Mediators and moderators of scapular focused interventions on shoulder disability in individuals with chronic shoulder pain: Secondary analysis of a randomized controlled trial

Abstract

Objective

To explore potential mediators of treatment effect and moderators of scapular-focused interventions on shoulder disability at 12-week follow-up in patients with chronic shoulder pain.

Design

Secondary analysis of a randomized controlled trial that compared the effects of scapular movement training with standardized exercises.

Methods

Sixty-four participants with chronic shoulder pain were enrolled in the trial, of whom 58 completed the 12-week follow-up and were included in these analyses. Shoulder disability was assessed with Disabilities of the Arm, Shoulder and Hand (DASH) questionnaire. Potential mediators were scapular upward rotation, kinesiophobia, and fear avoidance beliefs. Potential moderators were kinesiophobia, fear avoidance beliefs, and duration of symptoms.

Results

Kinesiophobia, fear avoidance, and scapular upward rotation did not (p >0 .05) mediate the effects of scapular focused interventions on shoulder disability. Our sensitivity analyses indicated that the findings are likely to change because there were some small residual confounding factors between those mediators and the shoulder disability scores at 12 weeks. Kinesiophobia, fear avoidance, and duration of symptoms did not moderate the effects of the intervention on shoulder disability.

Conclusions

Kinesiophobia, fear avoidance, duration of symptoms, and scapular upward rotation did not mediate or moderate the effects of scapular-focused interventions on shoulder disability at 12-week follow-up.

Keywords

physical therapy rehabilitation scapula dyskinesis shoulder impingement syndrome

Introduction

Shoulder pain is a common disorder in the general population,^1–3 with impact on quality of life, work productivity, daily functional activities, and healthcare expenses over time.^4,5 Scapular dyskinesis is frequently observed in individuals with shoulder pain and scapular-focused interventions are commonly recommended and conducted in clinical practice to treat this condition.^6,7

Therapeutic exercises, including scapular-focused exercises, are supported and recommended by systematic reviews and clinical practice guidelines,^7–10 which often focus on improving scapular movement pattern.^11–14 However, it is unclear whether improvements in shoulder disability are mediated or moderated by scapular kinematics, psychological factors, and duration of symptoms. Therapeutic exercises or physical activity may cause changes in pain symptoms or disability indirectly, by changing psychological variables (e.g. kinesiophobia, fear avoidance beliefs).^15–17 In addition, previous studies described that patients’ characteristics, such as duration of symptoms and psychological factors, may influence the effects of therapeutic exercises on shoulder disability.^18–22 Nevertheless, the role of those potential mediators and moderators of scapular-focused interventions is still unknown.

A recent randomized controlled trial (RCT) investigated whether the scapular movement training is superior to standardized exercises in the treatment of individuals with shoulder pain.²³ The scapular movement training aimed to educate the individuals about the proper scapular position in the rest position and during arm elevation, and the muscle activation during the elevation of the arm, and to improve the pattern of scapular movement during arm elevation and during a relevant functional activity chosen by the patient. The findings suggested that both interventions significantly improved pain intensity, disability, kinesiophobia, and fear avoidance beliefs, but with no clinically important differences between groups. Mediation and moderation analyses can help us to understand whether the intervention failed to cause improvements in potential mediators and whether there is a subgroup of patients who may benefit more from scapular-focused exercises.

Mediation analysis is an approach to study causal mechanisms, which can parse a treatment effect into indirect (effect on an outcome that is carried through a selected mediator) and direct effects (remaining effect not explained by the selected mediator).^24–26 Moderators (treatment effect modifiers) are baseline characteristics of individuals that are associated with differential treatment response. Moderators influence the relationship between treatment and outcome (i.e. the effect of treatment on individuals depends on their value of the moderator, which precedes treatment and is not associated with treatment).²⁷

The aim of this secondary analysis of a RCT was to explore potential mediators and moderators of scapular-focused interventions on shoulder disability at 12-week follow-up in patients with chronic shoulder pain.

Methods

Design

This is a secondary analysis of data from a previous RCT that compared scapular movement training with standardized exercises for patients with chronic shoulder pain (NCT03528499).

Participants

Participants were eligible if they were aged between 18 to 60 years, primary complaint of shoulder pain for at least 3 months with a Numerical Pain Rating Scale score of 3 points^23,28 or greater during arm elevation, a positive Scapular Assistance Test (SAT), presence of scapular dyskinesis, and 150° or more of active arm elevation. Scapular dyskinesis was assessed and SAT was performed as previously described.^29,30 Individuals were excluded if they presented history of surgery, fracture, instability and/or dislocation of the shoulder, signs of adhesive capsulitis, massive rotator tears (positive drop-arm test), body mass index greater than 28 kg/m², pregnancy, tingling or numbness in the upper limb reproduced by the upper limb tension test or cervical compression test, self-reported tape allergy, neurologic or systemic illness, and physical therapy and corticosteroid injection in the last 6 and 3 months, respectively.^23,28,31

Interventions

The details of the interventions tested within the trial have been described elsewhere.^23,28 Both intervention groups involved 16 individual sessions, twice a week, over 8 weeks. Scapular movement training group involved education on scapular position and movements impairments and optimal muscle activation during the first week, which included verbal, tactile, and visual feedback with support of anatomical models, EMG biofeedback of lower trapezius, middle trapezius, and serratus anterior, and mirror for visual feedback. Following the education phase, the treatment included exercises with slow, conscious, and paced movements of arm elevation and a functional activity selected by the patient, which aimed to improve the pattern of scapular movement.

