Effectiveness of Telematic Behavioral Techniques to Manage Anxiety, Stress and Depressive Symptoms in Patients with Chronic Musculoskeletal Pain: A Systematic Review and Meta-Analysis

Anxiety, depressive symptoms and stress have a significant influence on chronic musculoskeletal pain. Behavioral modification techniques have proven to be effective to manage these variables; however, the COVID-19 pandemic has highlighted the need for an alternative to face-to-face treatment. We conducted a search of PubMed, the Cumulative Index to Nursing and Allied Health Literature (CINAHL), Web of Science, APA PsychInfo, and Psychological and Behavioural Collections. The aim was to assess the effectiveness of telematic behavioral modification techniques (e-BMT) on psychological variables in patients with chronic musculoskeletal pain through a systematic review with meta-analysis. We used a conventional pairwise meta-analysis and a random-effects model. We calculated the standardized mean difference (SMD) with the corresponding 95% confidence interval (CI). Forty-one randomized controlled trials were included, with a total of 5018 participants. We found a statistically significant small effect size in favor of e-BMT in depressive symptoms (n = 3531; SMD = −0.35; 95% CI −0.46, −0.24) and anxiety (n = 2578; SMD = −0.32; 95% CI −0.42, −0.21) with low to moderate strength of evidence. However, there was no statistically significant effect on stress symptoms with moderate strength of evidence. In conclusion, e-BMT is an effective option for the management of anxiety and depressive symptoms in patients with chronic musculoskeletal pain. However, it does not seem effective to improve stress symptoms.


Introduction
The COVID-19 pandemic has shaken our lives and jeopardized the treatment of countless patients with chronic pain [1,2]. Chronic pain patients have shown a significant increase in their perceived pain in comparison with the pre-pandemic period [3], as well as an increase in depressive symptoms, anxiety, loneliness, tiredness and catastrophizing [3]. Nearly half of a sample of 2423 chronic pain patients had moderate to severe psychological distress [4]. The worsening of mental health in patients with chronic pain is not without consequences; these variables have been linked to higher pain catastrophizing, pain-related fear and avoidance, and a higher risk of misuse of opioids [5,6].
These patients need follow-up, a close relationship with health professionals and appropriate treatment, but social distancing prevents them from doing so [1]. Chronic pain patients had higher self-isolation than participants without pain during the pandemic [3]. Because it does not require being physically present, telerehabilitation, or the therapeutic use of technological devices, has been recommended for chronic pain management worldwide [2]. Over the last few decades, behavioral modification techniques (BMT) have showed to be effective in the management of psychological variables in chronic pain patients [7,8]. However, it is not clear if telematic BMT (e-BMT) is also effective to improve psychological variables and if it is as effective as in-person BMT. Some previous systematic reviews have assessed the effect of telerehabilitation based on BMT on variables such as pain intensity, disability, disease impact, physical function, pain-related fear of movement, and psychological distress [9][10][11][12], showing promising results.
The primary aim of this systematic review with meta-analysis was to evaluate the effectiveness of e-BMT compared with usual care/waiting list or in-person BMT in psychological variables. Secondly, we aimed to sub-analyze the results by intervention parameters and diagnostic conditions. The main reason for the secondary aim was because the "BMT" label includes a large range of interventions and so we can isolate effects by intervention or by clinical entities.

Materials and Methods
This systematic review and meta-analysis was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) 2020 statement [13]. This systematic review was registered prospectively in an international database (PROS-PERO), where it can be accessed (CRD42021278086).

Search Strategy
The search strategy of this systematic review is the same as another systematic review from our research group on this topic, which is currently under review. The search for studies was performed using Medline (PubMed), the Cumulative Index to Nursing and Allied Health Literature (CINAHL), Web of Science, APA PsychInfo, and Psychological and Behavioural Collections, from inception to (30) August 2021. In addition, we manually checked the references of the studies included in the review and checked the studies included in systematic reviews related to this topic. The search was also adapted and performed in Google Scholar due to its capacity to search for relevant articles and grey literature [14]. No restrictions were applied to any specific language. The different search strategies used are detailed in Appendix A.1.
Two independent reviewers (CVR and FCM) conducted the search using the same methodology, and the differences were resolved by consensus moderated by a third reviewer (JCG). We used Rayyan software to organize studies, assess studies for eligibility and remove duplicates [15].

