve physical performance, self-efficacy, or reduce caregiver stress. Therefore, we conducted a systematic review and meta-analysis of randomized controlled trials (RCTs) to investigate the efficacy of telerehabilitation in patients recovering from hip fractures.

Methods

This systematic review and meta-analysis adhered to the recommendations outlined in the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analysis) [22] statement, together with the Cochrane Handbook [23]. The study protocol has been registered with the PROSPERO platform for systematic reviews under the reference number CRD42023420790.

Literature search

A thorough computerized search was conducted to identify RCTs published in both English and Chinese databases. The databases encompassed PubMed, Embase, The Cochrane Library, Web of Science, CNKI, Wan Fang, and VIP Database, spanning from their inception up to October 2023. Two authors conducted independent and systematic searches in the aforementioned databases, employing a search strategy that combined Medical Subject Heading (MeSH) terms with their corresponding free-text keywords, as outlined below: “Telerehabilitation” OR “Tele-rehabilitation” OR “Remote Rehabilitation” OR “Virtual Rehabilitation” OR “Telemedicine” OR “Virtual Medicine” OR “Tele Intensive Care” OR “Mobile Health” OR “mHealth” OR “Telehealth” OR “Mobile smart healthcare” AND “Hip fracture” OR “Hip replacement” OR “THA” OR “Intertrochanteric Fractures” OR “Femur Trochlear Fracture” OR “Subtrochanteric Fractures” OR “Fracture of trochanteric femur” OR “Neck fracture of femur” OR “Intertrochanteric fracture of femur”. The detailed search strategy is provided in Table S1.

Inclusion and exclusion criteria

Inclusion criteria were determined based on the PICOS framework as follows: (1) P (Participants): patients after hip fracture or total hip arthroplasty (THA); (2) I (intervention): the treatment group used any online form of telerehabilitation intervention such as smartphone and computer; (3) C (control): the control group used usual care measures; (4) O (outcomes): outcome measures included at least one of the following indicators: Harris Hip Score (HHS), Medical Compliance Behavior Scale (MCBS), Euro Quality of life Five Dimensions Questionnaire (EQ-5D), Self-Rating Anxiety Scale (SAS), the Medical Outcomes Study item short from health survey (SF-36), the Timed Up and Go Test (TUG), and the Functional Independence Measure (FIM); (5) S (study design): randomized controlled trials (RCTs). Exclusion criteria covered the following aspects: (1) The study subjects did not suffer from hip fracture or did not receive THA. (2) Intervention methods were not related to telerehabilitation interventions. (3) The study design was meta-analysis, review articles, guidelines, conference abstracts/posters, or studies with insufficient data.

Screening and data extraction

The articles were transferred to a citation management tool (Endnote X9; Thompson Reuters, Philadelphia, PA) to facilitate the automatic removal of duplicate records. Subsequently, in accordance with the aforementioned inclusion and exclusion criteria, two researchers conducted independent screenings of the titles and abstracts, resorting to the full text when deemed necessary. After the screening process, a Cochrane data extraction form was employed to retrieve the following data and information from each study: (1) fundamental study details like the title, first author’s name, publication year, and sample source; (2) key characteristics of the included sample, encompassing sample size, age, gender distribution, and diagnostic criteria; (3) particulars regarding the intervention, including specific measures, intervention duration, follow-up duration, and dropout rate; (4) outcome indicators. To ensure precision, the accuracy of data extractions was verified by two authors.

Quality assessment

The risk of bias in RCTs was assessed using the revised Cochrane risk of bias, version 2 (RoB2) [24]. This quality rating process was conducted independently by two researchers. In the event of any disagreements, resolution was achieved through discussion with an additional reviewer. RoB2 comprises five distinct domains, which encompass the assessment of bias during the randomization process, bias due to deviations from established interventions, bias arising from missing outcome data, bias pertaining to outcome measures, and bias associated with the selective reporting of outcomes. Each of these domains is rated as “low risk of bias,” “some concerns,” or “high risk of bias.” When all domains indicate a low risk of bias, the overall risk of bias is also categorized as low. If the assessment results in some concerns without any instances of high risk of bias, then the overall risk of bias is labeled as some concerns. However, the presence of a “high risk of bias” in any domain results in an overall high risk of bias.