The standardized exercise group involved stretching exercises targeting the UT,³¹ posterior region of shoulder, and pectoralis minor,³² and the following strengthening exercises using elastic resistance bands (Theraband^®): prone extension,³³ prone horizontal abduction with external rotation,³³ serratus punch,³⁴ and side-lying external rotation.³³

Outcome measure

Shoulder disability was assessed with the Brazilian version of Disabilities of the Arm, Shoulder and Hand (DASH). The DASH is composed of 30 items that assess the ability to perform activities using the upper limb, and the final score ranges from 0 to 100, with lower scores indicating lower disability. The minimal clinically important difference in the DASH score is 10.8 points and the inter-rater and intra-rater reliability are 0,93 and 0,99, respectively.^35,36

Potential mediators

The selection of mediators was based on the theoretical rationale that both scapular upward rotation and psychological factors may cause improvement role in shoulder disability outcomes.^37,38 Previous observational studies found that changes in scapular upward rotation have been associated with shoulder pain and disability.^37,38 Scapular movement training focused on improving scapular movement patterns which would cause improvements in shoulder disability outcomes.²³ Psychological factors such as kinesiophobia and fear-avoidance beliefs were considered as putative mediators due to their association with higher pain intensity and greater disability in individuals with chronic shoulder pain.^39–42 These factors have been suggested to mediate the effects of therapeutic exercises on clinical outcomes (Figure 1).^43–46

Figure 1.

Directed acyclic graph representing the causal model with mediator of treatment effect on clinical outcomes (shoulder disability). *Mediators tested included: kinesiophobia, fear-avoidance beliefs, and scapular upward rotation.

Mediation analysis was conducted to identify the causal mechanisms through which ascapular-focused interventions affects shoulder disability. Such analyses allow researchers to estimate the direct and indirect effects of scapular-focused interventions.^24–26 The effect of a treatment analyzed through a selected mediator is considered the indirect effect, while the effect not determined by the selected mediator is the direct effect.⁴⁷

The selected mediators and theoretical rationale for the potentially important mediators are described in Table 1. The mediators were selected prior to the analysis. The scapular upward rotation was not considered in the moderation analysis because all individuals presented scapular dyskinesis and positive SAT. All mediators were measured at baseline and at the 12th week, except for scapular kinematics, which was measured in the 8th week.

Table 1.

Description of potential mediators and/or moderators and their theoretical causal mechanism.

Construct/measure	Theoretical rationale	Outcome measure
Kinesiophobia	Kinesiophobia is associated with increased disability and may influence the course of disability over time.^48,49 Effects of conservative interventions may be mediated by kinesiophobia.⁵⁰ Exercise may have positive effects on kinesiophobia,⁵¹ and contribute to reduced disability.	Tampa scale for kinesiophobia,^52,53 which is a valid and reliable instrument, with intraclass correlation coefficient (ICC) of 0.93.⁵³
Fear avoidance beliefs	Fear-Avoidance beliefs are associated with increased disability and may influence the course of disability over time and the effects of therapeutic exercises.^39,54 Effects of exercises may be mediated by fear avoidance.¹⁷ Therapeutic exercises may have positive effects on fear avoidance beliefs,⁵¹ which may contribute to reduced disability.	Fear avoidance beliefs questionnaire (physical activity subscale),^52,55 which is a valid and reliable instrument, with ICC of 0.97.⁵³
Scapular kinematics (upward rotation)	It is theorized that abnormal scapular movement is associated with increased shoulder pain and disability. Scapular-focused exercises aim to improve scapular kinematics, which may reduce pain and disability. The upward rotation presents smaller measurement error relative to overall movement compared to internal/external rotation and posterior/anterior tilt.⁵⁶	TrakSTAR hardware (Ascension Technology Corporation, Burlington, VT) integrated with the MotionMonitor software (Innovative Sports Training, Inc., Chicago, IL).^56,57
Duration of symptoms	Long duration of symptoms may contribute to the chronification of shoulder pain and poor treatment outcome.^58,59

Potential moderators

Through moderation analyses, researchers can estimate the interaction between baseline characteristics of patients with chronic shoulder pain, such as duration of symptoms, kinesiophobia, and fear avoidance beliefs, and clinical outcomes, following scapular-focused interventions. Such analyses assist in identifying which group of patients may respond better to a given treatment.²⁴

The following variables were assessed as potential moderators of the treatment effect: kinesiophobia, fear avoidance beliefs, and duration of symptoms. The work subscale of Fear Avoidance Beliefs Questionnaire was not considered due to possible floor effect in individuals with shoulder pain (Figure 2),⁵² and only the physical activity subscale was used in this mediation and moderation analysis.

Figure 2.

Directed acyclic graph representing the causal model with moderator of treatment effect on clinical outcomes (shoulder disability). *Moderators tested included: kinesiophobia, fear-avoidance beliefs, and scapular upward rotation.

Missing data

Sixty-four participants with chronic shoulder pain were included in the original trial, 59 (92.1%) completed the 8-week follow-up, and 58 (90.6%) completed 12-week follow-up, which resulted in missing data of 6 participants. In addition, the scapular kinematics data of 2 individuals were excluded from analysis due to noise in the electromagnetic signal, which also resulted in missing data in the analysis of scapular kinematics. We assumed missing data was missing at random (MAR) and implemented conditional multiple imputation by using the package mice⁶⁰ (R software),⁶¹ allowing for a maximum interaction of 50.

