The physical and psychological effectiveness of mHealth-based physiotherapy for patients with chronic non-specific low back pain: A systematic review and meta-analysis

Abstract

Background

To determine (a) the effect of mHealth-based physiotherapy for patients with chronic non-specific low back pain (CNLBP) on fear, anxiety, depression, and self-efficacy; (b) which is the most effective on improving individuals’ pain intensity and physical disability through the comparison of the efficacy of mHealth-based physiotherapy with outpatient-based physiotherapy, home-based physiotherapy with simple supervision or unsupervision, and waiting-list group.

Methods

The systematic review and meta-analysis of randomized controlled trials (RCTs) was conducted in PubMed, MEDLINE (via Ovid), Scopus, Embase, the Physiotherapy Evidence Database (PEDro), Web of Science, and the Cochrane Central Register of Controlled Trials (CENTRAL) until September 20, 2024. Two independent reviewers (LQ and CZW) extracted information about origin, characteristics of study participants, eligibility criteria, characteristics of interventions, outcome measures and main results.

Results

A total of 37 RCTs involving 5832 participants were included. The risk of bias was generally low in the included studies. The results indicated that mHealth-based physiotherapy for individuals’ CNLBP was more effective in reducing pain intensity (standardized mean difference [SMD] −0.32, 95% CI −0.48 to −0.17; P < 0.001), improving physical disability (SMD −0.30, 95% CI −0.42 to −0.18; P < 0.001), and decreasing fear-avoidance (SMD −0.28, 95% CI −0.47 to −0.09; P = 0.004). However, the mHealth-based physiotherapy was less effective on decreasing anxiety (SMD 0.29, 95% CI 0.06–0.52; P = 0.01) and remained unclear in decreasing depression (SMD 0.13, 95% CI −0.05 to 0.31; P = 0.16) and improving self-efficacy (SMD 0.14, 95% CI −0.06 to 0.34; P = 0.18). In subgroup analyses of pain intensity and physical disability, mHealth-based physiotherapy for individual with CNLBP was more effective than outpatient-based physiotherapy, home-based physiotherapy with simple supervision or unsupervision and waiting-list groups.

Conclusion

Our meta-analysis suggested that mHealth-based physiotherapy holds significant potential for reducing pain intensity and fear-avoidance, and improving physical disability in individuals with CNLBP compared to traditional physiotherapy model. However, its effect was less on reducing anxiety and depression, and improving self-efficacy.

Keywords

Low back pain mobile health telerehabilitation physiotherapy mHealth

Introduction

Low back pain (LBP) is a condition characterized by discomfort, muscular strain, or inflexibility in the lumbosacral region, which may occasionally extends down the lower extremities.^1,2 The Global Burden of Disease (GBD) study reports that by 2020, the global prevalence of LBP will exceed 619 million people, a 60% increase from 1990, and is expected to reach 843 million by 2050.³ Studies showed 85% to 90% of individuals with LBP has unclear triggering factors, and these cases were referred to as non-specific LBP (NLBP).² LBP is classified as acute (less than 6 weeks), subacute (6–12 weeks), and chronic (more than 12 weeks) based on pain duration.⁴

Previous studies indicated that individuals with chronic NLBP (CNLBP) would develop depression, anxiety and fear-avoidance due to repeated pain and physical disability, and long-term psychological dysfunction might further exacerbate pain intensity and physical dysfunction, forming a vicious circle.⁵ Lim et al. showed that exercise, education and cognitive behavioral therapy (CBT) are the first-line treatment for CNLBP based on their summary and analysis of 30 global clinical practice guidelines (CPGs) related to LBP.⁶ However, the efficacy of physiotherapy for patients with CNLBP has been underestimated under traditional physiotherapy models, which necessitate that patients receive qualified physiotherapy at outpatient clinics or specialized orthopedic institutions.⁷ This may be due to the fact that a large number of individuals with CNLBP do not have access to receive professional physiotherapy services due to time restriction, the high expense of treatment and traveling, and transportation inconvenience.⁸ Moreover, The CPGs showed that the crucial role of home-based exercises and daily self-management for patients with CNLBP is often undermined due to lack of supervision and guidance.^8,9 Therefore, it is both urgent and critical for individuals with CNLBP to explore innovative physiotherapy approaches.

The advent of mobile health (mHealth) technologies offers a promising solution to these challenge.^9,10 Nowadays, mHealth technologies, including smartphone applications, wearable devices, and online platforms, are extensively used to provide remote guidance and supervision for home-based exercise protocols and education in individuals with CNLBP.¹¹ Previous systematic reviews have demonstrated the effects of app-based physiotherapy on pain intensity reduction and physical function improvement in individuals with CNLBP.^12–14 However, the outcomes of previous research varied according to the different interventions of control group.^15–18

Psychological factors have a strong association to the pain and physical disability of individuals with CNLBP.⁵ Nevertheless, the effect of mHealth-based physiotherapy is still unclear on improving depression, anxiety, and fear-avoidance. So far, previous studies has reported conflicting results on the mental health benefits of mHealth-based physiotherapy.^15,16 Additionally, although studies have shown that the effect of mHealth-based physiotherapy is strongly associated with individuals’ self-efficacy, which reflects their perceptions and confidence in their ability to exercise,^17,18 the efficacy of mHealth-based physiotherapy in enhancing self-efficacy has not been thoroughly investigated.

Based on previous research findings,^5,12–14 the objective of this systematic review and meta-analysis was to explore the efficacy of mHealth-based physiotherapy for patients with CNLBP in terms of psychological outcomes, including depression, anxiety, self-efficacy and fear-avoidance. Furthermore, our research team explored the efficacy of mHealth-based physiotherapy in reducing individuals’ pain intensity and improving physical disability compared with outpatient-based physiotherapy, home-based physiotherapy with simple supervision or unsupervision, and waiting-list groups.

Methods

Registration and approval

This review has been registered with PROSPERO (CRD4202458173). The literature search, reporting guidelines, and research implementation process all adhered to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines.¹⁹

Search strategy

Randomized controlled trials (RCTs) meeting the inclusion criteria were included in this review. The inclusion criteria are defined according to the “Population, Intervention, Comparison, and Outcomes” (PICO) model, as outlined in Table 1. There were no restrictions on publication year or language.

Table 1.

Inclusion criteria defined by PICOS.

PICOS question item	Inclusion criteria
Participants	Adults aged ≥18 years diagnosed with chronic non-specific low back pain
Intervention	mHealth-based physiotherapy, including exercise programs and guidance through smartphone applications, web-based interventions, internet-based systems, or “sensors + smartphone applications” for physical activity
Comparison	Outpatient-based physiotherapy, including physical activities guided by physiotherapists, face-to-face exercise training, and physiotherapy at clinics Home-based physiotherapy, including home-based exercise programs with simple supervision or unsupervision No intervention group, including waiting-list or no-treatment groups
Outcomes	Primary outcomes, including physical disability and pain intensity Secondary outcomes, including fear-avoidance, anxiety, depression, and self-efficacy
Design	Randomized controlled trials

Information sources

Our research team has identified relevant RCTs up to September 20, 2024 in the following databases: PubMed, MEDLINE (via Ovid), Scopus, Embase, the Physiotherapy Evidence Database (PEDro), Web of Science, and the Cochrane Central Register of Controlled Trials (CENTRAL). The reference lists of the included studies were also screened. The entire search process, including risk of bias assessment, data extraction and analysis, and completion of writing this article, was carried out until October 20, 2024.

Search strategy

The search strategy consisted of controlled vocabulary and keywords tailored to each database's requirements. The exact research terms used were as follows: (“low back pain” or “back pain” or “Backache”) and (“Mobile Applications” or “mobile health” or “digital health” or “telerehabilitation”) and (“Randomized controlled trial” or “systematic review” or “meta-analysis”). A detailed search strategy for PubMed can be found in Supplement 1. The research team selected updated searches from the initially used databases to refine the study results. Additionally, reference lists from prior reviews were consulted to identify other relevant studies.

Screening process

The studies obtained from the search were uploaded to Endnote X9 (Thomson Research Soft). After removing duplicates, two researchers (SWH and ZYH) independently screened the titles and abstracts to identify relevant trials. Full texts of potentially relevant studies were retrieved and independently reviewed by two assessors, who evaluated whether they met the inclusion criteria according to the predefined standards. In case of any disagreement between the assessors regarding inclusion, a third reviewer (ZHQ) acted as an arbitrator until a consensus was reached. For trials with unclear intervention or design, further clarification was sought from the corresponding author.

Risk of bias assessment

The quality of individual trials was assessed using the PEDro scale.²⁰ The PEDro scale consists of 11 items designed to score the methodological quality of RCTs. Each item, except for the first (whether inclusion criteria and sources are stated), contributes one point, for a total score of 10. Items 2 to 9 assess internal validity concerning random allocation, allocation concealment, baseline comparability, blinding, adequate follow-up and intention-to-treat analysis. Items 10 and 11 evaluate the completeness of statistical reporting. Published PEDro scores were extracted from the PEDro database, and trials without a PEDro score were assessed by two independent reviewers (SWH and ZYH), both trained in the PEDro scale.

