This paper is in the following e-collection/theme issue:

Digital Health Reviews (1174) Digital Pain Assessment and Management (145)

Published on in Vol 24, No 9 (2022): September

Preprints (earlier versions) of this paper are available at https://preprints.jmir.org/preprint/37869, first published .
Digital Health Interventions for Musculoskeletal Pain Conditions: Systematic Review and Meta-analysis of Randomized Controlled Trials

Digital Health Interventions for Musculoskeletal Pain Conditions: Systematic Review and Meta-analysis of Randomized Controlled Trials

Digital Health Interventions for Musculoskeletal Pain Conditions: Systematic Review and Meta-analysis of Randomized Controlled Trials

Authors of this article:

Pim Peter Valentijn1, 2 Author Orcid Image ;   Liza Tymchenko1 Author Orcid Image ;   Teddy Jacobson3 Author Orcid Image ;   Jakob Kromann3 Author Orcid Image ;   Claus W Biermann3 Author Orcid Image ;   Mohamed Atef AlMoslemany4 Author Orcid Image ;   Rosa Ymkje Arends1, 5 Author Orcid Image
Article Authors Cited by (13) Tweetations (24) Metrics

    Review

    1Essenburgh Research & Consultancy, Essenburgh Group, Harderwijk, Netherlands

    2Department of Health Services Research, School for Public Health and Primary Care, Faculty of Health, Medicine and Life Sciences, Maastricht University, Maastricht, Netherlands

    3Nordic Netcare, Copenhagen, Denmark

    4Faculty of Medicine, Menoufia University, Menoufia, Egypt

    5University of Applied Sciences Utrecht, Utrecht, Netherlands

    Corresponding Author:

    Pim Peter Valentijn, MSc, PhD

    Essenburgh Research & Consultancy

    Essenburgh Group

    Ceintuurbaan 2-40

    Harderwijk, 3847 LG

    Netherlands

    Phone: 31 0341 217101

    Email: valentijn@essenburgh.com


    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

    J Med Internet Res 2022;24(9):e37869

    doi:10.2196/37869

    Keywords

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


    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.

    ‎
    Figure 1. Rainbow model for digital health interventions [14].
    View this figure

    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.


    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).

    ‎
    Figure 2. Flowchart of search strategy and study selection process.
    View this figure

    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).

    Table 1. Clusters of digital health interventions (N=56).
    Rainbow model intervention characterizationTotal 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 financing5 (9)0 (0)5 (21).006a

    		Data collection, management, and use23 (41)3 (9)20 (83)<.001b
    Professional domain

    		Client identification and registration47 (84)25 (78)22 (92).18

    		Health care provider decision support23 (41)7 (22)16 (67)<.001c

    		Telemedicine55 (98)32 (100)23 (96).25

    		Health care provider communication12 (21)8 (25)4 (17).46

    		Referral coordination4 (7)0 (0)4 (17).02b

    		Health care provider training23 (41)13 (41)10 (42).94
    Client domain

    		Targeted client communication42 (75)26 (81)16 (67).22

    		Personal health tracking38 (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).

    ‎
    Figure 3. 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.
    View this figure

    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).

    ‎
    Figure 4. Effect of digital health on pain. SMD: standardized mean difference.
    View this figure
    Table 2. Summary of findings and assessment of the quality of evidence for outcomes (N=56).
    OutcomesStudies, n (%)Certainty assessmentEffectCertainty


    		Study designRisk of biasInconsistencyIndirectnessImprecisionOther considerationsIndividuals (n=9359), n (%)SMDa rate (95% CI)
    Pain (follow-up: mean 25 weeks)37 (66.1)Randomized trialsSeriousbNot seriousNot seriousNot seriousNone5323 (56.9)0.19 (0.06 to 0.31)Moderate
    Disability and function (follow-up: mean 27 weeks)30 (53.6)Randomized trialsSeriousbNot seriousNot seriousNot seriousNone4849 (51.8)0.14 (0.03 to 0.25)Moderate
    Quality of life (follow-up: mean 25 weeks)24 (42.9)Randomized trialsNot seriousNot seriousNot seriousNot seriousNone3995 (42.5)0.22 (0.07 to 0.36)High
    Emotional functioning (follow-up: mean 29 weeks)24 (42.9)Randomized trialsSeriousbSeriouscNot seriousNot seriousNone3814 (40.8)0.24 (0.12 to 0.35)Low
    Self-management (follow-up: mean 26 weeks)21 (37.5)Randomized trialsSeriousbNot seriousNot seriousNot seriousNone2857 (30.5)0.14 (0.05 to 0.24)Moderate
    Global improvement (follow-up: mean 42 weeks)4 (7.1)Randomized trialsSeriousbNot seriousNot seriousSeriousdNone795 (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.