Statistical analysis

Sample size estimation

This is a post-hoc, secondary analysis of a previous trial,²³ which had an appropriate sample size to identify a minimum difference of 5.5^o in the scapular posterior tilt between the two groups. Despite this being a post-hoc and exploratory study, we estimated the required sample size for a mediation analysis under two different scenarios. In the first scenario, we assumed a large treatment-mediator and mediator-outcome effect (r = 0.6) in the second scenario, we assumed a moderate treatment-mediator and mediator-outcome effect (i.e. r = 0.3). We also assumed there was: (a) no exposure-mediator confounding, given this is a randomized trial; (b) a moderate confounding for the mediator-outcome (r = 0.3).⁶² We set power at 0.8.

The results of our sample size estimation suggested that: a minimum of 80 participants (40 per group) were required if there were large treatment-mediator and mediator-outcome effects. In case of moderate treatment-mediator and mediator-outcome effects, a minimum of 342 participants would be required. Those analyses suggest our present study was powered for detecting a large mediating effect but lacked power for detecting a moderate mediating effect.

Mediation

Mediation analyses was performed using the counterfactual framework,^63,64 assuming that there were: (1) no unmeasured confounding between the intervention-mediator relationship; (2) no unmeasured confounding between intervention-outcome relationship; (3) no unmeasured confounding between mediator and outcome; (4) no mediator-outcome confounder that is affected by the intervention; and (5) presence of temporal sequence from exposure (i.e. intervention), to the mediator to the outcome.

Assumptions #1 and #2 are satisfied because the participants were randomly allocated to interventions tested within the trial. Mediators were not randomized, so this path was potentially confounded. Hence, duration of symptoms was considered as confounder, and it was included as covariates in the outcome regression models. The baseline measures of potential mediators and baseline measures of the outcome measure were also included in the regression models (i.e. DASH score at the baseline) and included those as covariates in the outcome regression models.

Mediation analysis was performed using structural equation modeling to estimate the average causal mediation effect, average direct effect, average total effect, and proportion mediated.⁶⁴ Two linear models were fitted: the mediator model and the outcome model. The mediator model considered the potential mediator the dependent variable and treatment allocation the independent variable. The outcome model considered DASH scores at 12-week follow-up as the dependent variable and treatment allocation as the independent variable. The mediate disability (mediation package)⁶⁵ was used to compute the average causal mediation effect, average direct effect, average total effect, and proportion mediated, with bootstrap simulations (n = 1000) to compute 95% confidence intervals.⁶⁵

The first regression model included the mediator as the dependent variable; treatment as the independent variable, and duration of symptom as covariate. The second regression model included the DASH score at 12-week follow-up as dependent variable, and treatment as the independent variable and duration of symptoms as covariate.

Indirect and direct effects may be dependent on treatment allocation. Therefore, an interaction term (treatment×mediator) was included in the outcome model. The average causal mediation effects was calculated for each conditional on treatment exposure (x = 1 and x = 0) and their respective marginal effects because interaction effects can lead to significant bias on indirect and direct estimates of treatment effect.^66,67 The function “mediate” was used in R software to assess the total, direct, indirect effects and the proportion mediated.⁶⁸

To assess the robustness of our mediation analyses, sensitivity analyses was conducted to the no unmeasured confounding assumption⁶⁹ because we cannot rule out the possibility of confounding for the mediator-outcome relationship.^66,70 Sensitivity analyses was conducted using the medsens function (mediation package).⁶⁹

Moderation

Linear regression models were used to assess whether selected characteristics of participants at baseline moderated the effect of randomized interventions on pain and disability. Linear regressions were performed using the DASH score at 12 weeks follow-up as the dependent variable. Treatment allocation was included as the independent variable, while the duration of symptoms, fear avoidance, kinesiophobia, and DASH score at baseline as covariates.

For assessing whether treatment effects were moderated by one of the chose moderators, we included a standard interaction term between the selected moderator and group intervention. Standard diagnostic plots were used to assess regression assumptions of linearity and homoscedasticity. Interaction between exposure (i.e. treatment) and the potential moderator through plots of DASH score at 12 weeks versus the potential moderator were also visually assessed.

Descriptive statistics were conducted to summarize participants’ characteristics (i.e. number and percentage for describing categorical variables, mean and standard deviations for describing continuous variables). The R Software⁶¹ was used for conducting all analyses and set alpha at 0.05.

Results

The baseline characteristics of the participants are described in Table 2.

Table 2.

Baseline characteristics of participants.

Characteristics	Standardized exercises group (n = 32)	Scapular movement training group (n = 32)
Age, years	37.0 ± 13.0	38.3 ± 11.9
Height, cm	167.9 ± 9.7	171.1 ± 8.5
Body mass, kg	69.6 ± 10.7	73.4 ± 10.1
Sex, n (%)
Female/male	19 (59.3)/13 (40.6)	15 (46.8)/17 (53.1)
Duration of symptoms, months	28.7 ± 43.3	21.1 ± 24.2
Unilateral symptoms, n (%)	24 (75.0)	27 (84.4)
Most painful side, n (%)
Dominant	19 (59.4)	19 (59.4)

Values are reported as mean ± SD, unless otherwise indicated.