Data collection process

Two reviewers (LQ and CZW) independently extracted all data using standardized forms developed for this review. The extracted data included basic information (e.g. author, publication year, country, and study type), characteristics of the study population (e.g. age, gender, sample size), methodological details (e.g. study design, participant characteristics, interventions, outcome measures), and results (e.g. data format, pre- and post-intervention scores, change scores).

Statistical analysis

If fewer than three studies were included, a systematic review was conducted. If a sufficient number of studies were included, a meta-analysis was performed using RevMan 5.4 (Nordic Cochrane Center) and Stata 17.0 (StataCorp) to estimate effect sizes, conduct subgroup analyses, and perform sensitivity analysis. Weighted mean differences (WMD) with 95% confidence intervals (95% CI) were used as effect sizes when the same measurement methods were employed; standardized mean differences (SMD) with 95% CI were used when measurement methods differed.²¹ The interpretation of SMD is as follows: SMD < 0.2 is small; 0.2 ≤ SMD < 0.5 is moderate; 0.5 ≤ SMD < 0.8 is large; SMD ≥ 0.8 is very large.

For studies reporting means and standard deviations (SDs) indirectly, mean differences (MD) and SMD were calculated according to the Cochrane Handbook. Heterogeneity was assessed by visually inspecting point estimates and overlapping confidence intervals, as well as using χ² tests and I² statistics. If I² > 50% and P < 0.01, a random-effects model was used for the meta-analysis; otherwise, if I² ≤ 50% and P > 0.01, a fixed-effects model was applied. Subgroup analyses based on different control groups were conducted to explore sources of heterogeneity. Sensitivity analysis was performed to assess the robustness of the results, and funnel plots and Egger's test were used to evaluate publication bias, using Stata 17.0 (StataCorp) as the analysis tool.

Results

Study selection process

A total of 3445 potentially relevant studies were identified, including PubMed (639), MEDLINE (via Ovid) (123), CENTRAL (1020), Embase (191), Web of Science (1296), Scopus (176), and PEDro (134). After removing duplicates and screening titles and abstracts, 176 articles met the criteria for full-text review based on the inclusion and exclusion criteria. Subsequently, 37 studies were selected for final inclusion in the analysis (see Figure 1).

Figure 1.

Flowchart of studies selection.

Characteristics of included studies

This review included 37 articles, all of which were RCTs published between 2004 and 2024. The total sample size across the studies was 5832 participants (3541 in the experimental group and 2291 in the control group). These studies primarily compared mHealth-based physiotherapy with outpatient-based physiotherapy (13/37, 35.2%), home-based physiotherapy with simple or no supervision (12/37, 32.4%), and waiting-list groups (12/37, 32.4%). The studies were conducted in the following regions: Europe (17/37, 45.9%), Asia (9/37, 24.3%), Africa (2/37, 5.4%), Oceania (4/37, 10.8%), North America (4/37, 10.8%), and South America (1/37, 2.7%). The follow-up periods across the studies ranged from 8 weeks to 4 months (see Tables 2 to 4). The primary and secondary outcome measures included pain intensity, physical disability, fear, anxiety, depression, and self-efficacy.

Table 2.

Summary of the intervention characteristic of the included studies (mHealth-based physiotherapy vs. outpatient-based physiotherapy).

Author (year)	Population and setting	Intervention and assessment	The content of the intervention	Dropout and outcome
Yang 2019²²	Sample size: 8 Average age: 40.75 years Average BMI: 22.3 Female: 50.0% Pain duration: over 3 mouths Country: China	Experimental group: self-management via “Pain Care APP “ + physiotherapy（n = 5) Control group: physiotherapy (n = 3) Duration: 4 weeks Assessment: baseline, week 2 and week 4	Self-management via APP: home-based exercise with APP 4 times daily Physiotherapy: consisted of manual therapy, electrophysical therapy, and traction	Dropout: 0 Experimental group:0; Control group: 0 VAS^a PSEQ^b RMDQ^c SF-36^d
Weise 2022²³	Sample size: 213 Average age: 57.40 years Female: 51.8% Pain duration: over 3 mouths Country: Germany	Experimental group：self-directed and home-based exercise program via “ the ViViRA APP” + physiotherapy (n = 108) Control group：face-to-face physiotherapy guided by physiotherapists (n = 105) Duration: 12 weeks Assessment: baseline, week 2, week 6 and week 12	Exercise program : 4 times per day for 12 consecutive weeks	Dropout:35.3% Experimental group:37.0%; Control group:32.4% NPRS^e
Villatoro-Luque 2023²⁴	Sample size: 68 Average age: 43.07 years Average BMI: 22.3 Female: 50.0% Pain duration: over 12 weeks Country: Spain	Experimental group: home-based exercise via “Whatsapp APP” (n = 34) Control group: clinic-based exercise under the guidance of physiotherapists (n = 34) Duration: 8 weeks Assessment: baseline and week 8	Home-based exercise program: 6 times per week for 8 weeks Clinic-based exercise program: the interval of 2 days between treatment sessions for 8 weeks	Dropout: 2.9% Experimental group:2.9%; Control group: 2.9% VAS^a TSK^f
Toelle 2019²⁵	Sample size: 86 Average age: 38.7 years Average BMI: 24.9 Female: 70.2% Pain duration: over 3 months Country: China	Experimental group: back pain-specific education + daily physiotherapy/physical exercise + mindfulness and relaxation techniques +"Kaia back pain APP” (n = 42) Control group: one face-to-face session of physical exercises per week (n = 44) Duration: 12 weeks Assessment: baseline, week 6 and week 12	Exercise program: daily exercise for 12 weeks	Dropout: 7.9% Experimental group:11.3%; Control group: 4.2% NPRS^e HFAQ^g VR-12-PCS^h VR-12-MCSⁱ
Shi 2024⁷	Sample size: 54 Average age: 42.0 years Average BMI: 22.5 Female: 59.3% Pain duration: over 3 months Country: China	Experimental group: home-based strength and stretching exercises + “HIRS APP” (n = 27) Control group: three face-to-face sessions of physical exercises under the guidance of PTs (n = 27) Duration: 8 weeks Assessment: baseline, week 2, week 4 and week 8	Home-based exercise program: at least 3 times per week Clinic-based exercise program: the interval of 2 days between treatment sessions for 8 weeks	Dropout: 7.9% Experimental group:11.0%; Control group: 18.5% ODI^j NPRS^e FABQ^k SF-36^d
Sandal 2021²⁶	Sample size: 461 Average age: 47.5 years Average BMI: 27.6 Female: 55% Pain duration: over 8 weeks Country: Norway and Denmark	Experimental group: self-management: “selfBACK APP” + physical activity + strength and flexibility exercises + daily education messages + general information about low back pain (n = 232) Control group: clinic-based physiotherapy (n = 229) Duration: 9 months Assessment: baseline, week 6, month 3, month 6, and month 9	Exercise program: 3–5 sessions every week	Dropout: 10.0% Experimental group: 8.7%; Control group: 11.2% RMDQ^c PSEQ^b NPRS^e FABQ^k
Priebe 2020²⁷	Sample size: 1245 Average age: 40.7 years Average BMI: 26.5 Female: 64.7% Pain duration: over 12 weeks Country: Germany	Experimental group: “Kaia back pain APP” + educational program + physiotherapy + mindfulness (n = 933) Control group: Standard of clinic-based care (n = 312) Duration: 3 months Assessment: baseline and month 3	Not found detailed exercise program	Dropout: 22.0% Experimental group: 27.2%; Control group: 16.4% DASS^l NPRS^e VR-12-PCS^h VR-12-MCSⁱ
Park 2023²⁸	Sample size: 100 Average age: 35.5 years Average BMI: 25.65 Female: 40.0% Pain duration: over 3 months Country: Germany	Experimental group: therapeutic exercises + “DrAI APP” (n = 50) Control therapy: conventional physical therapy modalities and therapeutic exercises (n = 50) Duration: 4 weeks Assessment: baseline and week 4	Conventional physical therapy modalities including heat, ultrasound, and transcutaneous electrical nerve stimulation (TENS) mobilization, and manipulation Exercises program including stretching, sling exercise and core stability training	Dropout: 10.0% Experimental group: 8.7%; Control group: 11.2% RMDQ^c ODI^j NPRS^e SF-12^m
Mbada 2019²⁹	Sample size: 47 Average age: 48.8 years Average BMI: 27.1 Female: 72.3% Pain duration: over 3 months Country: Nigeria	Experimental group: home-based McKenzie extension + back care education + video APP (n = 21) Control group: back care education + clinic-based McKenzie extension (n = 26) Duration: 6 months Assessment: baseline, month 3 and month 6	Exercise program: exercise 3 times every week	Dropout: 21.8% Experimental group: 9.0%; Control group: 34.6% RMDQ^c VAS^a SF-12^m
Koppenaal 2022³⁰	Sample size: 208 Average age: 47.7y Average BMI: 26.1 Female: 49.0% Pain duration: over 3 months Country: Netherlands	Experimental group: physical activity + educational information + “Keele STarT Back Screening Tool APP” (n = 104) The face-to-face physiotherapy group: face-to-face physiotherapy including physical activity (n = 104) Duration: 3 months Assessment: baseline and month 3	Exercise program: goal-oriented exercise 3 times every week	Dropout: 10.0% Experimental group: 13.5%; Control group: 6.7% ODI^j NPRS^e FABQ^k PCS^e GSES^o
Karaduman 2023³¹	Sample size: 66 Average age: 45.6y Average BMI: 26.3 Female: 45.5% Pain duration: over 3 weeks Country: Turkey	Experimental group 1: the clinic-based stabilization exercises + the face-to-face guidance of the physiotherapist. (n = 22) Experimental group 2: the home-based stabilization exercises + the guidance of the physiotherapist via videoconferencing (n = 22) Control group: the home-based stabilization exercises without guidance (n = 22) Duration: 4 weeks Assessment: baseline and week 4	Exercise program: exercise 3 days a week per 4 weeks	Dropout: 0% Experimental group: 0%; Control group: 0% ODI^j NPRS^e TSK^f
Holden 2024³²	Sample size: 47 Female: 52.2% Pain duration: over 3 weeks Country: Australia	Experimental group: 8-hour face-to-face training program + physiotherapy + Interviewing-based app (n = 26) Control group: usual physiotherapy (n = 21) Duration: 6 weeks Assessment: Baseline and 6 weeks	Not found detailed exercise program	Dropout: 13.0% Experimental group: 19.2%; Control group: 5% ODI^j NPRS^e PSEQ^b
Geraghty 2018³³	Sample size: 83 Average age: 55.3y Female: 58.6% Pain duration: over 3 months Country: UK	Experimental group 1: “the SupportBack internet” + goal-oriented exercise + usual care (n = 27) Experimental group 2: one session per week + goals-oriented exercise + self-management via “supportback internet” + usual care (n = 29) Control group： physiotherapy or pain clinics (n = 27) Duration: 6 weeks Assessment: Baseline and 6 weeks	Not found detailed exercise program	Dropout: 16.1% Experimental group 1: 7.1%; Experimental group 2: 16.7%; Control group: 24.1% RMDQ^c PCSⁿ NPRS^e TSK^f
Geraghty 2024³⁴	Sample size: 825 Average age: 55.3y Female: 58.6% Pain duration: over 3 months Country: UK	Experimental group 1: “the SupportBack internet” + goal-oriented exercise + usual care (n = 274) Experimental group 2: one session per week + goals-oriented exercise + self-management via “supportback internet” + usual care (n = 275) Control group： physiotherapy or pain clinics (n = 276) Duration: 12 months Assessment: Baseline, 3 months, 6 months, 12 months	Not found detailed exercise program	Dropout: 16.1% Experimental group 1: 7.1%; Experimental group 2: 16.7%; Control group: 24.1% RMDQ^c PCSⁿ NPRS^e TSK^f
Fatoye 2020³⁵	Sample size: 48 Average age: 47.36 years Female: 38.9% Pain duration: over 3 months Country: UK	Experimental group: telerehabilitation-based McKenzie extension exercise + back care education (n = 21) Control group: clinic-based McKenzie extension protocol + back care education (n = 27) Duration: 8 weeks Assessment: baseline, week 4 and week 8	Exercise program: exercise 3 times weekly	Dropout: 0% Intervention group: 0%; Control group: 0% ODI^j
Cui 2023¹⁵	Sample size: 140 Average age: 52.5 years Average BMI: 28.30 Female: 67.9% Pain duration: over 12 weeks Country: USA	Experimental group: telerehabilitation-based exercise and education + digital care program (n = 70) Control group: in-person physiotherapy (n = 70) Duration: 2 months Assessment: Baseline and 2 months	Exercise program: exercise session 3 times every week	Dropout: 10.0% Experimental group 1: 8.0%； Control group: 12.0% ODI^j GAD-7^p NPRS^e FABQ^k PHQ-9^q

