JMIR J Med Internet Res Journal of Medical Internet Research 1438-8871 JMIR Publications Toronto, Canada v24i9e37869 36066943 10.2196/37869 Review Review Digital Health Interventions for Musculoskeletal Pain Conditions: Systematic Review and Meta-analysis of Randomized Controlled Trials Kukafka Rita Rovetta Alessandro Valentijn Pim Peter MSc, PhD 1
Essenburgh Research & Consultancy Essenburgh Group Ceintuurbaan 2-40 Harderwijk, 3847 LG Netherlands 31 0341 217101 valentijn@essenburgh.com
2 https://orcid.org/0000-0002-9788-1275 Tymchenko Liza MSc 1 https://orcid.org/0000-0003-4877-4390 Jacobson Teddy BSc 3 https://orcid.org/0000-0002-1814-0138 Kromann Jakob BSc 3 https://orcid.org/0000-0003-1701-3492 Biermann Claus W MSc, PhD 3 https://orcid.org/0000-0001-8103-0794 AlMoslemany Mohamed Atef 4 https://orcid.org/0000-0002-7525-2638 Arends Rosa Ymkje MSc, PhD 1 5 https://orcid.org/0000-0002-7854-1371 Essenburgh Research & Consultancy Essenburgh Group Harderwijk Netherlands Department of Health Services Research School for Public Health and Primary Care, Faculty of Health, Medicine and Life Sciences Maastricht University Maastricht Netherlands Nordic Netcare Copenhagen Denmark Faculty of Medicine Menoufia University Menoufia Egypt University of Applied Sciences Utrecht Utrecht Netherlands Corresponding Author: Pim Peter Valentijn valentijn@essenburgh.com 9 2022 6 9 2022 24 9 e37869 10 3 2022 14 4 2022 17 5 2022 25 7 2022 ©Pim Peter Valentijn, Liza Tymchenko, Teddy Jacobson, Jakob Kromann, Claus W Biermann, Mohamed Atef AlMoslemany, Rosa Ymkje Arends. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 06.09.2022. 2022

This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on https://www.jmir.org/, as well as this copyright and license information must be included.

 Background

Digital health solutions can provide populations with musculoskeletal pain with high-reach, low-cost, easily accessible, and scalable patient education and self-management interventions that meet the time and resource restrictions.

 Objective

The main objective of this study was to determine the effectiveness of digital health interventions for people with musculoskeletal pain conditions (ie, low back pain, neck pain, shoulder pain, knee pain, elbow pain, ankle pain, and whiplash).

 Methods

A systematic review and meta-analysis was conducted. We searched PubMed and Cochrane Central Register of Controlled Trials (from 1974 to August 2021) and selected randomized controlled trials of digital health interventions in the target population of patients with musculoskeletal pain with a minimum follow-up of 1 month. A total of 2 researchers independently screened and extracted the data.

 Results

A total of 56 eligible studies were included covering 9359 participants, with a mean follow-up of 25 (SD 15.48) weeks. In moderate-quality evidence, digital health interventions had a small effect on pain (standardized mean difference [SMD] 0.19, 95% CI 0.06-0.32), disability (SMD 0.14, 95% CI 0.03-0.25), quality of life (SMD 0.22, 95% CI 0.07-0.36), emotional functioning (SMD 0.24, 95% CI 0.12-0.35), and self-management (SMD 0.14, 95% CI 0.05-0.24).

 Conclusions

Moderate-quality evidence supports the conclusion that digital health interventions are effective in reducing pain and improving functioning and self-management of musculoskeletal pain conditions. Low-quality evidence indicates that digital health interventions can improve the quality of life and global treatment. Little research has been conducted on the influence of digital health on expenses, knowledge, overall improvement, range of motion, muscle strength, and implementation fidelity.

 Trial Registration

PROSPERO CRD42022307504; https://tinyurl.com/2cd25hus

 eHealth models of care mobile health mHealth digital health pain telehealth telemedicine disability function quality of life mobile phone Introduction

Musculoskeletal conditions are considered the leading cause of global morbidity and have substantial individual, societal, and economic implications [1]. Musculoskeletal conditions account for one-fifth of the world’s total number of years lived with disability [1]. The burden of musculoskeletal conditions is predicted to increase dramatically in the coming years because of the aging population in Western countries. Musculoskeletal conditions include a broad range of health conditions affecting the bones, joints, muscles, and spine, as well as rare autoimmune conditions. Common symptoms include pain, stiffness, and loss of mobility and dexterity, which often interfere with people’s ability to perform daily activities. In the global research on the burden of disease, low back and neck pain were responsible for 70% of impairments [2]. The management of musculoskeletal pain conditions requires an evidence-informed innovative care model that stimulates self-management, including daily activities, self-care, patient-professional collaboration, and a collaborative practice model [3].

For musculoskeletal pain conditions, there has been increasing interest in integrating digital health interventions to accomplish the triple aim of better health outcomes, better patient experiences, and smarter use of health service resources. Various studies have found moderate-quality evidence that digital health interventions have a positive clinical benefit in the management of musculoskeletal conditions leading to pain and functional disability [4-7]. However, owing to differences in content, duration, and delivery, it is difficult to draw strong conclusions about the effectiveness of digital health interventions. Hence, little is known about which type or combination of digital health solutions is superior [5,8-10]. This lack of information serves as a barrier to identifying key characteristics aligned with effective and ineffective digital health solutions and their wider implementation. Recently, the World Health Organization (WHO) published a taxonomy for the standardization of various digital health interventions and vocabulary [11]. Although taxonomy is a useful tool to differentiate between the different types of digital interventions, it cannot distinguish between the micro, meso, and macro factors that influence digital health innovation and implementation [12]. This calls for a broad overview of the evidence by outlining digital health solutions at the patient, professional, provider, and system levels, as described by the Rainbow Model (Figure 1) [13]. It is important to identify the most effective type of digital health intervention and, in turn, the most efficacious combination of components (eg, patient, provider, organizational, and system level) for clinical and managerial responses to the evidence, as well as for policy decision-making.

Following the Rainbow Model of Integrated Care (RMIC) and WHO digital health taxonomy, we comprehensively analyzed the effectiveness of digital health interventions for musculoskeletal pain conditions in published randomized controlled trials (RCTs) and assessed the extent to which differences in outcomes may be explained by the different types of interventions.

Rainbow model for digital health interventions [14].

 Methods

A systematic review was conducted according to a protocol registered on PROSPERO (International Prospective Register of Systematic Reviews; registration number 307504) and the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines [15].

Literature Search

We searched the electronic databases PubMed and the Cochrane Library using musculoskeletal pain condition–specific and digital health–specific text words and Medical Subject Headings (Tables S1 and S2 in Multimedia Appendix 1 [16-71]) from their inception to August 2021.