Mediators

The findings of the mediation analyses are presented in Table 3. Changes in kinesiophobia, fear avoidance, and scapular upward rotation during arm elevation and lowering at scapular plane did not present direct or indirect mediator effect on shoulder disability in individuals with chronic shoulder pain. Sensitivity analyses suggest small violations of the no unmeasured confounding assumption (e.g. r ≥ 0.1) may change our estimate of mediation effect (Supplemental material).

Table 3.

Potential mediators of treatment effect.

Mediator	Path a (95% CI)	Path b (95% CI)	Total effect (95% CI)	p-value	Direct effect (95% CI)	p-value	Indirect effect (95% CI)	p-value	Proportion mediated (95% CI)	p-value
Kinesiophobia	−2.9(−5.7, 0.0)	13.5(−27.6, 54.7)	3.4(−3.6, 10.8)	0.34	3.5(−3.4, 10.5)	0.33	−0.1(−2.9, 2.6)	0.94	0.0(−2.5, 3.7)	0.99
Fear avoidance (physical activity)	−0.2(−3.4, 3)	3.4(−7, 13.9)	3(−3.9, 10.3)	0.44	3.4(−2.8, 10)	0.3	−0.3(−4.4, 3.5)	0.84	0.0(−3.7, 5.8)	0.89
Scapular upward rotation
Concentric phase of flexion at 60°	−0.5(−3.8, 2.8)	−4.7(−19.7, 10.4)	3.3(−4.2, 11.2)	0.4	3.0(−4.2, 10.5)	0.41	0.3(−1.9, 2.6)	0.77	0.0(−1.8, 1.9)	0.75
Eccentric phaseof flexion at 60°	2.0(−1.6, 5.6)	1.4(−13.8, 16.5)	3.1(−4.2, 10.4)	0.41	3.6(−3.6, 10.9)	0.33	−0.4(−2.8, 1.2)	0.58	0.0(−2.2, 2.5)	0.8
Concentric phase of abduction at 60°	−0.1(−2.9, 2.8)	19.1(2.1, 36.1)	3.4(−3.9, 11.1)	0.35	3.4(−3.8, 10.9)	0.36	0.0(−1.5, 1.4)	0.96	0.0(−1.1, 1.4)	0.94
Eccentric phase of abduction at 60°	1.3(−2.3, 4.9)	11.5(−5.1, 28.2)	2.9(−4, 9.9)	0.41	2.9(−3.9, 9.5)	0.39	0.0(−1.7, 1.7)	0.99	0.0(−1.2, 1.5)	0.93
Concentric phase of scaption at 60°	−0.6(−3.6, 2.3)	−1.1(−17.4, 15.1)	3.2(−3.9, 10.7)	0.38	2.8(−4, 9.8)	0.45	0.4(−2.2, 3.1)	0.7	0.1(−2, 2.9)	0.68
Eccentric phase of scaption at 60°	2.2(−1.5, 5.9)	−2.7(−18.2, 12.8)	3.4(−3.6, 11.1)	0.39	4.0(−3.1, 12.1)	0.29	−0.7(−3.3, 1)	0.45	−0.1(−3.1, 2.5)	0.67

Moderator

The findings of the moderation analyses are presented in Table 4. Kinesiophobia (effect = 0.34, 95% CI: −0.87 to 1.55), fear avoidance (effect = 0.27, 95% CI: −0.97 to 1.52), and duration of symptoms (effect = 0.00, 95% CI: −0.22 to 0.23) did not moderate the treatment effect of interventions.

Table 4.

Moderators of DASH score changes to interventions.

Moderator	Β (95% CI)	Interaction p-value
Kinesiophobia	0.34 (−0.87, 1.55)	0.57
Fear avoidance	0.27 (−0.97, 1.52)	0.65
Duration of symptoms	0.00 (−0.22, 0.23)	0.98

Discussion

We conducted this exploratory secondary analysis a recently published trial, (NCT03528499), to explore whether kinesiophobia, fear-avoidance beliefs, and scapular upward rotation during arm elevation and lowering at scapular plane mediated treatment effects, or whether kinesiophobia, fear avoidance, and duration of symptoms moderated treatment effects. Our results showed no mediation effect on shoulder disability, and that kinesiophobia, fear avoidance, and duration of symptoms did not moderate treatment effects in patients with chronic shoulder pain being treated with scapular movement training or standardized exercises.

The results of this study did not find any mediating effect of scapular upward rotation on shoulder disability. Our results suggest scapular-focused interventions did not cause changes in scapular upward rotation (i.e. path a). The findings of this study are in accordance with a previous study,⁷¹ which suggests that scapular kinematics do not mediate the effects of scapular exercises on shoulder pain or disability in individuals with shoulder pain. The role of scapular dyskinesis has been questioned^72–74 as the pattern of altered scapular kinematics is not consistent among patients with shoulder pain, scapular dyskinesis may be present in asymptomatic individuals, the changes in muscle activity or scapular movements may be a consequence of painful symptoms in shoulder region.^{72,73,75–77} In addition, no association has been found between changes in pain or function scores and changes in scapular upward/downward rotations or scapular anterior/posterior tilt.⁷⁸ On the other hand, a previous study found that changes in scapular movement patterns were fairly correlated with disability scores in patients with shoulder subacromial pain, but those correlations may be confounded.⁷⁹