VAS: Visual Analog Scale.

PSEQ: Pain Self-efficacy Questionnaire.

RMDQ: Roland–Morris Disability Questionnaire.

SF-36: 36-item Short Form.

NPRS: Numerical Pain Rating Scale.

TSK: Tampa scale for kinesiophobia.

HFAQ: Hannover Functional Ability Questionnaire.

VR-12-PCS: Veterans RAND 12-Item Health Survey-Physical component summary score.

ⁱ

VR-12-MCS: Veterans RAND 12-Item Health Survey-Mental component summary score.

ODI: Owestry Disability Index.

FABQ: Fear-Avoidance Beliefs Questionnaire.

DASS: Depression Anxiety Stress Scale.

SF-12-MH: 12-item Short-Form Health Survey-Mental Health.

ⁿ

PCS: Pain Catastrophizing Scale.

GSES: General Self-Efficacy Scale.

GAD-7: 7-item Generalized Anxiety Disorder Scale.

PHQ-9: Patient Health Questionnaire-9.

Table 3.

Summary of the intervention characteristic of the included studies (mHealth-based physiotherapy vs. home-based physiotherapy with simple supervision or unsupervision).

Author (year)	Population and setting	Intervention and assessment	The content of the intervention	Dropout and outcome
Zheng 2022 (1)³⁶	Sample size: 37 Average age: 35.2 years Average BMI: 21.5 Female: 75.7% Pain duration: over 3 mouths Country: China	Intervention group: “AUK exercise system APP” + Core stable training + self-compassion training (psychotherapy) (n = 19) Control group ："AUK exercise system” APP + Core stable training + 4 face-to-face sessions (n = 18) Duration: 16 weeks Assessment: baseline, week 4 and week 16	Core stable training: one face-to-face session + home-based exercise at least 3 times every week Self-compassion: 4 face-to-face sessions + home-based intervention	Dropout: 8% Experimental group:11.1%; Control group: 5.2% RMDQ^a NPRS^b GAD-7^c PHQ-9^d PCS^e PSEQ^f
Zheng 2022 (2)³⁷	Sample size: 40 Average age: 34.5 years Average BMI: 21.9 Female: 65.0% Pain duration: over 12 weeks Country: China	Experimental group: “Ding Talk APP” + stretching and strengthening training + patient education (n = 20) Control group : “Ding Talk APP” + stretching and strengthening training (n = 20) Duration: 18 weeks Assessment: baseline, week 6 and week 18	Stretching and strengthening training : at least 3 times per week Patient education: Cognitive behavior therapy + one online session per week	Dropout: 7.5% Experimental group:10%; Control group: 5% RMDQ^a NPRS^b GAD-7^c SF-36^g PCS^e SDS^h
Özden 2022¹⁷	Sample size: 50 Average age: 41.2 years Average BMI: 25.85 Female: 60.0% Pain duration: over 3 months Country: Ireland	Experimental group: home-based exercise program via “Fizyoweb APP” (n = 25) Conventional group: paper-based and home-based exercise program (n = 25) Duration:8 weeks Assessment: baseline and week 8	Exercise program : six regular, three optional and two booster sessions exercise every week	Dropout: 0% Experimental group: 0%; Control group: 0% ODIⁱ VAS^j SF-36^g TSK^k
Özden 2023³⁸	Sample size: 44 Average age: 44.9 years Average BMI: 25.57 Female: 56.85% Pain duration: over 3 months Country: Turkey	Experimental group: home-based exercise program via “PhysioAnalyst APP” for 8 weeks (n = 22) Control group: 4-week exercise program with APP + 4-week exercise program without APP (n = 22) Duration: 8 weeks Assessment: baseline, week 4 and week 8	Exercise program : two sets of 10 repetitions of exercise movements every day	Dropout: 0% Experimental group: 0%; Control group: 0% ODIⁱ PCS^e VAS^j
Mesa-Castrillon 2024³⁹	Sample size: 97 Average age: 63.18 years Average BMI: 29.7 Female: 59% Pain duration: over 3 months Country: Australia	Experimental group: home-based physical activity plan and progressive resistance exercise program via “PhysiApp APP” (n = 47) Control group: education + home-based exercise without APP (n = 50) Duration: 6 months Assessment: baseline, month 3 and month 6	Exercise program: 30- minute exercise program three times a week	Dropout: 21.8% Experimental group: 9.0%; Control group: 34.6% RMDQ^a NPRS^b SF-12^l PSEQf
López-Marcos 2024¹⁶	Sample size: 90 Average age: 50.6 years Female: 58.8% Pain duration: over 12 weeks Country: Spain	Experimental group: physical exercise including warm-up and core stabilization training via “Healthy Back App” (n = 45) Control group: 6 face-face sessions including health education and therapeutic exercise (n = 45) Duration: 12 weeks Assessment: Baseline, week 4 and week 12	Exercise program： exercise sessions 3 times every week	Dropout: 8.2% Experimental 1 group: 8.2%; Control group: 8.2% SF-12^l NPRS^b ODIⁱ
Ji 2022⁴⁰	Sample size: 40 Average age: 72.3y Average BMI: 22.0 Female: 45.5% Pain duration: over 3 weeks Country: China	Experimental group： Pilates exercises via video + “PBU device” (n = 20) control group: Pilates exercises program via video (n = 20) Duration: 8 weeks Assessment: baseline and week 8	Exercise program: 60-minute exercise sessions twice a week	Dropout: 0% Experimental group: 0%; Control group: 0% ODIⁱ VAS^j
POZO-CRUZ2024⁴¹	Sample size: 100 Average age: 46.17 years Female: 13.0% Pain duration: over 3 months Country: Spain	Experimental group: web-based exercises and postural interventions + personal e-mail interventions + standard care (n = 50) Control group: standard care (n = 50) Duration: 9 months Assessment: Baseline and 9 months	Exercise program: exercise daily for 9 months	Dropout: 10.0% Experimental group: 8.0%； Control group: 12.0% RMDQ^a
Lara-Palomo 2022⁴²	Sample size: 74 Average age: 47.9 years Female: 58.1% Pain duration: over 3 months Country: Spain	Experimental group： electroanalgesia and McKenzie exercises via an internet-based system (n = 39) Control group: transcutaneous electrical nerve stimulation （TENS）+ McKenzie Method exercises (n = 35) Duration: 8 weeks Assessment: baseline and 2 months	Exercise program: 3 exercise sessions weekly for 8 weeks	Dropout: 10.0% Intervention group: 12.5%； Control group: 7.5% ODIi RMDQ^a VAS^j TSK^k SF-36^g