    ‎
    Figure 5. Effect of digital health on disability and function. SMD: standardized mean difference.
    View this figure
    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.

    ‎
    Figure 6. Effect of digital health on quality of life. SMD: standardized mean difference.
    View this figure
    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.

    ‎
    Figure 7. Effect of digital health on emotional functioning. SMD: standardized mean difference.
    View this figure
    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).

    ‎
    Figure 8. Effect of digital health on self-management. SMD: standardized mean difference.
    View this figure

    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).


    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.

    Acknowledgments

    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.

    Authors' Contributions

    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.

    Conflicts of Interest

    None declared.

    ‎
    Multimedia Appendix 1

    Supplementary tables and figures of search strategy and study results.

    DOCX File , 425 KB
    1. 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 Jun;78(6):844-848. [CrossRef] [Medline]
    2. Whiteford HA, Degenhardt L, Rehm J, Baxter AJ, Ferrari AJ, Erskine HE, et al. Global burden of disease attributable to mental and substance use disorders: findings from the Global Burden of Disease Study 2010. Lancet 2013 Nov 09;382(9904):1575-1586. [CrossRef] [Medline]
    3. 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 Sep;38(4):401-405. [CrossRef] [Medline]
    4. Dario AB, Moreti Cabral A, Almeida L, Ferreira ML, Refshauge K, Simic M, et al. Effectiveness of telehealth-based interventions in the management of non-specific low back pain: a systematic review with meta-analysis. Spine J 2017 Sep;17(9):1342-1351. [CrossRef] [Medline]
    5. 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 Jun 05;22(6):e15617 [FREE Full text] [CrossRef] [Medline]
    6. 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 May;24(4):257-262. [CrossRef] [Medline]
    7. 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 May;31(5):625-638. [CrossRef] [Medline]
    8. Slattery BW, Haugh S, Francis K, O'Connor L, Barrett K, Dwyer CP, et al. Protocol for a systematic review with network meta-analysis of the modalities used to deliver eHealth interventions for chronic pain. Syst Rev 2017 Mar 03;6(1):45 [FREE Full text] [CrossRef] [Medline]
    9. 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 Jun;31(6):470-492. [CrossRef] [Medline]
    10. 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 Mar;19(3):709-720. [CrossRef] [Medline]
    11. Recommendations on digital interventions for health system strengthening. World Health Organization. Geneva, Switzerland: World Health Organization; 2019 Jun 6.   URL: https://www.who.int/publications/i/item/9789241550505 [accessed 2021-07-06]
    12. 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 Dec 21;19(1):984 [FREE Full text] [CrossRef] [Medline]
    13. 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 Mar 22;13:e010 [FREE Full text] [CrossRef] [Medline]
    14. Rainbow Model for Integrated Care. Essenburgh Group. 2021.   URL: https://www.essenburgh.com/en/rainbow-model-of-integrated-care [accessed 2021-07-06]
    15. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JP, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. J Clin Epidemiol 2009 Oct;62(10):e1-34 [FREE Full text] [CrossRef] [Medline]
    16. Paganini S, Lin J, Kählke F, Buntrock C, Leiding D, Ebert DD, et al. A guided and unguided Internet- and mobile-based intervention for chronic pain: health economic evaluation alongside a randomised controlled trial. BMJ Open 2019 Apr 09;9(4):e023390 [FREE Full text] [CrossRef] [Medline]
    17. Fatoye F, Gebrye T, Fatoye C, Mbada CE, Olaoye MI, Odole AC, et al. The clinical and cost-effectiveness of telerehabilitation for people with nonspecific chronic low back pain: randomized controlled trial. JMIR Mhealth Uhealth 2020 Jun 24;8(6):e15375 [FREE Full text] [CrossRef] [Medline]
    18. Brattberg G. Self-administered EFT (Emotional Freedom Techniques) in individuals with fibromyalgia: a randomized trial. Integr Med 2008;7(4):30-35.