Systematic reviews have indicated that scapular-focused exercises may present small and not clinically important benefits on pain intensity and disability compared to the general approach,^7,80,81 and that there are conflicting findings or no changes on scapular movement or position.⁸¹ However, clinicians still recommend scapulothoracic muscle strengthening exercises or scapular movement training based on the rationale that patients with shoulder pain frequently present weak or altered muscle activity of scapulothoracic muscles or impaired scapular movement.^82–86 Nevertheless, the improvements in disability of the patients of this study were not explained by changes in scapular upward rotation, which suggests that clinicians should not focus on changing scapular upward rotation during the treatment of patients with shoulder pain. Therefore, further studies should consider the analysis of additional outcomes, such as pain intensity and self-perception of change in health condition, and additional mediators or moderators, including pain catastrophizing, self-efficacy, anxiety, depression, patient's expectations, scapulothoracic muscle activity or muscle strength, and tendon structural changes, all of which may play an important role in shoulder rehabilitation.

Previous studies demonstrated that psychological factors play an important role in the development and prognosis of shoulder pain,^22,39 and kinesiophobia and fear-avoidance beliefs were associated with high intensity of shoulder pain and disability.^22,39,40,87 Therapeutic exercises may reduce pain-related fear in individuals with chronic musculoskeletal disorders.^51,88 Our findings showed that patients’ kinesiophobia and fear-avoidance beliefs did not mediate nor moderated the effects of scapular-focused interventions. The lack of mediation may be explained by the small changes observed in kinesiophobia and fear-avoidance outcomes at 12-week follow-up. In addition, the changes in those variables did not exceed the minimal detectable change of those outcome measures.^52,89–91 Another possible explanation is that the Tampa Scale for Kinesiophobia and the Fear-Avoidance Beliefs Questionnaire were originally developed for individuals with chronic low back pain and may exhibit floor effects when used in populations with shoulder pain.⁵²

Riley et al. (2020) conducted mediation analyses that suggested that improvement in positive psychological factors (i.e. self-efficacy) may be more meaningful for functional outcome than improvements in negative psychological factors (e.g. fear avoidance beliefs) in individuals with low back pain. Unfortunately, in the original trial, other psychological factors were not measured, including pain catastrophizing, anxiety, depression, self-efficacy, and patients’ expectations, which might have provided valuable information on mechanisms of therapeutic exercises on shoulder rehabilitation. Therefore, future studies should also measure psychological variables (e.g. self-efficacy, anxiety, depression, pain catastrophizing) at different time points to estimate whether they mediate treatment effects of scapular-focused interventions.

Previous studies indicated that the duration of symptoms predicts poorer outcomes after treatment.^58,92–94 Interestingly, the duration of symptoms did not moderate the effects of scapular-focused interventions on shoulder disability. Only individuals with chronic symptoms were included in the original study, potentially accounting for the lack of moderation effects by the duration of symptoms on shoulder disability.

This study provides important insights, describing that treatment effects were not mediated by kinesiophobia, fear-avoidance beliefs, or scapular upward rotation. Furthermore, it describes whether treatment effects were not moderated by kinesiophobia, fear avoidance, or duration of symptoms. The information on how interventions work may assist the delivering effective treatments and reaching better outcomes.

There are limitations to our study. The sample size was calculated for comparison of the effects between scapular movement training and standardized exercises. The secondary analyses reported in this study were not planned a priori and were only powered to detect large mediating effects. Therefore, the sample size was insufficient to detect small or moderate effects. There are different methods for estimating sample size for mediation studies, with different trade-offs between those methods.⁹⁵ Given the little literature in this area and despite the limitations of our study, our findings are informative for researchers planning future projects in this area. Multiple analyses were conducted which increased the risk of type I error. However, our goal was to explore potential mechanisms of action of the interventions tested within the original trial, and confirmation analyses on larger data sets should be conducted before the results of this study are interpreted with certainty. Nevertheless, these are valuable for improving the design of future trials. The analyses considered only the changes in shoulder disability as the outcome with a limited number of mediators (scapular upward rotation, kinesiophobia, and fear avoidance beliefs) and moderators (kinesiophobia, fear avoidance beliefs, and duration of symptoms) that were collected in the original trial and due to limited sample size.⁹⁶ In addition, there are other factors that were not measured in this study (e.g. self-efficacy, anxiety, depression, pain catastrophizing) that might interact with the factors measured in this study. These unmeasured factors should be investigated in future studies.

In conclusion, the effects of scapular-focused interventions on shoulder disability were not mediated by changes in scapular kinematics, kinesiophobia, and fear avoidance beliefs. In addition, those effects were not moderated by the duration of symptoms, kinesiophobia and fear avoidance beliefs as measured at baseline.