RMDQ: Roland–Morris Disability Questionnaire.

NPRS: Numerical Pain Rating Scale.

GAD-7: 7-item Generalized Anxiety Disorder Scale.

PHQ-9: Patient Health Questionnaire-9.

PCS: Pain Catastrophizing Scale.

PSEQ: Pain Self-Efficiency Questionnaire.

SF-36: 36-item Short Form.

SDS: Self-Rating Depression Scale.

ⁱ

ODI: Owestry Disability Index.

VAS: Visual Analog Scale.

TSK: Tampa scale for kinesiophobia.

SF-12: 12-item Short-Form.

Table 4.

Summary of the intervention characteristic of the included studies (mHealth-based physiotherapy vs. waitlist group).

Author (year)	Population and setting	Intervention and assessment	The content of the intervention	Dropout and Outcome
Shebib 2019⁴³	Sample size: 177 Average age: 43.0 years Average BMI: 26.0 Female: 41.0% Pain duration: over 6 weeks Country: USA	Experimental group: sensor-guided physical exercise via APP + reading 1 to 2 educational articles per week (n = 113) Control group: receiving three educational articles + treatment-as-usual (n = 64) Duration: 12 weeks Assessment: baseline and week 12	Sensor-guided physical exercise program: 3 online sessions per week for 12 weeks	Dropout: 13.0% Experimental group:19.5%; Control group: 1.6% ODI^a VAS^b
Santos 2023⁴⁴	Sample size: 40 Average age: 67.8 years Average BMI: 29.2 Female: 82.5% Pain duration: over 6 weeks Country: Brazil	Experimental group: home-based exercise + education +"PAT-Back APP” (n = 20) Control group: a printed educational booklet + one face-to-face consultation with therapist (n = 20) Duration: 8 weeks Assessment: baseline and week 8	exercise program: 3–5 times per week	Dropout: 2.5% Experimental group: 0 ; Control group: 5% RMDQ^c NPRS^d DSC-E^e SES^f
Marcuzzi 2023⁴⁵	Sample size: 294 Average age: 50.6 years Average BMI: 26.8 Female: 58.8% Pain duration: over 3 months Country: Nigeria	Experimental group 1: physical activity + Strength and flexibility exercises + daily educational messages + additional resources and educational tools + APP (n = 97) Experimental group 2: educational messages + videos and descriptions of strength and flexibility exercises + additional resources and educational tools + e-Help website (n = 99) Control group: Any treatment related to low back pain (n = 98) Duration: 6 months Assessment: baseline, month 3 and month 6	Exercise program: 3–5 exercise sessions every week	Dropout: 12.6% Experimental 1 group: 18.8%; Control group: 0% RMDQ^c NPRS^d PSEQ^g
Licciardone 2020⁴⁶	Sample size: 100 Average age: 50.0 years Female: 84.4% Pain duration: over 3 months Country: USA	Experimental group: tailored intervention + usual care (n = 51) Control group: usual care (n = 49) Duration: 3 months Assessment: Baseline and month 3	Not found detailed exercise program	Dropout: 2.9% Experimental group: 1.9%; Control group: 2% NPRS^d ODI^a PROMIS-29^h
Kent 2015⁵	Sample size: 112 Average age: 43.3y Female: 54.5% Pain duration: over 3 weeks Country: Denmark	Experimental group: patient-tailored exercise + individualised movement pattern/postural rehabilitation + guidelines-based care + “ViMove motion-sensor system” (n = 58) Control Group: guidelines-based care (n = 54) Duration: 12 months Assessment: baseline, month 3 and month 12	Not found detailed exercise program	Dropout: 17.0% Experimental group: 22%; Control group: 11% RMDQ^c VAS^b FABQⁱ
Itoh 2022⁴⁷	Sample size: 99 Average age: 47.4y Average BMI: 23.9 Female: 44.4% Pain duration: over 3 months Country: Japan	Exercise group：the routine medical care + Patient Education + Exercise therapy + “the LINE app” (n = 48) Control group: Surgical Treatment + Surgical Treatment (n = 51) Duration: 12 weeks Assessment: baseline and week 12	Exercise program: approximately 1 to 3 minutes each day over 12 weeks	Dropout: 2.0% Experimental group: 4%; Control group: 0% NPRS^d TSK^j
Irvine 2015⁴⁸	Sample size: 597 Female: 60.0% Pain duration: over 3 months Country: USA	Experimental group 1: daily pain self-care activities + “the FitBack APP” (n = 199) Experimental group 2: received 8 program emails with content and prompts related to NLBP self-management (n = 199) Control group: only received assessment (n = 199) Duration: 4 months Assessment: Baseline, month 2 and month 4	Daily pain self-care activities including rest and relief, mindfulness, general fitness, and back pain-specific stretching and strength exercises	Dropout: 4.4% Experimental group 1: 8.0%; Experimental group 2: 2.5%； Control group: 2.5% TSK^j
Firzara 2023⁴⁹	Sample size: 42 Average age: 47.36 years Female: 38.9% Pain duration: over 3 months Country: UK	Experimental group: personalized physical activity + an education leaflet + physiotherapy + CBT (n = 21) Control group: conventional treatment (n = 21) Duration: 2 months Assessment: baseline and 2 months	Not found detailed exercise program	Dropout: 42.8% Experimental group: 28.1%； Control group: 58.1% RMDQ^c NPRS^d HADS^w
Chhabra 2018⁵⁰	Sample size: 93 Average age: 52.5 years Average BMI: 28.3 Female: 67.9% Pain duration: over 12 weeks Country: USA	Experimental group: physical activity and exercise adherence via “snapcare app” (n = 45) Control group： a written prescription including prescribed medicines and dosages and physical activity (n = 48) Duration: 12 weeks Assessment: Baseline and 12 weeks	Exercise program including back and aerobic exercises	Dropout: 2.2% Experimental group: 4.4%； Control group: 0% ODI^o RMDQ^a NPRS^b FABQ^q
Buhrman 2004⁵¹	Sample size: 51 Average age: 44.6 years Female: 62.5% Pain duration: over 3 months Country: Sweden	Experimental group: cognitive behavior therapy （CBT）+ physical exercise via email (n = 22) Control group：no intervention (n = 29) Duration: 3 months Assessment: baseline and 3 months	Exercise program: six regular and three optional sessions + 2 face-to-face coaching sessions	Dropout: 2.2% Experimental group: 8.8%； Control group: 0% HADS^k VAS^b
Amorim 2019⁵²	Sample size: 55 Average age: 58.3 years Average BMI: 28.1 Female: 53.1% Pain duration: over 3 months Country: Australia	Experimental group: physical activity + sedentary behavior information booklet + physical activities + “IMPACT app” (n = 31) Control group: booklet and brief advice + education (n = 24) Duration: 6 months Assessment: baseline and 6 months	Exercise including physical exercise, stretching and relaxation	Dropout: 2.2% Experimental group: 8.8%； Control group: 0% RMDQ^c DASS^l FABQⁱ NPRS^d
Almhdawi 2020⁵³	Sample size: 39 Average age: 58.3 years Average BMI: 28.1 Female: 53.1% Pain duration: over 3 months Country: Australia	Experimental group: office-based and home-based strengthening exercises + “Relieve my back app” (n = 20) Control group：nutritional education + traditional medical care (n = 19) Duration: 6 weeks Assessment: baseline and 6 weeks	Exercise program: 3-4 exercise sessions weekly	Dropout: 2.2% Intervention group: 8.8%； Control group: 0% ODI^a DASS^l VAS^b

ODI: Owestry Disability Index.