Statistical analysis

Statistical analysis was conducted using Stata software (version 15.1) and RevMan 5.4. Since all the extracted outcome indicators were continuous variables, we utilized the standardized mean difference (SMD) along with a 95% confidence interval (CI) as the effect size indicator. To gauge heterogeneity, we employed the I² statistic. When the I² value exceeded 50%, indicating substantial heterogeneity for continuous outcomes, we opted for a random-effects model for data analysis. We also conducted sensitivity analysis to identify potential sources of heterogeneity. Conversely, if the I² value did not surpass 50%, signifying low heterogeneity, we utilized a fixed-effects model for analysis. In cases where outcome indicators displayed significant heterogeneity, we systematically excluded the data from each study one by one to evaluate the impact of each study on both heterogeneity and the overall effect size. Subsequently, we conducted subgroup analyses to identify potential factors contributing to the observed heterogeneity. Additionally, publication bias was tested using funnel plots (if the number of included studies exceeded 8) and the Egger test. A statistical significance threshold of α = 0.05 was applied to all analyses.

Results

Search results

The initial database search yielded 1369 studies. After eliminating duplicates during the screening process, 753 studies were removed, leaving 474 studies for screening based on titles and abstracts. Following this initial screening, 44 studies underwent a full-text assessment. Ultimately, 17 RCTs met the inclusion criteria. The selection process is illustrated in Fig. 1.

Fig. 1.

Flowchart illustrating the study selection process following PRISMA guidelines

Study characteristics

Table 1 provides a summary of the characteristics and specifics of the included studies. These studies were published between 2014 and 2023, with sample sizes ranging from 26 to 237 patients, totaling 1577 participants. Among the eligible studies, 5 employed internet-based interventions [25–29], while 12 utilized telephone-based interventions [29–40]. The duration of these interventions spanned from 2 weeks to 24 months. Among the 17 studies, HHS score was reported in 9 articles [28, 32–35, 37, 39–41], MCBS scores in 3 trials [32, 35, 41], EQ-5D scores in 3 articles [29, 30, 36], SAS scales in 4 articles [31, 32, 35, 40], SF-36 in 5 articles [25, 32, 34, 35, 38], the TUG in 5 articles [25–27, 36, 39], and the FIM in 3 articles [26, 28, 40]. The follow-up time varied from 1 month to 6 months. The telerehabilitation interventions across the included studies exhibited inconsistency in terms of the types of technology employed, which encompassed telephone [30, 34, 39, 41] computer [25, 27–29, 31, 36, 38, 40], and WeChat [32, 33, 35, 37].

Table 1.