Supplemental Material

sj-docx-1-sel-10.1177_17585732251354920 - Supplemental material for Mediators and moderators of scapular focused interventions on shoulder disability in individuals with chronic shoulder pain: Secondary analysis of a randomized controlled trial

Supplemental material, sj-docx-1-sel-10.1177_17585732251354920 for Mediators and moderators of scapular focused interventions on shoulder disability in individuals with chronic shoulder pain: Secondary analysis of a randomized controlled trial by Danilo Harudy Kamonseki, Daniel Cury Ribeiro, Melina Nevoeiro Haik, Larissa Pechincha Ribeiro, Rafaela Firmino Almeida and Paula Rezende Camargo in Shoulder & Elbow

Footnotes

Author contributions

Each of the authors has contributed substantially and concurs with the content in the manuscript. Conceptualization: DHK, DCR, MNH, LPR, PRC. Data curation: DHK, DCR, MNH, PRC. Formal analysis: DCR. Funding acquisition: DHK, DCR, PRC. Investigation: DHK, LPR, RFA. Methodology: DHK, MNH, LPR, RFA, PRC. Project administration: DHK, MNH, RFA, PRC. Resources: DHK, PRC. Supervision: MNH, PRC. Validation: DCR, MNH, PRC. Visualization: DCR, MNH, LPR, PRC. Roles/Writing—original draft: DHK, DCR, MNH, RFA, PRC. Writing—review & editing: DHK, DCR, MNH, LPR, PRC.

Declaration of conflicting interests

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

Ethical approval

This is a secondary analysis of a randomized controlled trial (NCT03528499), which was approved by the Human Research Ethics Committee of the Federal University of São Carlos (reference number: 86974318.7.0000.5504).

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by Fundação de Amparo à Pesquisa do Estado de São Paulo—FAPESP (2018/04911-1; 2020/00771-0), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior—Brasil (CAPES), and Conselho Nacional de Desenvolvimento Científico e Tecnológico—CNPq (142373/2018-4). The research was conducted during tenure of The Sir Charles Hercus Health Research Fellowship of the Health Research Council of New Zealand [Grant number: 18/111] awarded to DCR.

Informed consent

All individuals signed a written consent before study enrollment and were aware of the main purpose of the study.

ORCID iDs

Danilo Harudy Kamonseki

Daniel Cury Ribeiro

Melina Nevoeiro Haik

Supplemental material

Supplemental material for this article is available online.

References

Luime

Koes

Hendriksen

IJM

, et al. Prevalence and incidence of shoulder pain in the general population; a systematic review. Scand J Rheumatol 2004; 33: 73–81.

Picavet

HSJ

Schouten

JSAG

. Musculoskeletal pain in the Netherlands: prevalences, consequences and risk groups, the DMC3-study. Pain 2003; 102: 167–178.

Feleus

Bierma-Zeinstra

SMA

Miedema

, et al. Incidence of non-traumatic complaints of arm, neck and shoulder in general practice. Man Ther 2008; 13: 426–433.

Hallman

Holtermann

Dencker-Larsen

, et al. Are trajectories of neck-shoulder pain associated with sick leave and work ability in workers? A 1-year prospective study. BMJ Open 2019; 9: e022006.

Marks

Comans

Bisset

, et al. Shoulder pain cost-of-illness in patients referred for public orthopaedic care in Australia. Aust Heal Rev 2019; 43: 540–548.

Cools

AMJ

Struyf

De Mey

, et al. Rehabilitation of scapular dyskinesis: from the office worker to the elite overhead athlete. Br J Sports Med 2014; 48: 692–697.

Saito

Harrold

Cavalheri

, et al. Scapular focused interventions to improve shoulder pain and function in adults with subacromial pain: a systematic review and meta-analysis. Physiother Theory Pract 2018; 34: 653–670.

Steuri

Sattelmayer

Elsig

, et al. Effectiveness of conservative interventions including exercise, manual therapy and medical management in adults with shoulder impingement: a systematic review and meta-analysis of RCTs. Brit J Sports Med 2017; 51: 1340–1347.

Pieters

Lewis

Kuppens

, et al. An update of systematic reviews examining the effectiveness of conservative physical therapy interventions for subacromial shoulder pain. J Orthop Sport Phys Ther 2020; 50: 131–141.

10.

Takeno

Glaviano

Norte

, et al. Therapeutic interventions for scapular kinematics and disability in patients with subacromial impingement: a systematic review. J Athl Train 2019; 54: 283–295.

11.

Ludewig

Kamonseki

Staker

, et al. Changing our diagnostic paradigm: movement system diagnostic classification. Int J Sports Phys Ther 2017; 12: 884–893.

12.

Hotta

Santos

McQuade

, et al. Scapular-focused exercise treatment protocol for shoulder impingement symptoms: three-dimensional scapular kinematics analysis. Clin Biomech 2018; 51: 76–81.

13.

Hotta

Gomes de Assis Couto

Cools

, et al. Effects of adding scapular stabilization exercises to a periscapular strengthening exercise program in patients with subacromial pain syndrome: a randomized controlled trial. Musculoskelet Sci Pract 2020; 49: 102171.

14.

de Oliveira Scatolin

Hotta

Cools

AMJ

, et al. Effect of conscious abdominal contraction on the activation of periscapular muscles in individuals with subacromial pain syndrome. Clin Biomech 2021; 84: 105349.

15.

Courbalay

Jobard

Descarreaux

, et al. Direct and indirect relationships between physical activity, fitness level, kinesiophobia, and health-related quality of life in patients with rheumatic and musculoskeletal diseases: a network analysis. J Pain Res 2021; 14: 3387–3399.

16.

Mansell

Hill

Main

, et al. Mediators of treatment effect in the back in action trial. Clin J Pain 2017; 33: 811–819.

17.

Chalder

Goldsmith

White

, et al. Rehabilitative therapies for chronic fatigue syndrome: a secondary mediation analysis of the PACE trial. Lancet Psychiatry 2015; 2: 141–152.

18.