VAS: Visual Analog Scale.

RMDQ: Roland–Morris Disability Questionnaire.

NPRS: Numerical Pain Rating Scale.

DSC-E: Depression Scale of the Center for Epidemiological Studies.

SES: Self-Efficacy Scale for Chronic Pain.

PSEQ: Pain Self-efficacy Questionnaire.

ROMIS-29：Patient-Reported Outcomes Measurement Information System-29.

ⁱ

FABQ: Fear-Avoidance Beliefs Questionnaire.

TSK: Tampa scale for kinesiophobia.

HADS: Hospital Anxiety and Depression Scale.

DASS: Depression Anxiety Stress Scale.

Risk of bias assessment

Among the 37 studies, the PEDro scores ranged from 5 to 8 points: 24.3% of the studies scored 5, 35.1% scored 6, 32.4% scored 7, and 8.1% scored 8 (see Table 5). The studies met the following PEDro criteria: random allocation (100%), allocation concealment (48.6%), baseline group comparability (97.3%), assessor blinding (35.1%), adequate follow-up (81.1%), intention-to-treat analysis, reporting of between-group comparison results (100%), and reporting of point estimates and feasibility (100%). One trial implemented blinding for participants, while another blinded the therapists, as this type of intervention is challenging to blind.

Table 5.

PEDro quality rating of included studies with Y/N entries for each criterion.

Study	Eligibility criteria and source	Random allocation	Concealed allocation	Groups similar at baseline	Blinding of subject	Blinding of therapists	Blinding of assessors	Adequate follow-up	Intention-to-treat analysis	Results of between- group comparison reported	Point measure and variability	Total score (0 to 10)
Zheng 2022 (1)³⁶	Y	Y	Y	Y	N	N	N	Y	Y	Y	Y	7
Zheng 2022 (2)³⁷	Y	Y	N	Y	N	N	N	Y	Y	Y	Y	6
Yang 2019²²	Y	Y	Y	Y	N	N	N	N	Y	Y	Y	6
Weise 2022²³	Y	Y	N	Y	N	N	N	Y	Y	Y	Y	7
Villatoro-Luque 2023²⁴	Y	Y	Y	Y	N	N	Y	Y	N	Y	Y	7
Toelle 2019²⁵	Y	Y	N	Y	N	N	N	Y	N	Y	Y	5
Shi 2024⁷	Y	Y	Y	Y	N	N	N	Y	Y	Y	Y	7
Shebib 2019⁴³	N	Y	Y	Y	N	N	N	N	Y	Y	Y	6
Santos 2023⁴⁴	Y	Y	N	Y	N	N	Y	Y	N	Y	Y	6
Sandal 2021²⁶	Y	Y	Y	Y	N	N	N	Y	Y	Y	Y	7
Priebe 2020²⁷	Y	Y	N	N	N	Y	N	Y	Y	Y	Y	6
Park 2023²⁸	Y	Y	N	Y	N	N	Y	N	N	Y	Y	5
Özden 2022¹⁷	Y	Y	Y	Y	N	N	Y	Y	Y	Y	Y	8
Özden 2023³⁸	N	Y	N	Y	N	N	Y	Y	N	Y	Y	6
Mesa-Castrillon 2024³⁹	Y	Y	N	Y	N	N	N	N	Y	Y	Y	5
Mbada 2019²⁹	Y	Y	Y	Y	N	N	N	Y	N	Y	Y	6
Marcuzzi 2023⁴⁵	Y	Y	Y	Y	N	N	N	Y	Y	Y	Y	7
López-Marcos 2024¹⁶	Y	Y	Y	Y	N	N	Y	Y	N	Y	Y	7
Licciardone 2020⁴⁶	Y	Y	N	Y	N	N	N	Y	Y	Y	Y	6
Koppenaal 2022³⁰	Y	Y	Y	Y	N	N	N	Y	Y	Y	Y	7
Kent 2015⁵	Y	Y	N	Y	N	N	N	N	Y	Y	Y	5
Karaduman 2023³¹	Y	Y	N	Y	N	N	N	Y	N	Y	Y	5
Ji 2022⁴⁰	Y	Y	N	Y	N	N	N	Y	N	Y	Y	5
Itoh 2022⁴⁷	Y	Y	N	Y	N	N	N	Y	Y	Y	Y	6
Irvine 2015⁴⁸	Y	N	N	Y	N	N	N	Y	Y	Y	Y	5
Holden 2024³²	Y	Y	N	Y	N	N	Y	Y	N	Y	Y	6
Geraghty 2018³³	Y	Y	N	Y	N	N	Y	N	Y	Y	Y	6
Geraghty 2024³⁴	Y	Y	Y	Y	N	N	Y	Y	N	Y	Y	7
Firzara 2023⁴⁹	Y	Y	N	Y	Y	N	N	Y	N	Y	Y	6
Fatoye 2020³⁵	Y	Y	Y	Y	N	N	N	Y	Y	Y	Y	7
POZO-CRUZ 2024⁴¹	Y	Y	Y	Y	N	N	N	Y	Y	Y	Y	7
Cui 2023¹⁵	Y	Y	Y	Y	N	N	N	Y	N	Y	Y	6
Chhabra 2018⁵⁰	Y	Y	Y	Y	N	N	Y	Y	Y	Y	Y	8
Lara-Palomo 2022¹⁶	Y	Y	Y	Y	N	N	Y	Y	Y	Y	Y	8
Buhrman 2004⁵⁶	Y	Y	N	Y	N	N	N	Y	N	Y	Y	5
Amorim 2019⁵²	Y	Y	Y	Y	N	N	Y	N	Y	Y	Y	7
Almhdawi 2020⁵³	Y	Y	N	Y	N	N	Y	Y	N	Y	Y	5

Meta-analysis results

Effect of mHealth-based physiotherapy on pain intensity

Among the 37 studies, 33 (89.2%) were included in the evaluation of the effect of mHealth-based physiotherapy on pain intensity.^{5,7,15–18,22–30,33,34,36–40,42–47,49–53} Twenty-three studies used the Numeric Pain Rating Scale (NPRS),^{7,15,16,23,25–28,30–34,36,37,39,44–47,49,50,53} and 10 studies used the Visual Analog Scale (VAS).^{5,16,17,22,29,38,40,42,43,51} The heterogeneity test revealed significant heterogeneity between the studies (I² = 83%, P < 0.001), therefore a random-effects model was used for the analysis. The meta-analysis results indicated that, compared with the control group, mHealth-based physiotherapy had a moderate benefit in reducing pain intensity (SMD −0.32, 95% CI −0.48 to −0.17; P < 0.001; see Figure 2^{7,15,17,18,22–30,33,34,36,37,39,40,42–47,49–53}). Besides, the mHealth-based physiotherapy has a superior effect than the outpatient-based physiotherapy (SMD −0.25, 95% CI −0.49 to −0.01; P = 0.04; see Figure 2^7,15,22–34), home-based physiotherapy with simple supervision or without supervision (SMD −0.43, 95% CI −0.85 to −0.01; P = 0.05; see Figure 2^{17,18,33,34,36–40,42}), waitlist group (SMD −0.35, 95% CI −0.59 to −0.10; P < 0.001; see Figure 2^{15,31,43–47,49–53}).

Figure 2.

Forest plot of the effect of mHealth-based physiotherapy on the improvement of pain intensity.

Effect of mHealth-based physiotherapy on physical disability

Among the 37 studies, 32 (86.5%) were included in the evaluation of the effect of mHealth-based physiotherapy on physical disability.^{5,7,15,17,18,22,25–40,42–47,49,50,52,53} Fifteen studies used the Roland–Morris Disability Questionnaire (RMDQ),^{5,15,22,26,29,33,34,36,37,39,44,45,47,49,50,52} 15 studies used the Oswestry Disability Index (ODI),^{7,15,17,18,28,30–32,35,38,40,42,43,46,53} and 2 studies used the Veterans RAND 12-Item Health Survey–Physical Component Summary Score (VR-12-PCS).^25,27 The heterogeneity test revealed significant heterogeneity between the studies (I² = 67%, P < 0.001), therefore a random-effects model was used for the analysis. The meta-analysis results showed that, compared with the control group, mHealth-based physiotherapy had a moderate benefit in improving physical disability (SMD −0.30, 95% CI −0.42 to −0.18; P < 0.001; see Figure 3^{7,15,17,18,22,25–40,42–50,52,53}). Besides, the mHealth-based physiotherapy has a superior effect than the outpatient-based physiotherapy (SMD −0.15, 95% CI −0.28 to −0.02; P = 0.007; see Figure 3^{7,15,22,25–35}), home-based physiotherapy with simple supervision or without supervision (SMD −0.57, 95% CI −1.02 to −0.12; P = 0.02; see Figure 3^{17,18,36–40,42}), waitlist group (SMD −0.32, 95% CI −0.51 to −0.14; P < 0.001; see Figure 3^{5,31,43–47,49,52,53}).