Study Eligibility Criteria
The selection criteria used in this systematic review and meta-analysis were based a Population, Intervention, Control, Outcomes, and Study design framework (PICOS). We included randomized controlled trials that have applied e-BMT through a technology device (Website, online, telephone or mobile application). The intervention could be applied alone or embedded with another treatment, only if the control group contains only the additional treatment. Control group could be usual care, waiting list, no intervention, or in-person equivalent BMT. The participants selected for the studies were patients older than 18 years with any kind of chronic musculoskeletal disorder. The participants' gender was irrelevant. We excluded patients with musculoskeletal pain due to oncologic or traumatic process. The measures used to assess the results were depressive symptoms, anxiety, and stress. Time of measurement was restrained to post-treatment results.

Selection Process and Data Extraction
The two phases of studies selection (title/abstract screening and full-text evaluation) were realized by two independent reviewers (CVR and FCM). First, they assessed the relevance of the studies regarding the study questions and aims, based on information from the title, abstract, and keywords of each study. If there was no consensus or the abstracts did not contain sufficient information, the full text was reviewed. In the second phase of the analysis, the full text was used to assess whether the studies met all the inclusion criteria. Differences between the two independent reviewers were resolved by a consensus process moderated by a third reviewer (JCG). Data described in the results were extracted by means of a structured protocol that ensured that the most relevant information was obtained from each study [16].

Risk of Bias and Methodological Quality Assessment
The Risk Of Bias 2 (RoB 2) tool was used to assess randomized trials [17]. It covers a total of 5 domains: (1) Bias arising from the randomization process, (2) Bias due to deviations from the intended interventions, (3) Bias due to missing outcome data, (4) Bias in measurement of the outcome, (5) Bias in selection of the reported result. The study will be categorized as having (a) low risk of bias if all domains shown low risk of bias, (b) some concerns if one domain is rated with some concerns without any with high risk of bias, and (c) high risk of bias, if one domain is rated as having high risk of bias or multiple with some concerns.
The studies' methodological quality was assessed using the PEDro scale [18], which assesses the internal and external validity of a study and consists of 11 criteria. The methodological criteria were scored as follows: yes (1 point), no (0 points), or do not know (0 points). The PEDro score for each selected study provided an indicator of the methodological quality (9-10 = excellent; 6-8 = good; 4-5 = fair; 3-0 = poor) [19]. We used the data obtained from the PEDro scale to map the results of the quantitative analyses.
Two independent reviewers (LSM and FCM) examined the quality and the risk of bias of all the selected studies using the same methodology. Disagreements between the reviewers were resolved by consensus with a third reviewer (JCG). Concordance between the results (inter-rater reliability) was measured using Cohen's kappa coefficient (κ) as follows: (1) κ > 0.7 indicated a high level of agreement between assessors; (2) κ = 0.5-0.7 indicated a moderate level of agreement; and (3) κ < 0.5 indicated a low level of agreement [20].

Quality of Evidence
The quality of evidence analysis was based on classifying the results into levels of evidence according to the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) framework, which is based on 5 domains: study design, imprecision, indirectness, inconsistency, and publication bias [21]. The assessment of the 5 domains was conducted according to GRADE criteria [22,23]. Evidence was categorized into the following 4 levels accordingly: (a) High quality. Further research is very unlikely to change our confidence in the effect estimate. All 5 domains are also met. (b) Moderate quality. Further research is likely to have an important impact on our confidence in the effect estimate and might change the effect estimate. One of the 5 domains is not met. (c) Low quality. Further research is very likely to have a significant impact on our confidence in the effect estimate and is likely to change the estimate. Two of the 5 domains are not met. (d) Very low quality. Any effect estimates highly uncertain. Three of the 5 domains are not met [22,23].
For the risk of bias domain, the recommendations were downgraded one level in the event there was an uncertain or high risk of bias and serious limitations in the effect estimate (more that 25% of the participants were from studies with high risk of bias, as measured by the RoB 2 scale). In terms of inconsistency, the recommendations were downgraded one level when the point estimates varied widely among studies, the confidence intervals showed minimal overlap or when the I 2 was substantial or large (greater than 50%). For the indirectness domain, recommendations were downgraded when severe differences in interventions, study populations or outcomes were found. (The recommendations were downgraded in the absence of direct comparisons between the interventions of interest or when there are no key outcomes, and the recommendation is based only on intermediate outcomes or if more than 50% of the participants were outside the target group.) For the imprecision domain, the recommendations were downgraded one level if there were fewer than 300 participants for the continuous data. Finally, the recommendations were downgraded due to strong influence of publication bias if the results changed significantly after adjusting for publication bias.