 Study Selection Overview

A total of 2 researchers (LT and MAA) worked separately on study selection, eligibility criteria evaluation, risk-of-bias assessment, and data extraction, and disagreements were resolved through iteration and discussion. If this failed, a third arbitrary resolution was performed by a third author (PPV). Studies were considered eligible if they were RCTs with a follow-up of ≥1 month; included participants aged >18 years with a musculoskeletal (chronic) pain condition (ie, low back pain, neck pain, shoulder pain, knee pain, elbow pain, ankle pain, or whiplash); and comprised an evaluation of a digital health intervention in the clinical, professional, organizational, or system domains of the RMIC [13]. Each intervention had to describe ≥1 digital health service according to the description of the WHO digital health taxonomy [11] (Table S3 in Multimedia Appendix 1 [16-71]). Non-English studies were excluded from this review.

 Data Extraction and Risk-of-Bias Assessment

For each included study, 2 researchers (LT and MAA) independently extracted the data using a standardized data extraction form. Any inconsistency was resolved through iteration and discussion. When the required data were not reported in the article, the researchers contacted the authors for the missing information. If the required data could not be provided, the study was included only for qualitative review. The following methodological risks of bias were assessed for each selected study: sequence generation; allocation concealment; blinding of outcome assessors, care providers, and participants; completeness of outcome data; intention-to-treat analysis; and sponsor involvement in authorship [72]. The Covidence software was used to manage data extraction and risk-of-bias assessments [73].

 Data Synthesis and Analysis

The primary outcomes included pain, functioning, and quality of life, as assessed using recognized and validated measures [74]. Cost, emotional functioning, overall progress, range of motion, muscle strength, knowledge, self-management, and process-related outcomes were all secondary outcomes of interest.

A 3-step method was used to identify distinct subgroups of digital health interventions according to the domains of the RMIC. First, the appropriate number of clusters was determined through a hierarchical agglomerative clustering analysis using the Euclidean distance and average silhouette methods, which measures the quality of a cluster. We tested for outliers by using the cluster membership of the distance method, which indicates how well an observation fits into the cluster that it has been assigned to [75]. No outliers were identified based on the results of this analysis. Second, a nonhierarchical cluster analysis based on the k-means algorithm was performed to validate the results of the hierarchical procedure by using the initial cluster centroid number from hierarchical clustering as a starting point [76,77]. This method establishes the presence of clusters by determining the average of all the data points in a cluster.

The grouping of the clusters was evaluated by performing a principal component analysis (PCA), which required data normalization, and the eigenvalues were calculated and analyzed in a biplot graph [78,79]. Assumptions of the PCA were tested following the procedure described by Kassambara [80] (ie, linearity of the data, level of measurement, and outliers). Finally, the clusters were visually evaluated using cluster plots and PCA. To provide an interpretation of the cluster, the cluster means of the digital health interventions were applied.

We used DerSimonian and Laird random-effects models to summarize the treatment effects and expressed the results as standardized mean differences (SMDs) for continuous outcomes using different scales together with 95% CIs. The SMD calculations were based on the effect differences between the baseline and last follow-up assessment [81]. In the systematic review, we included relevant studies; for the meta- and subgroup analyses, at least three independent studies were required to justify the meta-analysis [82].

Heterogeneity in treatment effects between studies was assessed using the restricted maximum likelihood method (I2) statistics, with I2 values of 25%, 50%, and 75% corresponding to low, moderate, and high levels of heterogeneity, respectively [83]. Potential sources of statistical heterogeneity were explored using a priori subgroup analysis to determine whether the intervention duration (1-12 months or >12 months) or setting (clinic or home-based) affected heterogeneity. Evidence of small study effects was assessed through visual examination of funnel plots [84]. We conducted a sensitivity analysis of primary outcomes by excluding studies according to the following criteria: (1) high risk of bias, (2) long follow-up (≥12 months), and (3) large sample size (>200 participants). We used a minimum of 10 independent studies [81].

Descriptive statistics were used to summarize the data, where mean and SD were reported for continuous data and frequencies and percentages for categorical data. The distribution of all continuous variables was checked. The statistical significance for subgroup and sensitivity analysis was calculated using the test for subgroup differences provided in the R Studio (version 2021.09.01) package meta. All analyses were performed using the statistical software R Studio (Build 372), and libraries dmetar, esc, tidyverse, meta, grid, robvis, pvclust, and factorextra were used [85].

 Quality of Evidence

The quality of evidence was rated for each pooled analysis by using the Grades of Recommendation, Assessment, Development, and Evaluation approach [86]. The quality of evidence was not downgraded for performance or detection bias as perfect blinding is considered problematic for complex digital health interventions [82]. For each comparison, 2 researchers (LT and MAA) independently rated the quality of evidence for each outcome as “high,” “moderate,” “low,” or “very low.” Discrepancies were resolved through iteration and discussion.

 Results Search Results and Study Characteristics

A total of 983 publications of potential interest were identified. Of the 983 publications, after removing 18 (1.83%) duplicates, 965 (98.17%) publications were selected for title and abstract screening. Subsequently, of the 965 publications, 64 (6.63%) were selected for full-text screening, and 56 (5.8%) RCTs were considered eligible for inclusion, assessing 9359 participants. Approximately 6% (4/64) of studies reported incomplete outcomes; therefore, they were excluded from the effect analysis (Figure 2).

Flowchart of search strategy and study selection process.

 Intervention Characteristics

The characteristics of the interventions in the included studies are summarized in Tables S4 and S5 in Multimedia Appendix 1. All studies performed interventions in the home base of the participants, and in some cases, the setting of the study was a combination of the home base of the participants and the primary care clinic (26/56, 46%). In addition, of the 56 studies, we included 30 (54%) studies targeting musculoskeletal pain conditions, 10 (18%) studies targeting chronic pain conditions, 9 (16%) studies targeting postsurgery rehabilitation participants, and 7 (13%) studies focusing on patients with arthritis. Most of the interventions focused on patient conditions, such as target client communication (42/56, 75% studies; 6806/9359, 72.72% participants) and personal health tracking (38/56, 68% studies; 5881/9359, 62.84% participants). Digital health interventions at the professional level included telemedicine (55/56, 98% studies; 9331/9359, 99.7% participants), client information and registration (47/56, 84% studies; 8041/9359, 85.92% participants), health care provider decision support (23/56, 41% studies; 4520/9359, 48.3% participants), health care training (23/56, 41% studies; 4569/9359, 48.82% participants), health care provider communication (12/56, 21% studies; 2901/9359, 31% participants), and referral coordination (4/56, 7% studies; 527/9359, 5.63% participants). None of the studies incorporated health care providers in a scheduled activity planning intervention. Only 9% (5/56) of studies were targeted at the organizational level. All of these studies included health financing interventions (2363/9359, 25.25% participants). Furthermore, some studies were targeted at the system level and included data collection, management, and use interventions (23/56, 41% studies; 4648/9359, 49.66% participants). The duration of the interventions ranged from 2 weeks to 12 months (median 12 weeks). Two distinctive subgroups of digital health interventions were identified in the 56 articles.

The first cluster (32/56, 57% studies; 4565/9359, 48.78% participants) included interventions mainly in professional and client domains, mostly performed (23/32, 72%) in the home base of the participant. The second cluster (24/56, 43% studies; 4794/9359, 51.22% participants) comprised interventions in the organizational, professional, and client domains, mostly performed (17/24, 71%) in the home base and clinic settings. Four statistically significant differences across the subgroups for digital health interventions were identified through the cluster differences analysis: targeted health care provider decision assistance; referral coordination; health finance; and data collection, management, and use. The 2 clusters were named based on the characteristics of their digital health interventions (Table 1).