Figure 3.

Forest plot of the effect of digital-based physiotherapy on the improvement of physical disability.

Effect of mHealth-based physiotherapy on depression, anxiety, fear-avoidance, and self-efficacy

Among the 37 studies, 17 (45.9%) were included in the evaluation of the effect of mHealth-based physiotherapy on depression^{7,15,17,18,22,25,27,36–39,42,44,46,49,51,52} The heterogeneity test revealed significant heterogeneity between the studies (I² = 62%, P < 0.001), therefore a random-effects model was used for the analysis. No significant difference was found between the intervention group and the control group (SMD 0.18, 95% CI 0.01–0.31; P = 0.16; see Figure 4^{7,15,17,18,22,25,27,36–39,42,44,46,49,51,52}).

Figure 4.

Forest plot of the effect of mHealth-based physiotherapy on decreasing depression.

Among the 37 studies, 8 (21.6%) were included in the evaluation of the effect of mHealth-based physiotherapy on anxiety.^{15,27,36,37,46,49,51,52} The heterogeneity test revealed significant heterogeneity between the studies (I² = 57%, P = 0.02), therefore a random-effects model was used for the analysis. The meta-analysis results indicated that mHealth-based physiotherapy was less effective than the control group in reducing anxiety (SMD 0.29, 95% CI 0.06–0.52; P = 0.01; see Figure 5^{15,27,36,37,46,49,51,52}).

Figure 5.

Forest plot of the effect of mHealth-based physiotherapy on decreasing anxiety.

Among the 37 studies, 13 (35.1%) were included in the evaluation of the effect of mHealth-based physiotherapy on fear-avoidance.^{7,15,17,18,24,30,31,33,34,36,37,47,48} The heterogeneity test revealed significant heterogeneity between the studies (I² = 71%, P < 0.001), therefore a random-effects model was used for the analysis. The meta-analysis results indicated that, compared with the control group, mHealth-based physiotherapy had a moderate benefit in reducing fear (SMD −0.28, 95% CI −0.47 to −0.09; P < 0.001; see Figure 6^{7,15,17,18,24,30,31,33,34,36,37,47,48}).

Figure 6.

Forest plot of the effect of mHealth-based physiotherapy on reducing fear-avoidance..

Among the 37 studies, 8 (21.6%) were included in the evaluation of the effect of mHealth-based physiotherapy on self-efficacy.^{22,26,30,32,34,36,44,45} The heterogeneity test revealed significant heterogeneity between the studies (I² = 66%, P < 0.001), therefore a random-effects model was used for the analysis. No significant difference was found between the intervention group and the control group (SMD 0.14, 95% CI −0.06 to 0.34; P = 0.18; see Figure 7^{22,26,30,32,34,36,44,45}).

Figure 7.

Forest plot of the effect of mHealth-based physiotherapy on the improvement of self-efficacy.

Sensitivity analysis

In this study, Stata 16.0 was used to conduct sensitivity analysis on pain intensity for evaluating the robustness and reliability of the results. The sensitivity analysis results demonstrated that excluding any single study did not impact the effect size of the mHealth-based physiotherapy on pain intensity (eFigure 1 in the Appendix 1), physical disability (eFigure 2 in the Appendix 1), depression (eFigure 3 in the Appendix 1), anxiety (eFigure 4 in the Appendix 1), fear-avoidance (eFigure 5 in the Appendix 1), and self-efficacy (eFigure 6 in the Appendix 1).

Publication bias

Funnel plots were used to assess publication bias in the effects of mHealth-based physiotherapy on pain intensity, physical disability, depression, anxiety, fear and self-efficacy and their funnel plots exhibited symmetrical patterns. In addition, our research team conduct Egger test for pain intensity (t33= −1.09, P = 0.286; eFigure 7 in the Appendix 1), physical disability (t32=-1.83, P = 0.077; eFigure 8 in the supplement 1), depression (t17 = 0.07, P = 0.946; eFigure 9 in the Appendix 1), anxiety (t8 = 0.93, P = 0.388; eFigure 10 in the Appendix 1), fear-avoidance (t13= −1.26, P = 0.388; eFigure 11 in the Appendix 1) and self-efficacy (t8 = 0.96, P = 0.375; eFigure 12 in the Appendix 1), and this results suggested no significant bias.

Discussion

Main findings

The aim of this systematic review and meta-analysis was to evaluate the effectiveness of mHealth-based physiotherapy for individuals with CNLBP in reducing pain intensity, improving physical disability, addressing psychological outcomes such as depression, anxiety, fear-avoidance and improving self-efficacy. In this research, mHealth-based physiotherapy demonstrated moderate benefits in reducing pain intensity (SMD −0.32, 95% CI −0.48 to −0.17; P < 0.001), improving physical disability (SMD −0.30, 95% CI −0.42 to −0.18; P < 0.001), and reducing fear-avoidance (SMD −0.28, 95% CI −0.47 to −0.09; P < 0.001) when compared to the control group. However, its impact remains unclear on improving individuals’ self-efficacy (SMD 0.14, 95% CI −0.06 to 0.34; P = 0.18) and decreasing depression (SMD 0.13, 95% CI −0.05 to 0.31; P = 0.16). In contrast, the effects of mHealth-based physiotherapy on decreasing anxiety (SMD 0.29, 95% CI 0.06–0.52; P = 0.01) were not as pronounced.

Effectiveness of mHealth-based physiotherapy for CNLBP patients

This review demonstrates that mHealth-based physiotherapy is more effective than outpatient-based physiotherapy, home-based physiotherapy with simple supervision or unsupervision and waiting-list groups in reducing pain intensity and improving physical disability, which is consistent with previous research findings.^9,10 This suggests that mHealth-based Physiotherapy may offer a more effective approach for individuals with NCLBP compared to the traditional physiotherapy. Furthermore, mHealth-based physiotherapy can address common barriers, such as cost, time, and transportation issues and improve work production and adherence, allowing NLBP patients to consistently and correctly follow home-based exercise protocol, ensuring exercise quality while providing qualified guidance to a larger number of patients.^9,10 Cui et al. found that mHealth-based technologies helps establish a strong alliance between patients and physical therapists, fosters frequent communication, and enhances patients’ understanding of LBP.¹⁵

As is known to us, individuals with CNLBP always develop depression, anxiety, and fear-avoidance due to long-term pain and physical disability, and these psychological factors further make a negative impact on pain intensity and physical disability, and quality of life.^54,55 The result of our research found the effect of mHealth-based physiotherapy has a moderate effect superior to the control group. However, mHealth-based physiotherapy did not show better effects in improving depression and anxiety. This may be due to the fact that the primary outcomes of studies included in this research were physical disability and pain intensity, and these studies more focused on reducing pain intensity and improving physical disability in individuals with CNLBP, not on alleviating depression and anxiety.^17,25,36,38 However, previous studies has shown that mHealth-based psychological interventions for individuals with CNLBP and depression can effectively decrease individuals’ anxiety and depression compared to usual care group, but the interventions of these studies did not lead to significant improvements in individuals’ pain intensity and physical disability, due to the absence of mHealth-based physiotherapy components.^55–58 For example, Baumeister et al. found after six months of mHealth-based psychological interventions, individuals with CNLBP and depression experience a more significant reduction of depression compared to the usual care group, but there was no notable improvement in their physical disability.⁵⁸ In the future, we could further explore the physical and psychological effect of mHealth-based physiotherapy combined with psychological interventions for individuals with CNLBP.

Previous research mostly concentrated on the effect of mHealth-based physiotherapy for individuals with CNLBP after short-term interval. For long-term follow-up, Sandal et al. found that app-based physiotherapy was more effective than outpatient-based physiotherapy in improving physical disability of NLBP patients after 9 months.²⁶ However, Geraghty et al. showed that there is no statistical difference between internet-based physiotherapy and outpatient-based physiotherapy, whether or not physiotherapist support was provided, in reducing pain intensity and improving physical disability.³⁴ These findings suggest that individuals with NLBP can effectively engage in self-management with the guidance of mHealth-based technologies.