Characteristics of included studies

Author/Year	Country	Sample size	Age (M ± SD)	Intervention		Intervention course	Follow-up time	Dropout (%)	Outcomes
				Treatment	Control
LL Li/ 2014	China	I: 100 C: 137	I: 66.20 ± 15.50 C: 67.30 ± 16.70	Telephone intervention with both patients and family caregivers	usual care	3 Months	Baseline, month 1,3,6(post-intervention)	4.8%	①②
DP Langford /2015	CA	I: 11 C: 15	I: 82.00 ± 9.00 C: 81.00 ± 10.00	Telephone follow-up provide recovery coaching	usual care	4 Months	Baseline, month 4 (post-intervention)	10%	③
YF Yang/ 2017	China	I: 59 C: 58	I: 74.52 ± 6.73 C: 75.91 ± 6.43	Telerehabilitation nursing of long-term guidance family hospital bed	usual care	2 Months	Baseline, month 2 (post-intervention)	NR	④⑤
SY Jiang/ 2018	China	I: 94 C: 102	I: 69.32 ± 7.61 C: 63.44 ± 6.54	Telephone follow-up and WeChat video	usual care	6 Months	Baseline, month 1,3,6 (post-intervention)	9.3%	①②④⑤
Q Ma/ 2018	China	I: 89 C: 91	I: 68.56 ± 11.71 C: 67.48 ± 12.23	Precede-proceed model combining WeChat platform rehabilitation	usual care	3 Months	Baseline, month 3 (post-intervention)	3.2%	①
CX Xu/ 2019	China	I: 50 C: 50	I: 60.40 ± 10.11 C: 61.54 ± 9.20	Mobile home orthopedic care platform	usual care	6 Months	Baseline, month 3,6 (post-intervention)	NR	① ⑤
JJ Zhang/ 2019	China	I: 32 C: 32	I: 39.21 ± 4.60 C: 39.23 ± 4.58	WeChat video remote guidance rehabilitation training	usual care	6 Months	Baseline, month 6 (post-intervention)	NR	①②④
S Eichler /2019	GER	I: 48 C: 39	I: 53.30 ± 7.00 C: 56.80 ± 5.70	The Meine Reha system	usual care	3 Months	Baseline, month 3 (post-intervention)	5.4%	⑤⑥
M Nelson / 2020	AUS	I: 35 C: 35	I: 62.00 ± 9.00 C: 67.00 ± 11.00	Technology-based home exercise program using an iPad application	usual care	6 Months	Baseline, month1,6(post-intervention)	15%	② ⑥
ZH Chen/ 2020	China	I: 44 C: 44	I: 74.62 ± 6.18 C: 73.44 ± 5.93	WeChat video remote guidance rehabilitation training	usual care	6 Months	Baseline, month6(post-intervention)	NR	①
M Ortiz-Piña /2021	ES	I: 28 C: 34	I: 75.86 ± 5.79 C: 80.38 ± 5.54	Tele-Rehabilitation- @ctivehip protocol	usual care	3 Months	Baseline, month 3(post-intervention)	12.68%	⑥⑦
CT Li/ 2022	HK	I: 15 C: 16	I: 76.50 ± 8.60 C:82.10 ± 9.70	Using the Caspar Health e-system	usual care	1 Month	Baseline, month 1(post-intervention)	3.2%	⑥
YY Zhang /2022	China	I: 27 C: 24	I: 77.00 ± 7.89 C: 75.17 ± 7.73	Home-oriented post-operative rehabilitation management system	usual care	3 Months	Baseline, month 1,3(post-intervention)	12.07%	①⑥⑦
Mora-T M/ 2022	ES	I: 30 C: 34	I: 75.77 ± 5.67 C: 80.38 ± 5.54	@ctivehip telerehabilitation program	usual care	3 Months	Baseline, month 3(post-intervention)	9.9%	③
QY Xu/ 2022	China	I: 30 C: 30	I: 75.10 ± 5.20 C: 74.20 ± 5.70	Rapid Rehabilitation Care Software	usual care	1 Month	Baseline, month 1(post-intervention)	NR	①⑤
YQ Chang/ 2022	China	I: 30 C: 29	I: 61.83 ± 8.51 C: 64.47 ± 7.66	The home tele-rehabilitation model based on smart	usual care	3 Months	Baseline, month 1,2,3(post-intervention)	7.8%	①⑥
WY Wu/ 2023	China	I: 43 C: 42	I: 74.28 ± 5.06 C: 72.00 ± 6.77	Home-Based Telerehabilitation Program	usual care	6 Months	Baseline, month 1,3,6(post-intervention)	3.3%	①④⑦

Notes: T, treatment; C, control. NR: No report, ①HHS: Harris Hip Score, ②MCBS: Medical Compliance Behavior Scale, ③EQ-5D: Euro Quality of life Five Dimensions Questionnaire, ④SAS: Self-Rating Anxiety Scale, ⑤SF-36: the MOS item short from health survey, ⑥TUG: the Timed Up and Go Test, ⑦FIM: the Functional Independence Measure

Risk of bias assessment

In terms of the overall bias assessment across these studies, 11.80% (2/17) studies [25, 36] were considered as low risk; 70.60% (12/17) studies [26, 28–34, 37–40] were considered some concerns, and 17.60% (3/17) studies [27, 35, 41] were considered as high risk. None of the 17 studies were deemed to have other bias concerns, with all of them classified as low risk in this regard. Figure 2 shows the risk of bias graph, and Fig. 3 illustrates the risk of bias summary.

Fig. 2.

Risk of bias graph for RoB2 (Rev. 2019) Excel output

Fig. 3.