Clusters of digital health interventions (N=56).

Rainbow model intervention characterization Total studies, n (%) Cluster 1: patient-provider–level digital health interventions (n=32), n (%) Cluster 2: patient-provider-organizational–level digital health interventions (n=24), n (%) Cluster differences (P value)
Organizational domain
Health financing 5 (9) 0 (0) 5 (21) .006a
Data collection, management, and use 23 (41) 3 (9) 20 (83) <.001b
Professional domain
Client identification and registration 47 (84) 25 (78) 22 (92) .18
Health care provider decision support 23 (41) 7 (22) 16 (67) <.001c
Telemedicine 55 (98) 32 (100) 23 (96) .25
Health care provider communication 12 (21) 8 (25) 4 (17) .46
Referral coordination 4 (7) 0 (0) 4 (17) .02b
Health care provider training 23 (41) 13 (41) 10 (42) .94
Client domain
Targeted client communication 42 (75) 26 (81) 16 (67) .22
Personal health tracking 38 (68) 21 (66) 17 (71) .69

aSignificant at level .01.

bSignificant at level .001.

cSignificant at level .05.

 Quality of Included Studies

The risk of bias in the included studies is summarized in Figure 3. Overall, there was a low risk of bias for 80.6% (316/392) of the items, an unclear risk for 8.7% (34/392), and a high risk for 10.7% (42/392).

Summary of the risks of bias in included studies. For each quality item, low risk means that sufficient data were reported in the study to allow the assessment of quality, and the study fulfilled the criteria for the quality item; high risk means that sufficient data were reported in the study to assess quality, but the study did not fulfill the criteria for the quality item; and unclear risk means that incomplete data for the quality item were reported. N/A: not applicable.

 Effect of Digital Health Interventions Pain

Of the 56 studies, 37 (66%; 5323/9359, 56.88% participants) reported the treatment effects on pain. Digital health interventions had a small effect on pain compared with standard care management (SMD 0.19, 95% CI 0.06-0.31; Figure 4). However, there was evidence of high heterogeneity between studies (I2=81%). There was evidence of different effects on pain based on different types of digital health interventions (patient-provider: SMD 0.07, 95% CI −0.04 to 0.19; patient-provider-organization: SMD 0.34, 95% CI 0.08-0.60; P value for subgroup difference=.05). The quality of the evidence for pain was rated as moderate (Table 2).

Effect of digital health on pain. SMD: standardized mean difference.

Summary of findings and assessment of the quality of evidence for outcomes (N=56).

Outcomes Studies, n (%) Certainty assessment Effect Certainty
Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Individuals (n=9359), n (%) SMDa rate (95% CI)
Pain (follow-up: mean 25 weeks) 37 (66.1) Randomized trials Seriousb Not serious Not serious Not serious None 5323 (56.9) 0.19 (0.06 to 0.31) Moderate
Disability and function (follow-up: mean 27 weeks) 30 (53.6) Randomized trials Seriousb Not serious Not serious Not serious None 4849 (51.8) 0.14 (0.03 to 0.25) Moderate
Quality of life (follow-up: mean 25 weeks) 24 (42.9) Randomized trials Not serious Not serious Not serious Not serious None 3995 (42.5) 0.22 (0.07 to 0.36) High
Emotional functioning (follow-up: mean 29 weeks) 24 (42.9) Randomized trials Seriousb Seriousc Not serious Not serious None 3814 (40.8) 0.24 (0.12 to 0.35) Low
Self-management (follow-up: mean 26 weeks) 21 (37.5) Randomized trials Seriousb Not serious Not serious Not serious None 2857 (30.5) 0.14 (0.05 to 0.24) Moderate
Global improvement (follow-up: mean 42 weeks) 4 (7.1) Randomized trials Seriousb Not serious Not serious Seriousd None 795 (5.5) 0.25 (−0.44 to 0.93) Low

aSMD: standardized mean difference.

bMost of the studies had a high frequency of other bias.

cLarge heterogeneity between studies (I2>50%).

d95% CI includes the possible benefits from both control and digital health interventions.

 Disability and Function

Of the 56 studies, data on disability and function were reported in 30 (54%) studies (4849/9359, 51.8% participants). Digital health interventions slightly improved the functioning of people with musculoskeletal conditions (SMD 0.14, 95% CI 0.03-0.25); however, there was considerable heterogeneity among studies (I2=66%; Figure 5). There was little evidence that different types of digital health interventions affected treatment effectiveness (patient-provider: SMD 0.11, 95% CI −0.06 to 0.28; patient-provider-organization: SMD 0.17, 95% CI 0.02-0.32; P value for subgroup difference=.58). The quality of the evidence for disability and functional outcomes was moderate.

Effect of digital health on disability and function. SMD: standardized mean difference.

 Quality of Life

Digital health interventions had a slightly positive effect on health-related quality of life (24/56, 43% studies; 3995/9359, 42.69% participants; SMD 0.22, 95% CI 0.07-1.36). There was evidence of high-level heterogeneity between studies (I2=63%; Figure 6). There was little evidence that different types of digital health interventions had differing effects on quality of life (patient-provider: SMD 0.16, 95% CI 0.02-0.30; patient-provider-organization: SMD 0.35, 95% CI −0.04 to 0.75; P value for subgroup difference=.30). The quality of evidence for the quality of life was graded as high.

Effect of digital health on quality of life. SMD: standardized mean difference.

 Emotional Functioning

Of the 56 studies, 24 (43%; 3814/9359, 40.75% participants) reported data on emotional functioning. Digital health interventions had a positive effect on emotional functioning compared with usual care (SMD 0.24, 95% CI 0.12-0.35); however, there was evidence of heterogeneity between studies (I2=71%; Figure 7). There was little evidence of different treatment effects for different types of interventions (patient-provider: SMD 0.21, 95% CI 0.12-0.30; patient-provider-organization: SMD 0.32, 95% CI −0.27 to 0.92; P value for subgroup difference=.60). The quality of evidence for emotional functioning was low.

Effect of digital health on emotional functioning. SMD: standardized mean difference.

 Self-management

Of the 56 studies, 21 (38%) reported treatment effects on self-management behavior (2857/9359, 30.5% participants). Evidence suggests that digital health interventions have a small positive effect on self-management behaviors compared with usual care (SMD 0.14, 95% CI 0.05-0.24; Figure 8) with moderate quality of evidence. There was little evidence that different types of interventions affected treatment effectiveness (patient-provider: SMD 0.19, 95% CI 0.07-0.30; patient-provider-organization: SMD 0.14, 95% CI −0.13 to 0.26; P value for subgroup difference=.19).

Effect of digital health on self-management. SMD: standardized mean difference.