    19. Allen KD, Arbeeva L, Callahan LF, Golightly YM, Goode AP, Heiderscheit BC, et al. Physical therapy vs internet-based exercise training for patients with knee osteoarthritis: results of a randomized controlled trial. Osteoarthritis Cartilage 2018 Mar;26(3):383-396 [FREE Full text] [CrossRef] [Medline]
    20. 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 Apr 08;162(7):792-796. [CrossRef] [Medline]
    21. Buhrman M, Fredriksson A, Edström G, Shafiei D, Tärnqvist C, Ljótsson B, et al. Guided Internet-delivered cognitive behavioural therapy for chronic pain patients who have residual symptoms after rehabilitation treatment: randomized controlled trial. Eur J Pain 2013 May;17(5):753-765. [CrossRef] [Medline]
    22. 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 Mar;24(1):94-100. [CrossRef] [Medline]
    23. Buhrman M, Skoglund A, Husell J, Bergström K, Gordh T, Hursti T, et al. Guided internet-delivered acceptance and commitment therapy for chronic pain patients: a randomized controlled trial. Behav Res Ther 2013 Jun;51(6):307-315. [CrossRef] [Medline]
    24. Bennell KL, Nelligan RK, Rini C, Keefe FJ, Kasza J, French S, et al. Effects of internet-based pain coping skills training before home exercise for individuals with hip osteoarthritis (HOPE trial): a randomised controlled trial. Pain 2018 Sep;159(9):1833-1842. [CrossRef] [Medline]
    25. 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 Jan;10(1):68-79. [CrossRef] [Medline]
    26. Buhrman M, Fältenhag S, Ström L, Andersson G. Controlled trial of Internet-based treatment with telephone support for chronic back pain. Pain 2004 Oct;111(3):368-377. [CrossRef] [Medline]
    27. Peters ML, Smeets E, Feijge M, van Breukelen G, Andersson G, Buhrman M, et al. 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 Nov;33(11):962-975 [FREE Full text] [CrossRef] [Medline]
    28. Chhabra HS, Sharma S, Verma S. Smartphone app in self-management of chronic low back pain: a randomized controlled trial. Eur Spine J 2018 Nov;27(11):2862-2874. [CrossRef] [Medline]
    29. Irvine AB, Russell H, Manocchia M, Mino DE, Cox Glassen T, Morgan R, et al. Mobile-Web app to self-manage low back pain: randomized controlled trial. J Med Internet Res 2015 Jan 02;17(1):e1 [FREE Full text] [CrossRef] [Medline]
    30. 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 Jul;21(6):1110-1120. [CrossRef] [Medline]
    31. 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 Apr 25;20(4):e156 [FREE Full text] [CrossRef] [Medline]
    32. 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 Jun;26(5):393-402. [CrossRef] [Medline]
    33. 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):44-49. [CrossRef] [Medline]
    34. 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 May;62(5):618-623 [FREE Full text] [CrossRef] [Medline]
    35. 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 Oct;65(10):1573-1581 [FREE Full text] [CrossRef] [Medline]
    36. 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 Oct 25;27:54 [FREE Full text] [CrossRef] [Medline]
    37. 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 Jan 7;2:1 [FREE Full text] [CrossRef] [Medline]
    38. 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 May 3;2:34 [FREE Full text] [CrossRef] [Medline]
    39. 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 [FREE Full text] [CrossRef] [Medline]
    40. Krein SL, Kadri R, Hughes M, Kerr EA, Piette JD, Holleman R, et al. Pedometer-based Internet-mediated intervention for adults with chronic low back pain: randomized controlled trial. J Med Internet Res 2013 Aug 19;15(8):e181 [FREE Full text] [CrossRef] [Medline]
    41. Chiauzzi E, Pujol LA, Wood M, Bond K, Black R, Yiu E, et al. painACTION-back pain: a self-management website for people with chronic back pain. Pain Med 2010 Jul;11(7):1044-1058. [CrossRef] [Medline]
    42. 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 May;43(6):500-505 [FREE Full text] [CrossRef] [Medline]
    43. Bennell KL, Nelligan R, Dobson F, Rini C, Keefe F, Kasza J, et al. 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 Apr 04;166(7):453-462. [CrossRef] [Medline]
    44. 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 Jan;47(1):41-47. [CrossRef] [Medline]