Risk of bias summary chart for RoB2 (Rev. 2019) Excel output

Meta-analysis results

Harris hip score (HHS)

The HHS, a widely utilized scale both domestically and internationally for evaluating hip functionality, comprises four key components: pain assessment, functional evaluation, deformity analysis, and joint mobility assessment [42]. Nine trials [28, 32–35, 37, 39–41] involving 1097 participants reported the outcome of HHS. The results revealed a substantial difference in hip function levels between the treatment and control groups (SMD = 1.05, 95% CI 0.64, 1.45) (Fig. 4; Table 2), demonstrating that telerehabilitation considerably improved patients’ hip function.

Fig. 4.

Meta-analysis of change from baseline in HHS

Table 2.

Outcomes index results

Outcomes	No. of studies	Size of effect,95% CI	I2	P value
HHS	9(26, 32–35, 37, 39–41)	1.05, 95% CI (0.64, 1.45)	89.5%	0.000
FIM	3(26, 28, 40)	1.38, 95% CI (1.08, 1.68)	0	0.565
TUG	5(25–27, 36, 39)	-0.06, 95% CI (-0.32, 0.20)	27.9%	0.235
EQ-5D	3(29, 30, 36)	1.65, 95% CI (0.99, 2.31)	0	0.547
SF-36	5(25, 32, 34, 35, 38)	1.04, 95% CI (0.42,1.65)	90.5%	0.000
SAS	4(31, 32, 35, 40)	-0.67,95% CI (-1.65, 0.31)	96%	0.000
MCBS	3(32, 35, 41)	1.23, 95% CI (0.71, 1.76)	85.1%	0.001

Subgroup analysis

In order to investigate the potential origin of heterogeneity, we performed subgroup analyses based on technology types, average age, and intervention duration. The results indicated that telerehabilitation had a significant effect in improving hip function regardless of treatment duration (1 month, SMD = 0.86, 95%CI (0.41, 1.32); 3 months, SMD = 0.40, 95%CI (0.17, 0.64); 6 months, SMD = 1.81, 95%CI (0.62, 3.00)). Notably, there was substantial heterogeneity observed for the 1-month (I² = 77%) and 6-month (I² = 95%) durations, while no such heterogeneity was observed for the 3-month duration (I² = 0%). This suggests that treatment duration could potentially be a source of heterogeneity in the effect of telerehabilitation. In a comprehensive analysis of RCTs with a follow-up period of one and six months, potential sources of heterogeneity were identified. First, in four studies with one-month follow-up [28, 32, 39, 40], the average age of patients exceeded 70 years. These old patients exhibited cognitive decline, which may have affected their familiarity with the smart telerehabilitation system. Additionally, biases in pain perception and communication could contribute to greater heterogeneity in hip function scores for this patient group. Second, three studies conducted a follow-up of six months [35, 37, 41], and the average patient age was 40, 68, and 76 years, respectively. The considerable variation in age across these studies, combined with hip replacements likely contributed to heterogeneity in Harris hip scores. When stratified by technology, it was evident that the group utilizing phones exhibited notably higher HHS scores in comparison to the control group (SMD = 1.12, 95% CI 0.64, 1.60). Conversely, such a difference was not observed within the Internet usage group. Substantial heterogeneity was observed for both the phone and Internet usage groups (Table 3).

Table 3.

Subgroup analyses of Harris hip score

Subgroup	Stratification	No. of studies	P value for heterogeneity	I2	Standardized mean differences	P value
Duration	1 month	4(26,30,37,38)	0.004	77%	0.86 (0.41,1.32)	0.0002
	3months	2(31,32)	0.360	0	0.40 (0.17,0.64)	0.0008
	6months	3(33,35,39)	< 0.001	95%	1.81 (0.62,3.00)	0.003
Technology	Internet	2(26,38)	0.007	86%	0.80 (-0.15,1.75)	0.100
	Phone	7(26,30–33,35,37)	< 0.001	91%	1.12 (0.64,1.60)	< 0.001
Average
age	< 60	1(33)	/	/	/	/
	60–69	5(30–32,37,39)	0.002	76.9%	0.70 (0.39,1.01)	0.020
	70–79	3(26,35,38)	0.003	90.5%	1.14 (0.22,2.06)	0.0006