 Qualitative Synthesis

The qualitative analysis showed that digital health interventions have little or no effect on global improvement compared with standard care management (4/56, 7% studies, 795/9359, 8.49% participants; SMD 0.25, 95% CI −0.44 to 1.93). There was evidence of heterogeneity between studies (I2=87%), with a low quality of evidence. In addition, data on the range of motion were provided from 4% (2/56) of investigations involving 2.24% (210/9359) of participants; however, the treatment effects were highly ambiguous (Table S6 in Multimedia Appendix 1 [16-71]). Furthermore, 4% (2/56) of studies reported no effect of digital health on muscle strength (Table S6 in Multimedia Appendix 1 [16-71]). Of the 56 studies, the effects of digital health interventions on knowledge were reported in 2 (4%) studies (774/9359, 8.27% participants), and 1 (2%) study reported a significant effect (Table S6 in Multimedia Appendix 1 [16-71]). One of the studies reported an effect on satisfaction scores among participants, and another reported recovery expectation rates during the intervention (Table S6 in Multimedia Appendix 1 [16-71]). A cost analysis of digital health interventions for individuals with musculoskeletal pain conditions was presented in 4% (2/56) of studies (349/9359, 3.73% participants). In both investigations, digital health interventions were cost-effective and efficient (Table S6 in Multimedia Appendix 1 [16-71]).

 Publication Bias, Subgroup, and Sensitivity Analyses

There was little evidence of funnel plot asymmetry in treatment effects for pain, disability and function, quality of life, and emotional functioning (Figures S1-S5 in Multimedia Appendix 1 [16-71]). In addition, there was little evidence that digital health interventions had different effects on pain, disability and function, quality of life, emotional functioning, and self-management based on the duration of intervention (pain P=.66; disability and function P=.94; quality of life P=.45; emotional functioning P=.42; and self-management P=.66) or study setting (pain P=.80; disability and function P=.05; quality of life P=.63; emotional functioning P=.06; and self-management P=.06). The sensitivity analysis showed that restricting analyses to studies with lower risks of bias (pain P=.15; disability and function P=.58; quality of life P=.26; and self-management P=.39), follow-up <12 months (pain P=.22; disability and function P=.66; quality of life P=.31; emotional functioning P=.85; and self-management P=.48), or a small sample size (pain P=.88; disability and function P=.74; quality of life P=.62; emotional functioning, P=.19; and self-management P=.85) provided no different treatment effects for pain, disability and function, quality of life, and self-management (Table S7 in Multimedia Appendix 1 [16-71]). However, the risk of bias resulted in different results for emotional functioning (P=.01).

 Discussion Principal Findings

To the best of our knowledge, this meta-analytic review is the first to systematically assess the effectiveness of digital health interventions among people with musculoskeletal pain conditions. Pain, functioning, quality of life, emotional functioning, and self-management were all found to have small positive effects on a diverse set of digital health interventions. There was evidence that multicomponent interventions targeted at the client, provider, and organization levels had greater effects on pain than interventions targeted only at the client and provider levels. There was little evidence that different types of digital health interventions had different effects on other outcomes. The lack of high-quality evidence on global improvement, range of motion, muscle strength, and knowledge reinforces the need for further research on digital health for musculoskeletal pain conditions.

 Comparison With Existing Evidence

Previous reviews have also reported evidence on the effects of digital health interventions for reducing pain in musculoskeletal conditions [4,5,87-89]. However, most of these studies focused solely on chronic pain [87,88] or generic musculoskeletal conditions [5,7]. Further research is needed to corroborate our findings linking compound digital health treatments at the patient, provider, and organizational levels to reduced pain symptoms.

Our findings highlighting that digital health interventions improve function are consistent with earlier reviews of studies involving patients with generic musculoskeletal conditions [5,7]. Reviews focusing on chronic and nonspecific low back pain populations have reported limited evidence on the effects of digital health interventions on improving function [4,88]. The complexity of (chronic) pain management and the small number of RCTs included in earlier evaluations could explain the disparity in results.

This review indicates that digital health interventions have little effect on health-related quality of life. Previous systematic reviews have been inconsistent in this regard. For instance, 2 reviews suggested nonsignificant quality of life effects on musculoskeletal and chronic pain conditions [7,87], whereas 1 review reported a significant improvement in quality of life among people with nonspecific low back pain [4]. The variability of results may be explained by the differences in target populations, quality of the study design, and number of RCTs included in previous studies.

Similarly, this study has shown favorable outcomes for the emotional functioning of digital health interventions for people with musculoskeletal pain [87,88]. However, the sensitivity analysis provides evidence of the risk of bias confounding the effects, which requires further investigation. In line with other studies, this review found that digital health interventions may increase self-management behavior [88].

In all the reviewed studies, there was only a minimal reference to the cost-effectiveness of digital health interventions for musculoskeletal pain conditions. We could only include 2 studies reporting a significant cost reduction of digital health interventions compared with usual care [16,17]. Future trials should further explore whether digital health interventions can improve health outcomes related to musculoskeletal pain at lower costs than usual care. Data reporting for global improvement, range of motion, muscle strength, knowledge, and the delivery process of digital health were notably underreported, as has been observed in other reviews [4,5,87-89].

 Strengths and Limitations

This is the first review to synthesize the types of digital health interventions reported in the literature and quantify their effectiveness and confidence in treatment effects across a broad range of outcome measures. The strength of this review is that it was theoretically grounded in the WHO taxonomy [11] and the RMIC [13] to classify ambiguous digital health interventions reported in the literature. However, some limitations of this study must be acknowledged. First, it must be noted that confounding factors carry an inherent risk of bias, as evidenced by the large statistical heterogeneity across the pooled results for pain, function, quality of life, and self-management. In addition, the effects found in this study could have been influenced by differences in measurement scales and not by real differences in variability among study populations [90,91]. This should be further investigated in future studies. Moreover, the content of digital health interventions, diagnostic groups, and control conditions varied considerably, potentially biasing the results. Therefore, generalizing the overall findings to the management of musculoskeletal pain conditions should be treated with caution. Second, although we used a broad search technique, this evaluation could have been hindered by language bias, as we only included English-language literature. This means that our search may not reflect all available digital health interventions for musculoskeletal pain conditions. Third, we did not find any evidence of publication bias. It should be noted that the Egger test could potentially be misleading when used with continuous outcome measures [92]. Finally, although we abstracted and summarized the essential components of the interventions, there was minimal information on the type and intensity of digital health interventions offered.

 Relevance for Clinical Practice and Research

A major finding was that digital health interventions targeted at the clinical, provider, and organizational levels were effective in reducing pain for musculoskeletal conditions. To date, most studies have focused on isolated digital interventions targeted at the patient-provider level, such as telemedicine or targeted client communication. Future research should focus on improving the longitudinal design and on different types of interventions, drawing on the recent WHO taxonomy and the RMIC. Our findings should encourage interest in implementing real-world evaluation designs of digital health models to improve health care delivery as digital health interventions become more prevalent. Moreover, none of the studies included in this review covered the full breadth of the triple aim of assessing health, quality of care, and cost outcomes in conjunction. This emphasizes the importance of creating a core triple-aim result set for digital health interventions, which includes a defined set of outcomes that measure user experience, intervention quality, and costs.

 Conclusions

This review provides moderate-quality evidence that digital health interventions are effective in reducing pain and improving functioning and self-management of musculoskeletal pain conditions. Low-quality evidence indicates that digital health can improve the quality of life and global treatment. Although evaluations of the effects of digital health on costs, knowledge, global improvement, range of motion, muscle strength, and implementation fidelity are limited, these findings point to the need for more primary research into the particular combination of digital interventions that health care providers could use effectively.