    45. Williams DA, Kuper D, Segar M, Mohan N, Sheth M, Clauw DJ. Internet-enhanced management of fibromyalgia: a randomized controlled trial. Pain 2010 Dec;151(3):694-702 [FREE Full text] [CrossRef] [Medline]
    46. Allen KD, Oddone EZ, Coffman CJ, Datta SK, Juntilla KA, Lindquist JH, et al. Telephone-based self-management of osteoarthritis: a randomized trial. Ann Intern Med 2010 Nov 02;153(9):570-579 [FREE Full text] [CrossRef] [Medline]
    47. Ruehlman LS, Karoly P, Enders C. A randomized controlled evaluation of an online chronic pain self management program. Pain 2012 Feb;153(2):319-330 [FREE Full text] [CrossRef] [Medline]
    48. Carpenter KM, Stoner SA, Mundt JM, Stoelb B. An online self-help CBT intervention for chronic lower back pain. Clin J Pain 2012 Jan;28(1):14-22 [FREE Full text] [CrossRef] [Medline]
    49. Dear BF, Titov N, Perry KN, Johnston L, Wootton BM, Terides MD, et al. 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 Jun;154(6):942-950. [CrossRef] [Medline]
    50. Naylor MR, Keefe FJ, Brigidi B, Naud S, Helzer JE. Therapeutic Interactive Voice Response for chronic pain reduction and relapse prevention. Pain 2008 Feb;134(3):335-345 [FREE Full text] [CrossRef] [Medline]
    51. 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 Oct;42(10):831-841. [CrossRef] [Medline]
    52. Kristjánsdóttir Ó, Fors EA, Eide E, Finset A, Stensrud TL, van Dulmen S, et al. 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 Mar 28;15(3):e72 [FREE Full text] [CrossRef] [Medline]
    53. 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 Feb 01;21(2):e68-e78 [FREE Full text] [CrossRef] [Medline]
    54. McCurry SM, Von Korff M, Morin CM, Cunningham A, Pike KC, Thakral M, et al. Telephone interventions for co-morbid insomnia and osteoarthritis pain: the OsteoArthritis and Therapy for Sleep (OATS) randomized trial design. Contemp Clin Trials 2019 Dec;87:105851 [FREE Full text] [CrossRef] [Medline]
    55. 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 Apr 03;8(4):e15806 [FREE Full text] [CrossRef] [Medline]
    56. 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 Sep;36(9):683-692. [CrossRef] [Medline]
    57. 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 Oct 13;22(10):e22326 [FREE Full text] [CrossRef] [Medline]
    58. Amorim AB, Pappas E, Simic M, Ferreira ML, Jennings M, Tiedemann A, 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 Feb 11;20(1):71 [FREE Full text] [CrossRef] [Medline]
    59. 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 Jul;96(7):1200-1207 [FREE Full text] [CrossRef] [Medline]
    60. Sharareh B, Schwarzkopf R. Effectiveness of telemedical applications in postoperative follow-up after total joint arthroplasty. J Arthroplasty 2014 May;29(5):918-22.e1. [CrossRef] [Medline]
    61. Moffet H, Tousignant M, Nadeau S, Mérette C, Boissy P, Corriveau H, et al. In-home telerehabilitation compared with face-to-face rehabilitation after total knee arthroplasty: a noninferiority randomized controlled trial. J Bone Joint Surg Am 2015 Jul 15;97(14):1129-1141. [CrossRef] [Medline]
    62. Piqueras M, Marco E, Coll M, Escalada F, Ballester A, Cinca C, et al. Effectiveness of an interactive virtual telerehabilitation system in patients after total knee arthoplasty: a randomized controlled trial. J Rehabil Med 2013 Apr;45(4):392-396 [FREE Full text] [CrossRef] [Medline]
    63. 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 Feb;23(2):239-247. [CrossRef] [Medline]
    64. 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 Jan 19;93(2):113-120. [CrossRef] [Medline]
    65. 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 Jun;1(2):207-211. [CrossRef]
    66. 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):195-198. [CrossRef] [Medline]
    67. 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):231-238 [FREE Full text] [CrossRef] [Medline]
    68. Priebe JA, Haas KK, Moreno Sanchez LF, Schoefmann K, Utpadel-Fischler DA, Stockert P, et al. Digital treatment of back pain versus standard of care: the cluster-randomized controlled trial, Rise-uP. J Pain Res 2020 Jul 17;13:1823-1838 [FREE Full text] [CrossRef] [Medline]
    69. 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):316-321. [CrossRef] [Medline]
    70. 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 Jan;27(1):28-39. [CrossRef] [Medline]