Functional independence measure (FIM)

The FIM score is employed as a post-discharge follow-up assessment for patients recovering from hip fractures [43]. A greater FIM score signifies an increased level of functional independence and decreased dependence, reflecting an enhanced capacity to engage in daily activities. In this meta-analysis of FIM, three studies [26, 28, 40] incorporating 219 patients were included. No notable heterogeneity was detected among these studies (I² = 0%), prompting the utilization of a fixed-effects model. The outcomes revealed a significant correlation between telerehabilitation intervention and enhanced FIM scores among hip fracture patients (SMD = 1.38, 95% CI 1.08, 1.68) (Fig. 5; Table 2).

Fig. 5.

Meta-analysis of change from baseline in FIM

Timed up and go test (TUG)

The TUG is a quick and easy test that assesses functional walking ability and predicts fall risk in older people. It requires no extra equipment or training [44]. Five studies [25–27, 36, 39], encompassing 329 participants, investigated the impact of telerehabilitation intervention on the TUG test. No substantial heterogeneity was observed among these studies (I² = 27.90%), leading us to employ a fixed-effects model. The analysis revealed that telerehabilitation intervention did not yield a significant enhancement in patients’ walking ability (SMD = -0.06, 95% CI -0.32, 0.20) (Fig. 6; Table 2).

Fig. 6.

Meta-analysis of change from baseline in TUG

Quality of life

Eight studies [25, 29, 30, 32, 34–36, 38] comprising a total of 690 patients, were incorporated into the quality-of-life meta-analysis. Different measurement scales (EQ-5D [29, 30, 36] and SF-36 [25, 32, 34, 35, 38]) were used for assessing patients’ quality of life in these studies. Among the studies we examined, the meta-analysis revealed that the telerehabilitation group exhibited a significant improvement in postoperative quality of life, as assessed by the SF-36 scale, in comparison to the control group (SMD = 1.65, 95% CI 0.99, 2.31). However, no such improvement was observed when using the EQ-5D scale. Nonetheless, the overall meta-analysis of quality of life showed that the telerehabilitation group significantly improved the quality of life in hip fracture patients in comparison to the control group (SMD = 1.04, 95% CI 0.42,1.65) (Fig. 7; Table 2).

Fig. 7.

Meta-analysis of Quality-of-Life Change from Baseline

Self-Rating anxiety scale (SAS)

SAS is widely used in clinical settings to assess patients’ anxiety status. The greater the score, the more evident the anxiety symptoms in the SAS [45]. Meta-analysis of the included four RCTs [31, 32, 35, 40] indicated that telerehabilitation did not significantly improve SAS scales compared to usual care (SMD = -0.67,95% CI -1.65, 0.31) (Fig. 8A; Table 2,).

Fig. 8.

Meta-analysis of change from baseline in A: SAS; B: MCBS

The medical compliance behavior scale (MCBS)

MCBS scale was developed by a Chinese scholar LL Li [41], and has shown excellent reliability [32]. The scale includes three aspects: patient behavior pattern, rehabilitation exercise and regular examination. A higher score indicates better compliance behavior. Three studies [32, 35, 41] were included in the meta-analysis of medical compliance behavior. Due to the presence of between-study heterogeneity (I² = 85.1%), a random-effects model was selected to combine the findings. This analysis indicated that telerehabilitation was linked to a higher level of medical compliance behavior (SMD = 1.23, 95% CI 0.71, 1.76) when compared to usual care (Fig. 8B; Table 2).

Sensitivity analysis

We performed four sensitivity analyses on outcome indicators exhibiting significant heterogeneity (HHS, SF-36, SAS, and MCBS) by systematically excluding one study at a time. In particular, the results of the sensitivity analysis for the MCBS outcome metrics showed that LL Li’s study [41] was the main source of heterogeneity. Following the exclusion of LL Li’s data, the studies displayed no substantial heterogeneity (I² = 7.00%), and the findings revealed a noteworthy disparity in MCBS levels between hip fracture patients and those receiving usual care (SMD = 1.52, 95% CI 1.23, 1.80). In the case of the remaining three outcome indicators, the omission of any single study did not significantly alter the results, underscoring the robustness of our pooled findings.