Supplementary tables and figures of search strategy and study results.

 Abbreviations PCA

principal component analysis

 PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

 PROSPERO

International Prospective Register of Systematic Reviews

 RCT

randomized controlled trial

 RMIC

Rainbow Model of Integrated Care

 SMD

standardized mean difference

 WHO

World Health Organization

This work was supported by a research grant from the Nordic Netcare Group (Copenhagen, Denmark). TJ, JK, and CWB are employees of the Nordic Netcare Group.

PPV, LT, and RYA conceptualized and designed the study; PPV, LT, and MAA acquired, analyzed, and interpreted the data; PPV drafted the manuscript; PPV, LT, MAA, TJ, JK, CWB, and RYA critically revised the manuscript for important intellectual content; LT and PPV conducted statistical analyses; PPV and LT supervised the study. All authors read and approved the final manuscript. PPV and LT had full access to all data in the study and take responsibility for the integrity of the data and accuracy of the data analysis.

None declared.

 Sebbag E Felten R Sagez F Sibilia J Devilliers H Arnaud L The world-wide burden of musculoskeletal diseases: a systematic analysis of the World Health Organization Burden of Diseases Database Ann Rheum Dis 2019 06 78 6 844 8 10.1136/annrheumdis-2019-215142 30987966 annrheumdis-2019-215142 Whiteford HA Degenhardt L Rehm J Baxter AJ Ferrari AJ Erskine HE Charlson FJ Norman RE Flaxman AD Johns N Burstein R Murray CJ Vos T Global burden of disease attributable to mental and substance use disorders: findings from the Global Burden of Disease Study 2010 Lancet 2013 11 09 382 9904 1575 86 10.1016/S0140-6736(13)61611-6 23993280 S0140-6736(13)61611-6 Briggs AM Towler SC Speerin R March LM Models of care for musculoskeletal health in Australia: now more than ever to drive evidence into health policy and practice Aust Health Rev 2014 09 38 4 401 5 10.1071/AH14032 25086678 AH14032 Dario AB Moreti Cabral A Almeida L Ferreira ML Refshauge K Simic M Pappas E Ferreira PH Effectiveness of telehealth-based interventions in the management of non-specific low back pain: a systematic review with meta-analysis Spine J 2017 09 17 9 1342 51 10.1016/j.spinee.2017.04.008 28412562 S1529-9430(17)30142-0 Hewitt S Sephton R Yeowell G The effectiveness of digital health interventions in the management of musculoskeletal conditions: systematic literature review J Med Internet Res 2020 06 05 22 6 e15617 10.2196/15617 32501277 v22i6e15617 PMC7305565 Jiang S Xiang J Gao X Guo K Liu B The comparison of telerehabilitation and face-to-face rehabilitation after total knee arthroplasty: a systematic review and meta-analysis J Telemed Telecare 2018 05 24 4 257 62 10.1177/1357633X16686748 28027679 Cottrell MA Galea OA O'Leary SP Hill AJ Russell TG Real-time telerehabilitation for the treatment of musculoskeletal conditions is effective and comparable to standard practice: a systematic review and meta-analysis Clin Rehabil 2017 05 31 5 625 38 10.1177/0269215516645148 27141087 0269215516645148 Slattery BW Haugh S Francis K O'Connor L Barrett K Dwyer CP O'Higgins S Egan J McGuire BE Protocol for a systematic review with network meta-analysis of the modalities used to deliver eHealth interventions for chronic pain Syst Rev 2017 03 03 6 1 45 10.1186/s13643-017-0414-x 28253909 10.1186/s13643-017-0414-x PMC5335823 Heapy AA Higgins DM Cervone D Wandner L Fenton BT Kerns RD A systematic review of technology-assisted self-management interventions for chronic pain: looking across treatment modalities Clin J Pain 2015 06 31 6 470 92 10.1097/AJP.0000000000000185 25411862 Kelly M Fullen B Martin D McMahon S McVeigh JG eHealth interventions to support self-management in people with musculoskeletal disorders: a scoping review protocol JBI Evid Synth 2021 03 19 3 709 20 10.11124/JBIES-20-00147 33725715 02174543-202103000-00018 Recommendations on digital interventions for health system strengthening World Health Organization 2019 6 6 2021-07-06 Geneva, Switzerland World Health Organization https://www.who.int/publications/i/item/9789241550505 Asthana S Jones R Sheaff R Why does the NHS struggle to adopt eHealth innovations? A review of macro, meso and micro factors BMC Health Serv Res 2019 12 21 19 1 984 10.1186/s12913-019-4790-x 31864370 10.1186/s12913-019-4790-x PMC6925468 Valentijn PP Schepman SM Opheij W Bruijnzeels MA Understanding integrated care: a comprehensive conceptual framework based on the integrative functions of primary care Int J Integr Care 2013 3 22 13 e010 10.5334/ijic.886 23687482 PMC3653278 Rainbow Model for Integrated Care Essenburgh Group 2021 2021-07-06 https://www.essenburgh.com/en/rainbow-model-of-integrated-care Liberati A Altman DG Tetzlaff J Mulrow C Gøtzsche PC Ioannidis JP Clarke M Devereaux PJ Kleijnen J Moher D The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration J Clin Epidemiol 2009 10 62 10 e1 34 10.1016/j.jclinepi.2009.06.006 19631507 S0895-4356(09)00180-2 Paganini S Lin J Kählke F Buntrock C Leiding D Ebert DD Baumeister H A guided and unguided Internet- and mobile-based intervention for chronic pain: health economic evaluation alongside a randomised controlled trial BMJ Open 2019 04 09 9 4 e023390 10.1136/bmjopen-2018-023390 30967405 bmjopen-2018-023390 PMC6500312 Fatoye F Gebrye T Fatoye C Mbada CE Olaoye MI Odole AC Dada O The clinical and cost-effectiveness of telerehabilitation for people with nonspecific chronic low back pain: randomized controlled trial JMIR Mhealth Uhealth 2020 06 24 8 6 e15375 10.2196/15375 32357128 v8i6e15375 PMC7381065 Brattberg G Self-administered EFT (Emotional Freedom Techniques) in individuals with fibromyalgia: a randomized trial Integr Med 2008 7 4 30 5 Allen KD Arbeeva L Callahan LF Golightly YM Goode AP Heiderscheit BC Huffman KM Severson HH Schwartz TA Physical therapy vs internet-based exercise training for patients with knee osteoarthritis: results of a randomized controlled trial Osteoarthritis Cartilage 2018 03 26 3 383 96 10.1016/j.joca.2017.12.008 29307722 S1063-4584(17)31381-X PMC6021028 Lorig KR Laurent DD Deyo RA Marnell ME Minor MA Ritter PL Can a Back Pain E-mail Discussion Group improve health status and