Advantages and limitations

The advantages of mHealth-based physiotherapy have been identified. Initially, previous studies have shown that individuals with CNLBP who experienced mHealth-based physiotherapy had better adherence and engagement compared to those receiving traditional physiotherapy model.^7,23,24 This may be due to the fact that mHealth-based technologies allow individuals with CNLBP to follow a flexible schedule for completing home-based exercise protocols, which undoubtedly offers convenience. Additionally, mHealth-based physiotherapy saves individuals’ time and expenses on traveling and treatment, thereby reducing their financial burden.²⁵

However, the widespread adoption of mHealth-based physiotherapy faces certain limitations, particularly in remote areas where many patients with CNLBP have limited internet access, preventing them from receiving mHealth-based physiotherapy. Besides, a large number of individuals with CNLBP are not aware of the importance of mHealth-based physiotherapy and find it challenging to accept this innovative model of physiotherapy.^8,9

From the methodological perspective, this study has several strengths. First, we included 37 relevant studies, providing a robust dataset to evaluate the physical and psychological effectiveness of mHealth-based physiotherapy. Besides, our research team minimized bias by having at least two independent reviewers assess the study inclusion criteria, data extraction, risk of bias, and certainty of the evidence. Nevertheless, there are several limitations to consider. First, there is some clinical heterogeneity in the interventions across the studies, but we have defined mHealth-based physiotherapy as home-based exercise programs guided by smartphone apps, internet-based systems, web interventions, and “app + sensors.” Additionally, our literature search was limited to English-language studies, which may have introduced language bias, potentially affecting the generalizability of the findings. Furthermore, due to the nature of the interventions, some studies may have been susceptible to bias, such as the lack of blinding for participants and therapists.

Conclusion

The results of this systematic review and meta-analysis strongly suggested that mHealth-based physiotherapy is more effective in reducing pain intensity and improving physical disability for patients with CNLBP than traditional physiotherapy model, especially outpatient-based physiotherapy, which means that mHealth-based physiotherapy can provide opportunities for individuals who lack access to professional physiotherapy services. Compared to traditional physiotherapy, this studies found that mHealth-based physiotherapy has a better effect in reducing individuals’ fear-avoidance, but it was less effective in alleviating depression and anxiety, and improving self-efficacy, which might be due to the fact that mHealth-based physiotherapy of studies included in this review did not combined with the psychological interventions.

Supplemental Material

sj-docx-1-jtt-10.1177_1357633X251340037 - Supplemental material for The physical and psychological effectiveness of mHealth-based physiotherapy for patients with chronic non-specific low back pain: A systematic review and meta-analysis

Supplemental material, sj-docx-1-jtt-10.1177_1357633X251340037 for The physical and psychological effectiveness of mHealth-based physiotherapy for patients with chronic non-specific low back pain: A systematic review and meta-analysis by Weihong Shi, Lixia Chen, Yuhang Zhang, Wangshu Yuan, Qing Li, Zhengwei Chen, Houqiang Zhang, Qiyang Feng, Yingshan Gan, Huiling Zhang, Di Liu and Ye Lin in Journal of Telemedicine and Telecare

Footnotes

Author contributions

The inception of the study was facilitated through the collaborative efforts of Weihong Shi, Lixia Chen and Wangshu Yuan. Di Liu, Yingshan Gan and Weihong Shi undertook the experimental procedures and were responsible for the composition of the manuscript. The quality of papers was evaluated by Weihong Shi and Yuhang Zhang. The data extraction of papers was conducted by Zhengwei Chen and Qing Li. Substantial contributions to the analysis and preparation of the manuscript were made by Qing Li, Houqiang Zhang and Huiling Zhang. Qiyang Feng executed the data analyses, and Zhengwei Chen played a crucial role in the analytical process through constructive discussions.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: National High-Level Hospital Clinical Research Funding (grant number 2022-PUMCH-B-053).

ORCID iDs

Weihong Shi

Houqiang Zhang

Data availability statement

The data generated during this study are available from the corresponding author upon reasonable request. All data supporting the findings of this study are included within this published article.

Supplemental material

Supplemental material for this article is available online.

References

Ansari

Sharma

Khan

. The potential role of matrix rhythm therapy in managing chronic low back pain: a scoping review. Cureus 2024; 16: e72088.. PMID: 39574995; PMCID: PMC11579626.

Alam

Ansari

Zaki

, et al. Effects of physical interventions on pain and disability in chronic low back pain with pronated feet: a systematic review and meta-analysis. Physiother Theory Pract 2025; 41: 390–404.. Epub 2024 Mar 3. PMID: 38433468.

GBD 2021 Low Back Pain Collaborators. Global, regional, and national burden of low back pain, 1990–2020, its attributable risk factors, and projections to 2050: a systematic analysis of the global burden of disease study 2021. Lancet Rheumatol 2023; 5: e316–e329.

Maher

Underwood

Buchbinder

. Non-specific low back pain. Lancet 2017; 389: 736–747.

Kent

Laird

Haines

. The effect of changing movement and posture using motion-sensor biofeedback, versus guidelines-based care, on the clinical outcomes of people with sub-acute or chronic low back pain-a multicentre, cluster-randomised, placebo-controlled, pilot trial. BMC Musculoskelet Disord 2015; 16: 31.

Lim

Mak

Ngai

, et al. Nonpharmacological spine pain management in clinical practice guidelines: a systematic review using AGREE II and AGREE-REX tools. J Orthop Sports Phys Ther 2025; 55: 1–14.. PMID: 39680669.

Shi

Zhang

Bian

, et al. The physical and psychological effects of telerehabilitation-based exercise for patients with nonspecific low back pain: prospective randomized controlled trial. JMIR Mhealth Uhealth 2024; 12: e56580.

Foster

. Barriers and progress in the treatment of low back pain. BMC Med 2011; 9: 08.

Krebs

Duncan

. Health app use among US mobile phone owners: a national survey. JMIR Mhealth Uhealth 2015; 3: e101.

10.

Brannon

Cushing

. A systematic review: is there an app for that? Translational science of pediatric behavior change for physical activity and dietary interventions. J Pediatr Psychol 2015; 40: 373–384.

11.

Rising

Jensen

Moser

, et al. Characterizing the US population by patterns of Mobile health use for health and behavioral tracking: analysis of the national cancer institute's health information national trends survey data. J Med Internet Res 2020; 22: e16299.

12.

de Melo Santana

Raffin Moura

Martins de Toledo

, et al. Efficacy of mHealth interventions for improving the pain and disability of individuals with chronic low back pain: systematic review and meta-analysis. JMIR Mhealth Uhealth 2023; 11: e48204.

13.

Liu

Cai

, et al. The efficacy of e-health in the self-management of chronic low back pain: a meta analysis. Int J Nurs Stud 2020; 106: 103507.

14.

Molina-Garcia

Mora-Traverso

Prieto-Moreno

, et al. Effectiveness and cost-effectiveness of telerehabilitation for musculoskeletal disorders: a systematic review and meta-analysis. Ann Phys Rehabil Med 2024; 67: 101791.

15.

Cui

Janela

Costa

, et al. Randomized-controlled trial assessing a digital care program versus conventional physiotherapy for chronic low back pain. NPJ Digit Med 2023; 6: 21..

16.

López-Marcos

Díaz-Arribas

Valera-Calero

, et al. The added value of face-to-face supervision to a therapeutic exercise-based app in the management of patients with chronic low back pain: a randomized clinical trial. Sensors (Basel) 2024; 24: 67.

17.

Özden

Sarı

Karaman

ÖN

, et al. The effect of video exercise-based telerehabilitation on clinical outcomes, expectation, satisfaction, and motivation in patients with chronic low back pain. Ir J Med Sci 2022; 191: 1229–1239.. Epub 2021 Aug 6. Erratum in: Ir J Med Sci.Jun, 2022 ;191(3):1469. doi: 10.1007/s11845-021-02797-8. PMID: 34357527.

18.

Villar-Alises

García-Muñoz

Matias-Soto

, et al. Ehealth interventions for managing spine pain-benefits for pain, quality of life, catastrophizing and fear avoidance beliefs: an overview of systematic reviews with meta-analysis. J Orthop Sports Phys Ther 2024; 54: 1–18.. PMID: 39605288.

19.

Jüni

Egger

. PRISMAtic reporting of systematic reviews and meta-analyses. Lancet 2009; 374: 1221–1223.. PMID: 19819375.

20.

Cashin

McAuley

. Clinimetrics: physiotherapy evidence database (PEDro) scale. J Physiother 2020; 66: 59.

21.

McKenzie

Salanti

Lewis

, et al. Meta-analysis and the cochrane collaboration: 20 years of the cochrane statistical methods group. Syst Rev 2013; 2: 80.

22.

Yang

Wei

, et al. Smartphone-Based remote self-management of chronic low back pain: a preliminary study. J Healthc Eng 2019; 2019: 4632946.

23.