Publication bias

No significant publication bias was detected for HHS (Egger, p = 0.117) and quality of life (Egger, p = 0.729) outcomes. Publication bias was not assessed for the remaining outcome indicators due to the inclusion of fewer than eight studies in these analyses.

Discussion

This comprehensive meta-analysis incorporated 17 RCTs to examine the impacts of telerehabilitation on hip function, functional independence, walking ability, quality of life, anxiety, and compliance behavior in post-hip fracture surgery patients. The outcomes of our meta-analysis revealed a significant positive influence of telerehabilitation on enhancing hip function, functional independence, compliance behavior, and quality of life when compared to the control group. However, no notable differences between groups were observed concerning walking ability and anxiety.

Hip fractures have become an increasingly serious health problem in the general population, especially for old people, and patients with hip fractures are primarily treated by hip surgery such as total hip replacement and internal fixation [46]. Physical rehabilitation after hip surgery is one of the recommended components of accelerated surgical rehabilitation and constitutes a pivotal component of postoperative care. Its primary objectives include optimizing patients’ functionality and independence while minimizing postoperative complications [47]. Ongoing research pertaining to telerehabilitation, a home-based rehabilitation approach following hip fracture surgery, suggests that family-based caregivers play a beneficial role in the postoperative rehabilitation of hip fracture patients [48]. This influence is notably observed with enhanced functional independence [48], improved postoperative pain and reduced complications [49]. However, the responsibility of tending to older individuals with hip fractures has adverse effects on the well-being of family caregivers [50]. These caregivers often express feelings of fatigue and exhaustion while also indicating unfulfilled support needs [51]. Among their reported challenges are a deficiency in information sharing, role ambiguity, and disarray in the discharge planning process [48, 52]. The advent of telerehabilitation enables patients to utilize visual equipment within their home setting, facilitating remote guidance from healthcare professionals at distant terminals. Patients can seek advice regarding rehabilitation knowledge, receive guidance, and benefit from interventions in various aspects of rehabilitation. This approach aims to achieve the desired outcome of personalized rehabilitation training [53]. The application of telerehabilitation care for postoperative hip fracture patients can overcome the spatial distance, make up for the lack of continuity of home care for postoperative patients, help the community and patients in poor and remote areas to solve the problems of few medical resources and low medical technology [54], and provide professional and timely rehabilitation guidance [55].

Hip function

Our study found that telerehabilitation significantly improved hip function in postoperative hip fracture patients, which is consistent with the results of the pilot study by Kalron A et al. [56] as well as the findings of all the 10 included RCTs [28, 32–35, 37, 39–41] on the effect of telerehabilitation on hip function. Improved hip function means improved joint mobility and accelerated postoperative rehabilitation for patients. Currently, the number of hip fractures is increasing in older adults [3]. Our subgroup analyses indicate that telerehabilitation can substantially enhance hip function in old patients. Notably, in the telerehabilitation group, functional improvement in the hip was more pronounced among patients with an average age of 70–79 years than those with an average age of 60–69 years. The results of subgroup analyses also showed that the telephone-based intervention was more effective than the Web-based intervention. Our analysis may be because most included studies are elderly people. Compared with computers, mobile smart phones are more convenient, fast, economical, accessible, simpler to operate, and can be accepted by more people, thereby improving the compliance of telephone-based intervention and achieving better functional recovery effects after telerehabilitation.

Functional independence and quality of life

Our meta-analysis results additionally indicate that telerehabilitation exerts a beneficial influence on enhancing patients’ functional independence and quality of life. These findings align with the outcomes of systematic evaluations conducted by Tsuge T et al. [57], Wang J et al. [58], and Bedra et al. [59]. Hip fractures have a profound impact on patients’ quality of life and activities of daily living [60]. Studies have shown that less than half of hip fracture patients eventually regain their pre-fracture level of physical functional independence [61]. Furthermore, the quality of life remains significantly compromised for an extended period following the surgical procedure [62]. Frequently, this diminished quality of life manifests as functional constraints within the physical, social, and emotional domains [63]. Telerehabilitation empowers patients with a high level of support and assistance from healthcare professionals, fostering increased self-confidence. It encourages patients to confidently engage in activities within their capabilities, ultimately enhancing their functional independence and, in turn, improving their overall quality of life.