Feleus

van Dalen

Bierma-Zeinstra

, et al. Kinesiophobia in patients with non-traumatic arm, neck and shoulder complaints: a prospective cohort study in general practice. BMC Musculoskelet Disord 2007; 8: 117.

19.

Bot

van der Waal

Terwee

, et al. Predictors of outcome in neck and shoulder symptoms: a cohort study in general practice. Spine (Phila Pa 1976) 2005; 30: E459–E470.

20.

Karels

Bierma-Zeinstra

SMA

Burdorf

, et al. Social and psychological factors influenced the course of arm, neck and shoulder complaints. J Clin Epidemiol 2007; 60: 839–848.

21.

van der Windt

DAWM

Kuijpers

Jellema

, et al.

Do psychological factors predict outcome in both low-back pain and shoulder pain?

Ann Rheum Dis 2007; 66: 313–319.

22.

Martinez-Calderon

Meeus

Struyf

, et al. The role of psychological factors in the perpetuation of pain intensity and disability in people with chronic shoulder pain: a systematic review. BMJ Open 2018; 8: e020703.

23.

Kamonseki

Haik

Ribeiro

, et al. Scapular movement training is not superior to standardized exercises in the treatment of individuals with chronic shoulder pain and scapular dyskinesis: randomized controlled trial. Disabil Rehabil 2023; 45: 2925–2935.

24.

Kraemer

. Messages for clinicians: moderators and mediators of treatment outcome in randomized clinical trials. Am J Psychiatry 2016; 173: 672–679.

25.

Lee

Herbert

Lamb

, et al. Investigating causal mechanisms in randomised controlled trials. Trials 2019; 20: 524.

26.

Mellor

Kasza

Grimaldi

, et al. Mediators and moderators of education plus exercise on perceived improvement in individuals with gluteal tendinopathy: an exploratory analysis of a 3-arm randomized trial. J Orthop Sport Phys Ther 2022; 52: 826–836.

27.

Kraemer

Wilson

Fairburn

, et al. Mediators and moderators of treatment effects in randomized clinical trials. Arch Gen Psychiatry 2002; 59: 877–883.

28.

Kamonseki

Haik

Camargo

. Scapular movement training versus standardized exercises for individuals with chronic shoulder pain: protocol for a randomized controlled trial. Braz J Phys Ther 2021; 25: 221–229.

29.

McClure

Tate

Kareha

, et al. A clinical method for identifying scapular dyskinesis, part 1: reliability. J Athl Train 2009; 44: 165–173.

30.

Rabin

Irrgang

Fitzgerald

, et al. The intertester reliability of the Scapular Assistance Test. J Orthop Sports Phys Ther 2006; 36: 653–660.

31.

Camargo

Alburquerque-Sendín

Avila

, et al. Effects of stretching and strengthening exercises, with and without manual therapy, on scapular kinematics, function, and pain in individuals with shoulder impingement: a randomized controlled trial. J Orthop Sports Phys Ther 2015; 45: 984–997.

32.

Cools

Johansson

Cagnie

, et al. Stretching the posterior shoulder structures in subjects with internal rotation deficit: comparison of two stretching techniques. Shoulder Elbow 2012; 4: 56–63.

33.

De Mey

Cagnie

Van De Velde

, et al. Trapezius muscle timing during selected shoulder rehabilitation exercises. J Orthop Sport Phys Ther 2009; 39: 743–752.

34.

Castelein

Cagnie

Parlevliet

, et al.

Serratus anterior or pectoralis minor: which muscle has the upper hand during protraction exercises?

Man Ther 2016; 22: 158–164.

35.

Orfale

Araújo

PMP

Ferraz

, et al. Translation into Brazilian Portuguese, cultural adaptation and evaluation of the reliability of the Disabilities of the Arm, Shoulder and Hand Questionnaire. Braz J Med Biol Res 2005; 38: 293–302.

36.

Franchignoni

Vercelli

Giordano

, et al. Minimal clinically important difference of the disabilities of the arm, shoulder and hand outcome measure (DASH) and its shortened version (QuickDASH). J Orthop Sports Phys Ther 2014; 44: 30–39.

37.

Lawrence

Braman

Ludewig

. Shoulder kinematics impact subacromial proximities: a review of the literature. Braz J Phys Ther 2020; 24: 219–230.

38.

Ludewig

Reynolds

. The association of scapular kinematics and glenohumeral joint pathologies. J Orthop Sport Phys Ther 2009; 39: 90–104.

39.

Martinez-Calderon

Struyf

Meeus

, et al. The association between pain beliefs and pain intensity and/or disability in people with shoulder pain: a systematic review. Musculoskelet Sci Pract 2018; 37: 29–57.

40.

Kamonseki

Pott-Junior

Haik

, et al. Pain-related fear phenotypes are associated with function of the upper limbs in individuals with shoulder pain. Musculoskelet Sci Pract 2021; 55: 102416.

41.

González Aroca

Díaz

ÁP

Navarrete

, et al. Fear-avoidance beliefs are associated with pain intensity and shoulder disability in adults with chronic shoulder pain: a cross-sectional study. J Clin Med 2023; 12: 3376.

42.

Ata

Tuncer

Kara

, et al. The relationship between kinesiophobia, balance, and upper extremity functions in patients with painful shoulder pathology. PM R 2024; 16: 1088–1094.

43.