Data Synthesis
The statistical analysis was conducted using RStudio software version 1.4.1717, which is based on R software version 4.1.1 [24,25]. To compare the outcomes reported by the studies, we calculated the standardized mean difference (SMD), as Hedge's g, over time and the corresponding 95% confidence interval (CI) for the continuous variables. It was interpreted as described by Hopkins et al. [26]. If necessary, CI and standard error (SE) were converted into standard deviation (SD) [27]. The estimated SMDs were interpreted as described by Hopkins et al. [26]; that is, we considered an SMD of 4.0 to represent an extremely large clinical effect, 2.0-4.0 represented a very large effect, 1.2-2.0 represented a large effect, 0.6-1.2 represented a moderate effect, 0.2-0.6 represented a small effect, and 0.0-0.2 represented a trivial effect.
We used the same inclusion criteria for the systematic review and the meta-analysis and included 3 additional criteria: (1) In the results, there was detailed information regarding the comparative statistical data of the exposure factors, therapeutic interventions, and treatment responses; (2) the intervention was compared with a similar control group; and (3) data on the analyzed variables were represented in at least 3 studies.
As we pooled different treatments, we could not assume that there was a unique true effect. So, we anticipated between-study heterogeneity and used a random-effects model to pool effect sizes. In order the calculate the heterogeneity variance τ 2 , we used the Restricted Maximum Likelihood Estimator as recommended for continuous outcomes [28,29]. To calculate the confidence interval around the pooled effect, we used Knapp-Hartung adjustments [30,31].
We estimated the degree of heterogeneity among the studies using Cochran's Q statistic test (a p-value < 0.05 was considered significant), the inconsistency index (I 2 ) and the prediction interval (PI) based on the between-study variance τ 2 [26]. Cochran's Q test allows us to assess the presence of between-study heterogeneity [32]. Despite its common use to assess heterogeneity, the I 2 index only represents the percentage of variability in the effect sizes not caused by a sampling error [33]. Therefore, as recommended, we additionally report PIs. The PIs are an equivalent to standard deviation and represent a range within which the effects of future studies are expected to fall based on current data [33,34].
To detect the presence of outliers that could potentially influence the estimated pooled effect and assess the robustness of our results, we applied an influence analysis based on the leave-one-out method [35]. If the study's results had an important influence on the pooled effect, we conducted a sensitivity analysis removing it or them. We additionally ran a drapery plot, which is based on p-value functions and gives us the p-value curve for the pooled estimate for all possible alpha levels [36].
To detect publication bias, we performed a visual evaluation of the Doi plot and the funnel plot [37], seeking asymmetry. We also performed a quantitative measure of the Luis Furuya Kanamori (LFK) index, which has been shown to be more sensitive than the Egger test in detecting publication bias in a meta-analysis of a low number of studies [38]. An LFK index within ±1 represents no asymmetry, exceeding ±1 but within ±2 represents minor asymmetry, and exceeding ±2 involves major asymmetry. If there was significant asymmetry, we applied a small-study effect method to correct for publication bias using the Duval and Tweedie trim and fill method [39].
For the qualitative analysis, we reported the between-group mean difference (MD) with the 95% CI for the outcomes of interest. If it was not reported by the authors, we calculated it [40].
For the qualitative analysis, we reported the between-group mean difference (MD) with the 95% CI for the outcomes of interest. If it was not reported by the authors, we calculated it [40].

Depressive Symptoms
According to the influence analyses, we conducted a sensitivity analysis without Dear et al. [43]. We found a statistically significant small effect size (32 RCTs; n = 3531; SMD = −0.35; 95% CI −0.46, −0.24) of e-BMT on depressive symptoms compared with usual care or waiting list, with significant heterogeneity (Q = 74.06 (p < 0.01); I 2 = 57% (36%, 71%); PI −0.82, 0.12) and a low strength of evidence ( Figure 2). Since PI crosses zero, we cannot be confident that future studies will not find contradictory results; however, the results appear to be robust to different p-value functions. With respect to the presence of publication bias, the funnel and Doi plots show an asymmetrical pattern, demonstrating minor asymmetry (LFK index = −1.62). When the sensitivity analysis is adjusted for publication bias, there is still a small significant effect. Statistical analyses are detailed in Appendix A.7. Subgroup analyses are detailed in Table 1a.