Anxiety and walking ability

The effect of telerehabilitation on improving patients’ anxiety and walking ability was not significant in this study. This is probably because anxiety relief often relies on in-depth face-to-face communication and real-time feedback between psychotherapists and patients, and the telerehabilitation system could not completely replace face-to-face emotional support and interactions [64, 65]. In addition, four studies on SAS outcome measures exhibited a wide age gap in participants enrolled. In one RCT, the average age of the subjects was about 40 years old, while the average age of subjects in the other three studies was about 70 years old. Age might lead to heterogeneity in anxiety levels after postoperative rehabilitation. The recovery of walking ability mainly depends on neuroplastic activation, and the effect of telerehabilitation training may be compromised by insufficient intensity or lack of multimodal stimulation (such as touch, proprioception). Previous studies have also shown that about 49-88.2% of patients with hip fractures exhibit varying degrees of fear of falling, which limits the walking ability of patients and enhances psychological stress caused by hip fracture pain, thereby inducing or aggravating the anxiety of patients [66, 67]. In addition, most telerehabilitation programs included in our study did not use wearable devices, which may have led to limitations in gait training and inaccurate scores in assessing patients’ walking ability. Although there are fewer pilot studies applying telerehabilitation to explore anxiety in patients after hip fracture surgery, we found that patients in the telerehabilitation group had an average SAS score about 5 points lower than the control group at 3 months postoperatively. We hypothesize that this improvement could be attributed to the accelerated rehabilitation process, which may have enhanced hip function, consequently improving patients’ quality of life and alleviating their concerns related to physical functioning.

Discussion of heterogeneity

While some studies exhibited data heterogeneity, we conducted appropriate sensitivity analyses and subgroup analyses to address this issue. To identify the origins of this heterogeneity, we carried out subgroup analyses based on factors such as the duration of hip function (1 month, 3 months, and 6 months), patients’ age, and the various telerehabilitation tools employed in the studies, respectively. Based on the subgroup analyses, it appears that this variability may be attributed to several factors, including the divergence in telerehabilitation tools employed across studies, significant variations in the mean age of the populations included in each study, and discrepancies in the timing of baseline hip function measurements. Additionally, it could be influenced by the inclusion of pilot studies with a higher risk of bias according to the quality assessment, geographical disparities in study settings, and other factors. However, it is important to note that we were unable to conduct subgroup analyses due to the limited number of studies available for inclusion. However, our analysis found that most subjects were Chinese, the number of subjects included was small, and intervention programs were inconsistent. These limitations may reduce the representativeness of the included studies to the overall population, thus affecting the validity of the results and conclusions of this study.

Possible obstacles

In telerehabilitation programs, participants were older and may be unfamiliar with smart devices. In addition, most telerehabilitation programs are mainly implemented by researchers, and there is a lack of professionals to implement telerehabilitation in a long-term and scientific manner and effectively record follow-up data. Given the above possible obstacles, cooperation with communities may be essential to install the telerehabilitation system for old patients with hip fractures in the community. The hospital regularly arranged for nursing personnel to carry out education, training, and quality control of the telerehabilitation system for community staff. Such methods can help solve the problem of poor digital literacy of old patients with hip fractures, improve compliance of postoperative rehabilitation training, and reduce the economic cost of rehabilitation training.

Strengths

Our systematic review and meta-analysis exhibit several notable strengths. Firstly, this systematic review is one of the few to investigate the positive impact of postoperative hip fracture rehabilitation in conjunction with telerehabilitation. Notably, the previous study by Tsuge T et al. [57] included only 7 RCTs and did not explore patients’ emotional well-being, hip joint function, functional independence, or medical treatment compliance. Our study supplements their findings [57]. Secondly, we meticulously compared all data with baseline measurements and calculated the final differences between pre- and post-intervention for each outcome indicator included in the meta-analysis. Thirdly, to enhance the quality of this meta-analysis, we exclusively considered RCTs and conducted a rigorous quality assessment using Version 2 of the Cochrane tool for assessing the risk of bias in randomized trials, RoB2. The overall risk of bias in the included studies was found to be low. Fourthly, the overall dropout rate for the programs included in our study remained low, not exceeding 20%. Lastly, the sensitivity and subgroup analyses conducted further bolster the robustness of our results.