    71. 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 Jun;44(7):581-587 [FREE Full text] [CrossRef] [Medline]
    72. Cumpston M, Li T, Page MJ, Chandler J, Welch VA, Higgins JP, et al. Updated guidance for trusted systematic reviews: a new edition of the Cochrane Handbook for Systematic Reviews of Interventions. Cochrane Database Syst Rev 2019 Oct 03;10:ED000142. [CrossRef] [Medline]
    73. Covidence: Accelerate Your Systematic Review. Covidence. 2017.   URL: https://www.covidence.org/ [accessed 2021-07-06]
    74. Dworkin RH, Turk DC, Farrar JT, Haythornthwaite JA, Jensen MP, Katz NP, IMMPACT. Core outcome measures for chronic pain clinical trials: IMMPACT recommendations. Pain 2005 Jan;113(1-2):9-19. [CrossRef] [Medline]
    75. Piot M. Chapter 11: Clustering, distance methods and ordination. In: Statistics in Climate Sciences. Bern, Switzerland: Departement Mathematik und Statistik, Universität Bern; 2014:11-1-1112.
    76. Hair JF. Multivariate Data Analysis: A Global Perspective. 7th edition. Hoboken, NJ, USA: Prentice Hall; 2009.
    77. 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 Oct 09;15:463 [FREE Full text] [CrossRef] [Medline]
    78. Leznik M, Tofallis C. Estimating invariant principal components using diagonal regression. University of Hertfordshire. 2005.   URL: http://uhra.herts.ac.uk/handle/2299/715 [accessed 2022-05-11]
    79. Gabriel KR. The biplot graphic display of matrices with application to principal component analysis. Biometrika 1971 Dec;58(3):453-467. [CrossRef]
    80. Kassambara A. Practical Guide to Cluster Analysis in R: Unsupervised Machine Learning. Scotts Valley, CA, USA: CreateSpace Independent Publishing Platform; 2017.
    81. Higgins JP, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, et al. Cochrane Handbook for Systematic Reviews of Interventions. Hoboken, NJ, USA: John Wiley & Sons; 2019.
    82. Valentijn PP, Pereira FA, Ruospo M, Palmer SC, Hegbrant J, Sterner CW, et al. Person-centered integrated care for chronic kidney disease: a systematic review and meta-analysis of randomized controlled trials. Clin J Am Soc Nephrol 2018 Mar 07;13(3):375-386 [FREE Full text] [CrossRef] [Medline]
    83. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ 2003 Sep 06;327(7414):557-560 [FREE Full text] [CrossRef] [Medline]
    84. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997 Sep 13;315(7109):629-634 [FREE Full text] [CrossRef] [Medline]
    85. R Core Team. R: a language and environment for statistical computing. R Foundation for Statistical Computing. 2013.   URL: http://r.meteo.uni.wroc.pl/web/packages/dplR/vignettes/intro-dplR.pdf [accessed 2021-07-06]
    86. Guyatt GH, Oxman AD, Vist GE, Kunz R, Falck-Ytter Y, Alonso-Coello P, GRADE Working Group. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. BMJ 2008 Apr 26;336(7650):924-926 [FREE Full text] [CrossRef] [Medline]
    87. Slattery BW, Haugh S, O'Connor L, Francis K, Dwyer CP, O'Higgins S, et al. 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 Jul 17;21(7):e11086 [FREE Full text] [CrossRef] [Medline]
    88. Moman RN, Dvorkin J, Pollard EM, Wanderman R, Murad MH, Warner D, et al. 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 Nov 01;20(11):2238-2255. [CrossRef] [Medline]
    89. 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 Oct 01;101(10):pzab167. [CrossRef] [Medline]
    90. 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 Feb 21;14:30 [FREE Full text] [CrossRef] [Medline]
    91. Deeks JJ, Higgins JP, Altman DG. Chapter 9: Analyzing data and undertaking meta-analyses. In: Higgins JP, Green S, editors. Cochrane Handbook for Systematic Reviews of Interventions. Version 5.1.0. London, UK: Cochrane Collaboration; 2011.
    92. Identifying publication bias in meta-analyses of continuous outcomes. Cochrane Training. 2020.   URL: https:/​/training.​cochrane.org/​resource/​identifying-publication-bias-meta-analyses-continuous-outcomes [accessed 2022-06-24]

    ‎
    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

    Edited by R Kukafka; submitted 10.03.22; peer-reviewed by A Rovetta; comments to author 14.04.22; revised version received 17.05.22; accepted 25.07.22; published 06.09.22

    Copyright

    ©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.

    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.