lower health care costs?: a randomized study Arch Intern Med 2002 04 08 162 7 792 6 10.1001/archinte.162.7.792 11926853 ioi10219 Buhrman M Fredriksson A Edström G Shafiei D Tärnqvist C Ljótsson B Hursti T Gordh T Andersson G Guided Internet-delivered cognitive behavioural therapy for chronic pain patients who have residual symptoms after rehabilitation treatment: randomized controlled trial Eur J Pain 2013 05 17 5 753 65 10.1002/j.1532-2149.2012.00244.x 23139021 Hørdam B Sabroe S Pedersen PU Mejdahl S Søballe K Nursing intervention by telephone interviews of patients aged over 65 years after total hip replacement improves health status: a randomised clinical trial Scand J Caring Sci 2010 03 24 1 94 100 10.1111/j.1471-6712.2009.00691.x 19422632 SCS691 Buhrman M Skoglund A Husell J Bergström K Gordh T Hursti T Bendelin N Furmark T Andersson G Guided internet-delivered acceptance and commitment therapy for chronic pain patients: a randomized controlled trial Behav Res Ther 2013 06 51 6 307 15 10.1016/j.brat.2013.02.010 23548250 S0005-7967(13)00048-X Bennell KL Nelligan RK Rini C Keefe FJ Kasza J French S Forbes A Dobson F Abbott JH Dalwood A Harris A Vicenzino B Hodges PW Hinman RS Effects of internet-based pain coping skills training before home exercise for individuals with hip osteoarthritis (HOPE trial): a randomised controlled trial Pain 2018 09 159 9 1833 42 10.1097/j.pain.0000000000001281 29794609 00006396-201809000-00019 Berman RL Iris MA Bode R Drengenberg C The effectiveness of an online mind-body intervention for older adults with chronic pain J Pain 2009 01 10 1 68 79 10.1016/j.jpain.2008.07.006 18774342 S1526-5900(08)00685-8 Buhrman M Fältenhag S Ström L Andersson G Controlled trial of Internet-based treatment with telephone support for chronic back pain Pain 2004 10 111 3 368 77 10.1016/j.pain.2004.07.021 15363881 00006396-200410000-00019 Peters ML Smeets E Feijge M van Breukelen G Andersson G Buhrman M Linton SJ Happy despite pain: a randomized controlled trial of an 8-week Internet-delivered positive psychology intervention for enhancing well-being in patients with chronic pain Clin J Pain 2017 11 33 11 962 75 10.1097/AJP.0000000000000494 28379873 PMC5636048 Chhabra HS Sharma S Verma S Smartphone app in self-management of chronic low back pain: a randomized controlled trial Eur Spine J 2018 11 27 11 2862 74 10.1007/s00586-018-5788-5 30324496 10.1007/s00586-018-5788-5 Irvine AB Russell H Manocchia M Mino DE Cox Glassen T Morgan R Gau JM Birney AJ Ary DV Mobile-Web app to self-manage low back pain: randomized controlled trial J Med Internet Res 2015 01 02 17 1 e1 10.2196/jmir.3130 25565416 v17i1e1 PMC4296097 Calner T Nordin C Eriksson MK Nyberg L Gard G Michaelson P Effects of a self-guided, Web-based activity programme for patients with persistent musculoskeletal pain in primary healthcare: a randomized controlled trial Eur J Pain 2017 07 21 6 1110 20 10.1002/ejp.1012 28464364 Mecklenburg G Smittenaar P Erhart-Hledik JC Perez DA Hunter S Effects of a 12-week digital care program for chronic knee pain on pain, mobility, and surgery risk: randomized controlled trial J Med Internet Res 2018 04 25 20 4 e156 10.2196/jmir.9667 29695370 v20i4e156 PMC5943627 Johnston M Foster M Shennan J Starkey NJ Johnson A The effectiveness of an acceptance and commitment therapy self-help intervention for chronic pain Clin J Pain 2010 06 26 5 393 402 10.1097/AJP.0b013e3181cf59ce 20473046 00002508-201006000-00006 Blixen CE Bramstedt KA Hammel JP Tilley BC A pilot study of health education via a nurse-run telephone self-management programme for elderly people with osteoarthritis J Telemed Telecare 2004 10 1 44 9 10.1258/135763304322764194 15006216 Ang DC Chakr R Mazzuca S France CR Steiner J Stump T Cognitive-behavioral therapy attenuates nociceptive responding in patients with fibromyalgia: a pilot study Arthritis Care Res (Hoboken) 2010 05 62 5 618 23 10.1002/acr.20119 20191481 Shigaki CL Smarr KL Siva C Ge B Musser D Johnson R RAHelp: an online intervention for individuals with rheumatoid arthritis Arthritis Care Res (Hoboken) 2013 10 65 10 1573 81 10.1002/acr.22042 23666599 Petrozzi MJ Leaver A Ferreira PH Rubinstein SM Jones MK Mackey MG Addition of MoodGYM to physical treatments for chronic low back pain: a randomized controlled trial Chiropr Man Therap 2019 10 25 27 54 10.1186/s12998-019-0277-4 31673330 277 PMC6814139 Shebib R Bailey JF Smittenaar P Perez DA Mecklenburg G Hunter S Randomized controlled trial of a 12-week digital care program in improving low back pain NPJ Digit Med 2019 1 7 2 1 10.1038/s41746-018-0076-7 31304351 76 PMC6550254 Toelle TR Utpadel-Fischler DA Haas KK Priebe JA App-based multidisciplinary back pain treatment versus combined physiotherapy plus online education: a randomized controlled trial NPJ Digit Med 2019 5 3 2 34 10.1038/s41746-019-0109-x 31304380 109 PMC6550294 Odole AC Ojo OD Is telephysiotherapy an option for improved quality of life in patients with osteoarthritis of the knee? Int J Telemed Appl 2014 2014 903816 10.1155/2014/903816 24778645 PMC3980780 Krein SL Kadri R Hughes M Kerr EA Piette JD Holleman R Kim HM Richardson CR Pedometer-based Internet-mediated intervention for adults with chronic low back pain: randomized controlled trial J Med Internet Res 2013 08 19 15 8 e181 10.2196/jmir.2605 23969029 v15i8e181 PMC3758050 Chiauzzi E Pujol LA Wood M Bond K Black R Yiu E Zacharoff K painACTION-back pain: a self-management website for people with chronic back pain Pain Med 2010 07 11 7 1044 58 10.1111/j.1526-4637.2010.00879.x 20545873 PME879 Buhrman M Nilsson-Ihrfeldt E Jannert M Ström L Andersson G Guided internet-based cognitive behavioural treatment for chronic back pain reduces pain catastrophizing: a randomized controlled trial J Rehabil Med 2011 05 43 6 500 5 10.2340/16501977-0805 21533329 Bennell KL Nelligan R Dobson F Rini C Keefe F Kasza J French S Bryant C Dalwood A Abbott JH Hinman RS Effectiveness of an Internet-delivered exercise and pain-coping skills training intervention for persons with chronic knee pain: a randomized trial Ann Intern Med 2017 04 04 166 7 453 62 10.7326/M16-1714 28241215 2605025 Leveille SG Huang A Tsai SB Allen M Weingart SN Iezzoni LI Health