Depressive Symptoms
According to the influence analyses, we conducted a sensitivity analysis without Dear et al. [43]. We found a statistically significant small effect size (32 RCTs; n = 3531; SMD = −0.35; 95% CI −0.46, −0.24) of e-BMT on depressive symptoms compared with usual care or waiting list, with significant heterogeneity (Q = 74.06 (p < 0.01); I 2 = 57% (36%, 71%); PI −0.82, 0.12) and a low strength of evidence ( Figure 2). Since PI crosses zero, we cannot be confident that future studies will not find contradictory results; however, the results appear to be robust to different p-value functions. With respect to the presence of publication bias, the funnel and Doi plots show an asymmetrical pattern, demonstrating minor asymmetry (LFK index = −1.62). When the sensitivity analysis is adjusted for publication bias, there is still a small significant effect. Statistical analyses are detailed in Appendix A7. Subgroup analyses are detailed in Table 1a.

Anxiety
According to the influence analyses, we conducted a sensitivity analysis without Trudeau et al. [62]. We found a statistically significant small effect size (21 RCTs; n = 2578; SMD = −0.32; 95% CI −0.42, −0.21) of e-BMT on anxiety compared with usual care or waiting list, with significant heterogeneity (Q = 33.47 (p = 0.04); I 2 = 37% (0%, 63%); PI −0.64, 0.00) and a moderate strength of evidence ( Figure 3). Since PI crosses zero, we cannot be confident that future studies will not find contradictory results; however, the results appear to be robust to different p-value functions. With respect to the presence of publication bias, the funnel and Doi plots show a symmetrical pattern, demonstrating no asymmetry (LFK index = −0.48). Statistical analyses are detailed in Appendix A.8. Subgroup analyses are detailed in Table 1b.

Stress
We found no statistically significant effect size (4 RCTs; n = 789; SMD = −0.13; 95% CI −0.28, 0.02) of e-BMT on stress compared with usual care or waiting list, with significant heterogeneity (Q = 1.33 (p = 0.72); I 2 = 0% (0%, 85%); PI −0.34, 0.07) and a moderate strength of evidence ( Figure 4). Since PI crosses zero, we cannot be confident that future studies will not find contradictory results. With respect to the presence of publication bias, the

Stress
We found no statistically significant effect size (4 RCTs; n = 789; SMD = −0.13; 95% CI −0.28, 0.02) of e-BMT on stress compared with usual care or waiting list, with significant heterogeneity (Q = 1.33 (p = 0.72); I 2 = 0% (0%, 85%); PI −0.34, 0.07) and a moderate strength of evidence (Figure 4). Since PI crosses zero, we cannot be confident that future studies will not find contradictory results. With respect to the presence of publication bias, the funnel and Doi plots show an asymmetrical pattern, demonstrating minor asymmetry (LFK index = −1.55). When the sensitivity analysis is adjusted for publication bias, there is no influence on the estimated effect. Statistical analyses are detailed in Appendix A.9.
GRADE's overall strength of the evidence is detailed in

Discussion
The primary aim of this systematic review with meta-analysis was to evalu effectiveness of e-BMT compared with usual care/waiting list or in-person BMT in of psychological variables. Secondly, we aimed to sub-analyze the results by interv parameters and diagnostic conditions. The main results found that e-BMT seems t effective option for the management of anxiety and depressive symptoms in patien musculoskeletal conditions causing chronic pain but not to improve stress sympto

Discussion
The primary aim of this systematic review with meta-analysis was to evaluate the effectiveness of e-BMT compared with usual care/waiting list or in-person BMT in terms of psychological variables. Secondly, we aimed to sub-analyze the results by intervention parameters and diagnostic conditions. The main results found that e-BMT seems to be an effective option for the management of anxiety and depressive symptoms in patients with musculoskeletal conditions causing chronic pain but not to improve stress symptoms. e-BMT does not seem to provide greater improvement than in-person BMT for psychological variables.
Several research studies have been published and have shown similar results to those found in this review with meta-analysis with regard to depressive and anxiety symptoms. For example, the rapid review conducted by Varker et al. [83] aimed to evaluate the effectiveness of e-BMT (by videoconference) and also through conventional mobile phone calls for people with high levels of anxiety and depression. The main results showed that both rehabilitation modalities produced significant positive results in terms of decreasing the levels of both psychological variables. In addition to this, the review conducted by McCall et al. [84] found that delivering psychological telematic interventions resulted in a significant decrease in depressive symptoms but could not be proven to be effective in comparison to face-to-face psychological intervention. Anxiety symptoms could not be assessed. This work included few studies, so the results have to be interpreted with caution.
In addition to being a possible alternative to in-person treatment, e-BMT appears to be a cost-effective technique compared to in-person BMT. De Boer et al. compared e-BMT and in-person BMT in patients with chronic pain and found that the costs of online CBT were EUR 199 lower than in-person BMT [85]. Similarly, Aspvall et al. found that after 6 months of follow-up in children and adolescents with obsessive compulsive disorder, there was a difference of USD 1688 in favor of e-BMT [86]. Healthcare systems and guidelines should seriously consider implementing e-BMT in the management of patients with musculoskeletal disorders causing chronic pain.