Limitations and prospects

Certainly, our study has some limitations that warrant acknowledgment. Firstly, we did not delve into the analysis of various types of telerehabilitation, leaving us without insights into how different modalities of telerehabilitation impact postoperative recovery from hip fractures. Future research will include a network meta-analysis specifically focusing on different forms of telerehabilitation to determine the most effective intervention. Secondly, among the 17 trials included in our study, there was inconsistency in the frequency of cycles and the duration of interventions within telehealth programs. Future studies on the effects of telerehabilitation interventions should be conducted with caution and efforts made to achieve greater consistency in the application of telerehabilitation models. Third, the current conclusions are based solely on 6-month data, and thus the long-term effects of telerehabilitation programs remain unclear. Future studies should investigate the long-term impacts of the telerehabilitation program on both hip function and psychological well-being. In addition, it must be recognized that the advantages offered by telerehabilitation, including time, space and cost savings, have the potential to be widely applied in clinical rehabilitation care and maintenance of continuity of care [68]. With the development of communication technology and the Internet, telerehabilitation provides better rehabilitation opportunities for both healthcare providers and patients than traditional rehabilitation methods. This study provides a valuable reference for the clinical implementation of telerehabilitation after hip fracture surgery. In the future, hospitals or clinics can refer to the telerehabilitation program included in this study and establish a standard and personalized rehabilitation program for patients with hip fractures using smartphones and networked VR devices. The effect of telerehabilitation can be observed through videos and wearable monitoring devices, and the accuracy of the subjects’ movements can be reverse-checked using biosensing and other technologies through voice and graphics.

Conclusion

The systematic review and meta-analysis conducted in this study indicate that telerehabilitation interventions can potentially serve as valuable supplements to standard clinical care, contributing to the enhancement of hip function and quality of life following hip fracture surgery. However, there was no significant difference in relieving patients’ anxiety and improving their walking ability. Therefore, further RCTs exploring the application of telerehabilitation interventions are warranted to provide more comprehensive insights into their effectiveness. Telerehabilitation can also bridge the gap between patients’ postoperative rehabilitation and the shortage of rehabilitation therapists or nursing staff and regional restrictions. Real-time monitoring of telerehabilitation (such as wearable devices), video guidance, and personalized training programs are feasible and safe and can improve the continuity and compliance of postoperative care. In particular, patients with limited mobility or living in remote areas, patients with chronic diseases requiring long-term rehabilitation care, and patients at high risk of postoperative complications may mostly benefit from telerehabilitation.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

RCTs

randomized controlled trials

RoB 2

Risk of Bias 2

THA

Total hip arthroplasty

HHS

Harris Hip Score

MCBS

Medical Compliance Behavior Scale

EQ-5D

Euro Quality of life Five Dimensions Questionnaire

SAS

Self-Rating Anxiety Scale

TUG

Timed Up and Go Test

FIM

Functional Independence Measure

SMD

Standardized mean difference

CI

Confidence interval

SAS

Self-Rating Anxiety Scale

Author contributions

Conceptualization: Huangyi Xiao, Jun Wang; Methodology: Huangyi Xiao, Lanmo Lu; Formal analysis and investigation: Wenshu Zeng, Ziyu Yan, Jiankun Yuan; Writing - original draft preparation: Huangyi Xiao; Writing - review and editing: Huangyi Xiao, Jun Wang; Funding acquisition: Huangyi Xiao; Resources: Huangyi Xiao; Supervision: Jun Wang, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project was funded by the 2023 Yunnan Provincial Education Department Scientific Research Fund (Grant No. 2023Y0492) and the 2023 Yunnan University of Chinese Medicine Nursing Special Fund Key Projects (Grant No. YZHZ202303). The fund was not involved in any study design, data collection, analysis and interpretation, report writing, and article submission for publication.

Data availability

Data is provided within the manuscript or supplementary information files.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.