coaching via an Internet portal for primary care patients with chronic conditions: a randomized controlled trial Med Care 2009 01 47 1 41 7 10.1097/MLR.0b013e3181844dd0 19106729 00005650-200901000-00006 Williams DA Kuper D Segar M Mohan N Sheth M Clauw DJ Internet-enhanced management of fibromyalgia: a randomized controlled trial Pain 2010 12 151 3 694 702 10.1016/j.pain.2010.08.034 20855168 00006396-201012000-00022 PMC2972378 Allen KD Oddone EZ Coffman CJ Datta SK Juntilla KA Lindquist JH Walker TA Weinberger M Bosworth HB Telephone-based self-management of osteoarthritis: a randomized trial Ann Intern Med 2010 11 02 153 9 570 9 10.7326/0003-4819-153-9-201011020-00006 21041576 153/9/570 Ruehlman LS Karoly P Enders C A randomized controlled evaluation of an online chronic pain self management program Pain 2012 02 153 2 319 30 10.1016/j.pain.2011.10.025 22133450 00006396-201202000-00013 PMC3514416 Carpenter KM Stoner SA Mundt JM Stoelb B An online self-help CBT intervention for chronic lower back pain Clin J Pain 2012 01 28 1 14 22 10.1097/AJP.0b013e31822363db 21681084 PMC3184315 Dear BF Titov N Perry KN Johnston L Wootton BM Terides MD Rapee RM Hudson JL The Pain Course: a randomised controlled trial of a clinician-guided Internet-delivered cognitive behaviour therapy program for managing chronic pain and emotional well-being Pain 2013 06 154 6 942 50 10.1016/j.pain.2013.03.005 23688830 S0304-3959(13)00095-X Naylor MR Keefe FJ Brigidi B Naud S Helzer JE Therapeutic Interactive Voice Response for chronic pain reduction and relapse prevention Pain 2008 02 134 3 335 45 10.1016/j.pain.2007.11.001 18178011 00006396-200802000-00012 PMC2693197 Del Pozo-Cruz B Adsuar JC Parraca J Del Pozo-Cruz J Moreno A Gusi N A Web-based intervention to improve and prevent low back pain among office workers: a randomized controlled trial J Orthop Sports Phys Ther 2012 10 42 10 831 41 10.2519/jospt.2012.3980 22951407 2794 Kristjánsdóttir ÓB Fors EA Eide E Finset A Stensrud TL van Dulmen S Wigers SH Eide H A smartphone-based intervention with diaries and therapist feedback to reduce catastrophizing and increase functioning in women with chronic widespread pain. part 2: 11-month follow-up results of a randomized trial J Med Internet Res 2013 03 28 15 3 e72 10.2196/jmir.2442 23538392 v15i3e72 PMC3636011 Janevic MR Shute V Murphy SL Piette JD Acceptability and effects of commercially available activity trackers for chronic pain management among older African American adults Pain Med 2020 02 01 21 2 e68 78 10.1093/pm/pnz215 31509196 5567560 PMC7049259 McCurry SM Von Korff M Morin CM Cunningham A Pike KC Thakral M Wellman R Yeung K Zhu W Vitiello MV Telephone interventions for co-morbid insomnia and osteoarthritis pain: the OsteoArthritis and Therapy for Sleep (OATS) randomized trial design Contemp Clin Trials 2019 12 87 105851 10.1016/j.cct.2019.105851 31614214 S1551-7144(19)30566-X PMC6886712 Hauser-Ulrich S Künzli H Meier-Peterhans D Kowatsch T A smartphone-based health care chatbot to promote self-management of chronic pain (SELMA): pilot randomized controlled trial JMIR Mhealth Uhealth 2020 04 03 8 4 e15806 10.2196/15806 32242820 v8i4e15806 PMC7165314 Kohns DJ Urbanik CP Geisser ME Schubiner H Lumley MA The effects of a pain psychology and neuroscience self-evaluation internet intervention: a randomized controlled trial Clin J Pain 2020 09 36 9 683 92 10.1097/AJP.0000000000000857 32520816 00002508-202009000-00006 Lam J Svensson P Alstergren P Internet-based multimodal pain program with telephone support for adults with chronic temporomandibular disorder pain: randomized controlled pilot trial J Med Internet Res 2020 10 13 22 10 e22326 10.2196/22326 33048053 v22i10e22326 PMC7592067 Amorim AB Pappas E Simic M Ferreira ML Jennings M Tiedemann A Carvalho-E-Silva AP Caputo E Kongsted A Ferreira PH 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 02 11 20 1 71 10.1186/s12891-019-2454-y 30744606 10.1186/s12891-019-2454-y PMC6371593 Skolasky RL Maggard AM Li D Riley 3rd LH Wegener ST Health behavior change counseling in surgery for degenerative lumbar spinal stenosis. Part I: improvement in rehabilitation engagement and functional outcomes Arch Phys Med Rehabil 2015 07 96 7 1200 7 10.1016/j.apmr.2015.03.009 25827657 S0003-9993(15)00257-9 PMC6153438 Sharareh B Schwarzkopf R Effectiveness of telemedical applications in postoperative follow-up after total joint arthroplasty J Arthroplasty 2014 05 29 5 918 22.e1 10.1016/j.arth.2013.09.019 24342278 S0883-5403(13)00700-6 Moffet H Tousignant M Nadeau S Mérette C Boissy P Corriveau H Marquis F Cabana F Ranger P Belzile ÉL Dimentberg R In-home telerehabilitation compared with face-to-face rehabilitation after total knee arthroplasty: a noninferiority randomized controlled trial J Bone Joint Surg Am 2015 07 15 97 14 1129 41 10.2106/JBJS.N.01066 26178888 97/14/1129 Piqueras M Marco E Coll M Escalada F Ballester A Cinca C Belmonte R Muniesa JM Effectiveness of an interactive virtual telerehabilitation system in patients after total knee arthoplasty: a randomized controlled trial J Rehabil Med 2013 04 45 4 392 6 10.2340/16501977-1119 23474735 Bini SA Mahajan J Clinical outcomes of remote asynchronous telerehabilitation are equivalent to traditional therapy following total knee arthroplasty: a randomized control study J Telemed Telecare 2017 02 23 2 239 47 10.1177/1357633X16634518 26940798 1357633X16634518 Russell TG Buttrum P Wootton R Jull GA Internet-based outpatient telerehabilitation for patients following total knee arthroplasty: a randomized controlled trial J Bone Joint Surg Am 2011 01 19 93 2 113 20 10.2106/JBJS.I.01375 21248209 93/2/113 Li LL Gan YY Zhang LN Wang YB Zhang F Qi JM The effect of post-discharge telephone intervention on rehabilitation following total hip replacement surgery Int J Nurs Sci 2014 06 1 2 207 11 10.1016/j.ijnss.2014.05.005 Tousignant M Moffet H Boissy P Corriveau H Cabana F Marquis F A randomized controlled trial of home telerehabilitation for post-knee arthroplasty J Telemed Telecare 2011 17 4 195 8 10.1258/jtt.2010.100602 21398389 jtt.2010.100602 Iles R Taylor NF Davidson M O'Halloran P Telephone coaching can increase activity levels for people with non-chronic low