Practical Implication
Concerning clinical implications, the results showed good results in favor of e-BMT. This gives us an effective treatment window in the COVID-19 era, so we are going to have a greater impact on patients with persistent pain. In addition, there is a decentralization of interventions, which may have some positive effects such as improving and increasing adherence to treatments due to easier accessibility, as well as lowering barriers to access or facilitating follow-up. Future studies should also focus on longer follow-ups to see this effectiveness and evaluate variables such as motivation or adherence to chronic pain treatments. Finally, telemedicine rehabilitation may lead to lower costs for both patients and therapists, which may reduce waiting lists for clinical treatments.

Limitations
We found limited evidence for depressive symptoms; true effects might be different from our estimated effects. We found the presence of publication bias for depressive and stress symptoms; however, adjustments did not influence the results. All the studies have a high risk of bias; results should be interpreted cautiously. Future studies should improve their design quality to enhance our trust in their results. We have pooled together different BMT and conditions. However, we also provided sub-analyses where depressive symptoms and anxiety are analyzed by treatment and by condition.

Conclusions
e-BMT is an effective option for the management of anxiety and depressive symptoms in patients with musculoskeletal conditions causing chronic pain and should be introduced when in-person intervention is not possible. However, it does not seem effective to improve stress symptoms.  APA PsychINFO-75 results. (web or websites or internet or online or Online Therapy or mobile or Mobile Applications or remote treatment or digital treatment or Digital Interventions or Internet-Based Intervention or Telerehabilitation or Telemedicine) AND (chronic pain or persistent pain or long term pain or long-term pain) AND (randomized controlled trials or rct or randomised control trials) NOT (systematic review or meta-analysis or literature review or review of literature) NOT (pediatric or child or children or infant or adolescent).
Web of science-49 studies. TI = (Web OR eearth OR melth OR remote treatment OR digital treatment OR Mobile Applications OR Software OR Online OR Telephone OR Cell phone OR estherapy OR Internet OR Online OR Telerehabilitation OR Internet-Based Intervention OR Telerehabilitation OR Telemedicine) AND TI = (Chronic pain) AND TI = (randomi?ed controlled trial* OR rct).
Google Scholar.
("web" OR "online" OR "internet" OR "mobile" OR "telerehabilitation" OR "telemedicine") AND [allintitle:"chronic pain" OR "persistent pain"] AND ("randomized controlled trial" OR "randomised controlled trial OR "RCT")-review.     There were no significant differences between group for anxiety and depressive symptoms.   There were no significant differences between e-CBT and face-to-face CBT in depressive symptoms. There were no significant differences for any outcomes. No statistically significant differences between groups for stress (p = 0.21). The intervention group showed significant interaction effects of time x group for anxiety (p = 0.03) and depressive symptoms (p = 0.001).  Intervention group showed medium effects on depressive symptoms. There were no statistically significant differences in anxiety and depressive symptoms. There were no statistically significant interactions for group-by-time on depressive symptoms.    Appendix A.5. Assessment of the Quality of the Studies Based on the PEDro Scale Table A3. PEDro scale. Notes: 1: subject choice criteria are specified; 2: random assignment of subjects to groups; 3: hidden assignment; 4: groups were similar at baseline; 5: all subjects were blinded; 6: all therapists were blinded; 7: all evaluators were blinded; 8: measures of at least one of the key outcomes were obtained from more than 85% of baseline subjects; 9: intention-to-treat analysis was performed; 10: results from statistical comparisons between groups were reported for at least one key outcome; 11: the study provides point and variability measures for at least one key outcome.