Weise

Zenner

Schmiedchen

, et al. The effect of an app-based home exercise program on self-reported pain intensity in unspecific and degenerative back pain: pragmatic open-label randomized controlled trial. J Med Internet Res 2022; 24: e41899.

24.

Villatoro-Luque

Rodríguez-Almagro

Aibar-Almazán

, et al. Telerehabilitation for the treatment in chronic low back pain: a randomized controlled trial. J Telemed Telecare 2023: 1357633X231195091. doi: https://doi.org/10.1177/1357633X231195091

25.

Toelle

Utpadel-Fischler

Haas

, et al. App-based multidisciplinary back pain treatment versus combined physiotherapy plus online education: a randomized controlled trial. NPJ Digit Med 2019; 2: 34.

26.

Sandal

Bach

Øverås

, et al. Effectiveness of app-delivered, tailored self-management support for adults with lower back pain-related disability: a selfBACK randomized clinical trial. JAMA Intern Med 2021; 181: 1288–1296.

27.

Priebe

Haas

Moreno Sanchez

, et al. Digital treatment of back pain versus standard of care: the cluster-randomized controlled trial, rise-uP. J Pain Res 2020; 13: 1823–1838.

28.

Park

Choi

, et al. Long-term effects of deep-learning digital therapeutics on pain, movement control, and preliminary cost-effectiveness in low back pain: a randomized controlled trial. Digit Health 2023; 9: 20552076231217817.

29.

Mbada

Olaoye

Dada

, et al. Comparative efficacy of clinic-based and telerehabilitation application of Mckenzie therapy in chronic low-back pain. Int J Telerehabil 2019; 11: 41–58.

30.

Koppenaal

Pisters

Kloek

, et al. The 3-month effectiveness of a stratified blended physiotherapy intervention in patients with nonspecific low back pain: cluster randomized controlled trial. J Med Internet Res 2022; 24: e31675.

31.

Karaduman

Ataş Balci

. The effects of in-person-supervised, tele-supervised, and unsupervised stabilization exercises on pain, functionality, and kinesiophobia in patients with chronic low back pain: a randomized, single-blind trial. Physiother Theory Pract 2024; 40: 2492–2502.

32.

Holden

O'Halloran

Davidson

, et al. Embedded motivational interviewing combined with a smartphone application to increase physical activity in people with sub-acute low back pain: a cluster randomised controlled trial. Braz J Phys Ther 2024; 28: 101091.

33.

Geraghty

AWA

Stanford

Stuart

, et al. Using an internet intervention to support self-management of low back pain in primary care: findings from a randomised controlled feasibility trial (SupportBack). BMJ Open 2018; 8: e016768.

34.

Geraghty

AWA

Becque

Roberts

, et al. Supporting self-management of low back pain with an internet intervention with and without telephone support in primary care (SupportBack 2): a randomised controlled trial of clinical and cost-effectiveness. Lancet Rheumatol 2024; 6: e424–e437.

35.

Fatoye

Gebrye

Fatoye

, et al. The clinical and cost-effectiveness of telerehabilitation for people with nonspecific chronic low back pain: randomized controlled trial. JMIR Mhealth Uhealth 2020; 8: e15375.

36.

Zheng

Liu

, et al. The effect of M-health-based core stability exercise combined with self-compassion training for patients with nonspecific chronic low back pain: a randomized controlled pilot study. Pain Ther 2022; 11: 511–528.

37.

Zheng

Liu

Zhang

, et al. Does m-health-based exercise (guidance plus education) improve efficacy in patients with chronic low-back pain? A preliminary report on the intervention's significance. Trials 2022; 23: 190.. PMID: 35241140; PMCID: PMC8892411.

38.

Özden

Güçlü

Tümtürk

, et al. The effect of visual feedback-based clinical monitoring application in patients with chronic low back pain: a randomized controlled trial. Eur Spine J 2024; 33: 505–516.

39.

Mesa-Castrillon

Simic

Ferreira

, et al. Effectiveness of an eHealth-delivered program to empower people with musculoskeletal pain in rural Australia: a randomized controlled trial. Arthritis Care Res (Hoboken). 2024; 76: 570–581.

40.

R. -X

Feng

Wang

Y. -Y

, et al. Application of online-pilates exercise based on information visualization training feedback in elderly people with low back pain: a randomized controlled trials. 2022 10th International Conference on Orange Technology (ICOT), Shanghai, China, 2022, pp. 1–4, doi: https://doi.org/10.1109/ICOT56925.2022.10008123.

41.

Del Pozo-Cruz

Adsuar

Parraca

, et al. A web-based intervention to improve and prevent low back pain among office workers: a randomized controlled trial. J Orthop Sports Phys Ther 2012; 42: 831–841..

42.

Lara-Palomo

Antequera-Soler

Matarán-Peñarrocha

, et al. Comparison of the effectiveness of an e-health program versus a home rehabilitation program in patients with chronic low back pain: a double blind randomized controlled trial. Digit Health 2022; 8: 20552076221074482.

43.

Shebib

Bailey

Smittenaar

, et al. Randomized controlled trial of a 12-week digital care program in improving low back pain. NPJ Digit Med 2019; 2: 1.

44.

Santos

AEDN

Nunes

ACL

Pereira

LSM

, et al. Physical activity supported by low-cost Mobile technology for back pain (PAT-back) to reduce disability in older adults: results of a feasibility study. Phys Ther 2024; 104: pzad153.. PMID: 37941491.

45.

Marcuzzi

Nordstoga

Bach

, et al. Effect of an artificial intelligence-based self-management app on musculoskeletal health in patients with neck and/or low back pain referred to specialist care: a randomized clinical trial. JAMA Netw Open 2023; 6: e2320400.

46.

Licciardone

Pandya

. Feasibility trial of an eHealth intervention for health-related quality of life: implications for managing patients with chronic pain during the COVID-19 pandemic. Healthcare (Basel) 2020; 8: 81.

47.

Itoh

Mishima

Yoshida

, et al. Evaluation of the effect of patient education and strengthening exercise therapy using a Mobile messaging app on work productivity in Japanese patients with chronic low back pain: open-label, randomized, parallel-group trial. JMIR Mhealth Uhealth 2022; 10: e35867.

48.

Irvine

Russell

Manocchia

, et al. Mobile-web app to self-manage low back pain: randomized controlled trial. J Med Internet Res 2015; 17: e1.

49.

Tun Firzara

Teo

Teh

, et al. Evaluation of an electronic clinical decision support system (DeSSBack) to improve low back pain management: a pilot cluster randomized controlled trial. Fam Pract 2023; 40: 742–752..

50.

Chhabra

Sharma

Verma

. Smartphone app in self-management of chronic low back pain: a randomized controlled trial. Eur Spine J 2018; 27: 2862–2874.

51.

Buhrman

Fältenhag

Ström

, et al. Controlled trial of internet-based treatment with telephone support for chronic back pain. Pain 2004; 111: 368–377.. PMID: 15363881.

52.

Amorim

Pappas

Simic

, et al. Integrating Mobile-health, health coaching, and physical activity to reduce the burden of chronic low back pain trial (IMPACT): a pilot randomised controlled trial. BMC Musculoskelet Disord 2019; 20: 71.

53.

Almhdawi

Obeidat

Kanaan

, et al. Efficacy of an innovative smartphone application for office workers with chronic non-specific low back pain: a pilot randomized controlled trial. Clin Rehabil 2020; 34: 1282–1291.

54.

Marshall

Joyce

Tseng

, et al. Changes in pain self-efficacy, coping skills, and fear-avoidance beliefs in a randomized controlled trial of yoga, physical therapy, and education for chronic low back pain. Pain Med 2022; 23: 834–843.

55.

Cruz-Díaz

Romeu

Velasco-González

, et al. The effectiveness of 12 weeks of pilates intervention on disability, pain and kinesiophobia in patients with chronic low back pain: a randomized controlled trial. Clin Rehabil 2018; 32: 1249–1257..

56.

Rutledge

Atkinson

Chircop-Rollick

, et al. Randomized controlled trial of telephone-delivered cognitive behavioral therapy versus supportive care for chronic back pain. Clin J Pain 2018; 34: 322–327.

57.

Sanabria-Mazo

Colomer-Carbonell

Borràs

, et al. Efficacy of videoconference group acceptance and commitment therapy (ACT) and behavioral activation therapy for depression (BATD) for chronic low back pain (CLBP) plus comorbid depressive symptoms: a randomized controlled trial (IMPACT study). J Pain 2023; 24: 1522–1540.. Epub 2023 Apr 25. PMID: 37105508.

58.

Baumeister

Paganini

Sander

, et al. Effectiveness of a guided internet- and Mobile-based intervention for patients with chronic back pain and depression (WARD-BP): a multicenter, pragmatic randomized controlled trial. Psychother Psychosom 2021; 90: 255–268.. Epub 2020 Dec 15. PMID: 33321501.