back pain: a randomised trial J Physiother 2011 57 4 231 8 10.1016/S1836-9553(11)70053-4 22093121 S1836-9553(11)70053-4 Priebe JA Haas KK Moreno Sanchez LF Schoefmann K Utpadel-Fischler DA Stockert P Thoma R Schiessl C Kerkemeyer L Amelung V Jedamzik S Reichmann J Marschall U Toelle TR Digital treatment of back pain versus standard of care: the cluster-randomized controlled trial, Rise-uP J Pain Res 2020 7 17 13 1823 38 10.2147/JPR.S260761 32765057 260761 PMC7381830 Kosterink SM Huis in 't Veld RM Cagnie B Hasenbring M Vollenbroek-Hutten MM The clinical effectiveness of a myofeedback-based teletreatment service in patients with non-specific neck and shoulder pain: a randomized controlled trial J Telemed Telecare 2010 16 6 316 21 10.1258/jtt.2010.006005 20798425 16/6/316 del Pozo-Cruz B Gusi N del Pozo-Cruz J Adsuar JC Hernandez-Mocholí M Parraca JA Clinical effects of a nine-month Web-based intervention in subacute non-specific low back pain patients: a randomized controlled trial Clin Rehabil 2013 01 27 1 28 39 10.1177/0269215512444632 22653374 0269215512444632 del Pozo-Cruz B Parraca J del Pozo-Cruz J Adsuar J Hill J Gusi N An occupational, Internet-based intervention to prevent chronicity in subacute lower back pain: a randomised controlled trial J Rehabil Med 2012 06 44 7 581 7 10.2340/16501977-0988 22674240 Cumpston M Li T Page MJ Chandler J Welch VA Higgins JP Thomas J Updated guidance for trusted systematic reviews: a new edition of the Cochrane Handbook for Systematic Reviews of Interventions Cochrane Database Syst Rev 2019 10 03 10 ED000142 10.1002/14651858.ED000142 31643080 Covidence: Accelerate Your Systematic Review Covidence 2017 2021-07-06 https://www.covidence.org/ Dworkin RH Turk DC Farrar JT Haythornthwaite JA Jensen MP Katz NP Kerns RD Stucki G Allen RR Bellamy N Carr DB Chandler J Cowan P Dionne R Galer BS Hertz S Jadad AR Kramer LD Manning DC Martin S McCormick CG McDermott MP McGrath P Quessy S Rappaport BA Robbins W Robinson JP Rothman M Royal MA Simon L Stauffer JW Stein W Tollett J Wernicke J Witter J IMMPACT Core outcome measures for chronic pain clinical trials: IMMPACT recommendations Pain 2005 01 113 1-2 9 19 10.1016/j.pain.2004.09.012 15621359 S0304-3959(04)00440-3 Piot M Chapter 11: Clustering, distance methods and ordination Statistics in Climate Sciences 2014 Bern, Switzerland Departement Mathematik und Statistik, Universität Bern 11-1 12 Hair JF Multivariate Data Analysis: A Global Perspective. 7th edition 2009 Hoboken, NJ, USA Prentice Hall Valentijn PP Ruwaard D Vrijhoef HJ de Bont A Arends RY Bruijnzeels MA Collaboration processes and perceived effectiveness of integrated care projects in primary care: a longitudinal mixed-methods study BMC Health Serv Res 2015 10 09 15 463 10.1186/s12913-015-1125-4 26450573 10.1186/s12913-015-1125-4 PMC4598962 Leznik M Tofallis C Estimating invariant principal components using diagonal regression University of Hertfordshire 2005 2022-05-11 http://uhra.herts.ac.uk/handle/2299/715 Gabriel KR The biplot graphic display of matrices with application to principal component analysis Biometrika 1971 12 58 3 453 67 10.2307/2334381 Kassambara A Practical Guide to Cluster Analysis in R: Unsupervised Machine Learning 2017 Scotts Valley, CA, USA CreateSpace Independent Publishing Platform Higgins JP Thomas J Chandler J Cumpston M Li T Page MJ Welch VA Cochrane Handbook for Systematic Reviews of Interventions 2019 Hoboken, NJ, USA John Wiley & Sons Valentijn PP Pereira FA Ruospo M Palmer SC Hegbrant J Sterner CW Vrijhoef HJ Ruwaard D Strippoli GF Person-centered integrated care for chronic kidney disease: a systematic review and meta-analysis of randomized controlled trials Clin J Am Soc Nephrol 2018 03 07 13 3 375 86 10.2215/CJN.09960917 29438975 CJN.09960917 PMC5967678 Higgins JP Thompson SG Deeks JJ Altman DG Measuring inconsistency in meta-analyses BMJ 2003 09 06 327 7414 557 60 10.1136/bmj.327.7414.557 12958120 327/7414/557 PMC192859 Egger M Davey Smith G Schneider M Minder C Bias in meta-analysis detected by a simple, graphical test BMJ 1997 09 13 315 7109 629 34 10.1136/bmj.315.7109.629 9310563 PMC2127453 R Core Team R: a language and environment for statistical computing R Foundation for Statistical Computing 2013 2021-07-06 http://r.meteo.uni.wroc.pl/web/packages/dplR/vignettes/intro-dplR.pdf Guyatt GH Oxman AD Vist GE Kunz R Falck-Ytter Y Alonso-Coello P Schünemann HJ GRADE Working Group GRADE: an emerging consensus on rating quality of evidence and strength of recommendations BMJ 2008 04 26 336 7650 924 6 10.1136/bmj.39489.470347.AD 18436948 336/7650/924 PMC2335261 Slattery BW Haugh S O'Connor L Francis K Dwyer CP O'Higgins S Egan J McGuire BE An evaluation of the effectiveness of the modalities used to deliver electronic health interventions for chronic pain: systematic review with network meta-analysis J Med Internet Res 2019 07 17 21 7 e11086 10.2196/11086 31317869 v21i7e11086 PMC6668295 Moman RN Dvorkin J Pollard EM Wanderman R Murad MH Warner D Hooten WM A systematic review and meta-analysis of unguided electronic and mobile health technologies for chronic pain-is it time to start prescribing electronic health applications? Pain Med 2019 11 01 20 11 2238 55 10.1093/pm/pnz164 31386151 5544261 de Oliveira Lima L Saragiotto BT Costa LO Nogueira LC Meziat-Filho N Reis FJ Self-guided Web-based pain education for people with musculoskeletal pain: a systematic review and meta-analysis Phys Ther 2021 10 01 101 10 pzab167 10.1093/ptj/pzab167 34174081 6309587 Takeshima N Sozu T Tajika A Ogawa Y Hayasaka Y Furukawa TA Which is more generalizable, powerful and interpretable in meta-analyses, mean difference or standardized mean difference? BMC Med Res Methodol 2014 02 21 14 30 10.1186/1471-2288-14-30 24559167 1471-2288-14-30 PMC3936842 Deeks JJ Higgins JP Altman DG Higgins JP Green S Chapter 9: Analyzing data and undertaking meta-analyses Cochrane Handbook for Systematic Reviews of Interventions. Version 5.1.0 2011 London, UK Cochrane Collaboration Identifying publication bias in meta-analyses of continuous outcomes Cochrane Training 2020 2022-06-24 https://training.cochrane.org/resource/identifying-publication-bias-meta-analyses-continuous-outcomes