Ryum

Stiles

. Changes in pain catastrophizing, fear-avoidance beliefs, and pain self-efficacy mediate changes in pain intensity on disability in the treatment of chronic low back pain. Pain Rep 2023; 8: e1092.

44.

Wood

Bejarano

Csiernik

, et al. Pain catastrophising and kinesiophobia mediate pain and physical function improvements with Pilates exercise in chronic low back pain: a mediation analysis of a randomised controlled trial. J Physiother 2023; 69: 168–174.

45.

Varallo

Scarpina

Giusti

, et al.

Does kinesiophobia mediate the relationship between pain intensity and disability in individuals with chronic low-back pain and obesity?

Brain Sci 2021; 11: 684.

46.

Mansell

den Hollander

Lotzke

, et al. A person-centred prehabilitation program based on cognitive behavioural physical therapy for patients scheduled for lumbar fusion surgery: a mediation analysis to assess fear of movement (kinesiophobia), self-efficacy and catastrophizing as mediators of health outcomes. Eur J Pain 2022; 26: 1790–1799.

47.

Joyce

Chernofsky

Lodi

, et al. Do physical therapy and yoga improve pain and disability through psychological mechanisms? A causal mediation analysis of adults with chronic low back pain. J Orthop Sport Phys Ther 2022; 52: 470–483.

48.

Luque-Suarez

Martinez-Calderon

Navarro-Ledesma

, et al. Kinesiophobia is associated with pain intensity and disability in chronic shoulder pain: a cross-sectional study. J Manipulative Physiol Ther 2020; 43: 791–798.

49.

Luque-Suarez

Martinez-Calderon

Falla

. Role of kinesiophobia on pain, disability and quality of life in people suffering from chronic musculoskeletal pain: a systematic review. Br J Sports Med 2019; 53: 554–559.

50.

Barnhoorn

Staal

van Dongen

RTM

, et al. Are pain-related fears mediators for reducing disability and pain in patients with complex regional pain syndrome Type 1? An explorative analysis onpain exposure physical therapy. PLoS One 2015; 10: e0123008.

51.

Martinez-Calderon

Flores-Cortes

Morales-Asencio

, et al. Conservative interventions reduce fear in individuals with chronic low back pain: a systematic review. Arch Phys Med Rehabil 2020; 101: 329–358.

52.

Kamonseki

Haik

Ribeiro

, et al. Measurement properties of the Brazilian versions of fear-avoidance beliefs questionnaire and Tampa scale of kinesiophobia in individuals with shoulder pain. PLoS One 2021; 16: e0260452.

53.

de Souza

da Silva Marinho

Siqueira

, et al. Psychometric testing confirms that the Brazilian-Portuguese adaptations, the original versions of the Fear-Avoidance Beliefs Questionnaire, and the Tampa Scale of Kinesiophobia have similar measurement properties. Spine (Phila Pa 1976) 2008; 33: 1028–1033.

54.

Janela

Costa

Molinos

, et al. Fear Avoidance Beliefs in upper-extremity musculoskeletal pain conditions: secondary analysis of a prospective clinical study on digital care programs. Pain Med 2023; 24: 451–460.

55.

de Abreu

Faria

CDCM

Cardoso

SMV

, et al. The Brazilian version of the Fear Avoidance Beliefs Questionnaire. Cad Saude Publica 2008; 24: 615–623.

56.

Haik

Alburquerque-Sendín

Camargo

. Reliability and minimal detectable change of 3-dimensional scapular orientation in individuals with and without shoulder impingement. J Orthop Sports Phys Ther 2014; 44: 341–349.

57.

Van Der Helm

FCT

Veeger

HEJ

, et al. ISB Recommendation on definitions of joint coordinate systems of various joints for the reporting of human joint motion – part II: shoulder, elbow, wrist and hand. J Biomech 2005; 38: 981–992.

58.

Chester

Shepstone

Daniell

, et al. Predicting response to physiotherapy treatment for musculoskeletal shoulder pain: a systematic review. BMC Musculoskelet Disord 2013; 14: 203.

59.

Struyf

Geraets

Noten

, et al. A multivariable prediction model for the chronification of non-traumatic shoulder pain: a systematic review. Pain Physician 2016; 19: 1–10.

60.

van Buuren

Groothuis-Oudshoorn

. Mice : multivariate imputation by chained equations in R. J Stat Softw 2011; 45: 1–67.

61.

R_Core_Team. A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing, http://www.r-project.org (2008).

62.

Vittinghoff

Neilands

. Sample size for joint testing of indirect effects. Prev Sci 2015; 16: 1128–1135.

63.

Valeri

VanderWeele

. Mediation analysis allowing for exposure–mediator interactions and causal interpretation: theoretical assumptions and implementation with SAS and SPSS macros. Psychol Methods 2013; 18: 137–150.

64.

VanderWeele

TJ.

Explanation in causal inference: Methods for mediation and interaction. New York: Oxford University Press, 2015.

65.

Tingley

Yamamoto

Hirose

, et al. Mediation : R package for causal mediation analysis. J Stat Softw 2014; 59: 1–13.

66.

Imai

Keele

Tingley

. A general approach to causal mediation analysis. Psychol Methods 2010; 15: 309–334.

67.

VanderWeele

. A unification of mediation and interaction. Epidemiology 2014; 25: 749–761.

68.

Imai

Keele

Tingley

, et al. Causal mediation analysis using R. In: HD

(ed.) Advances in social science research using R. New York: Springer-Verlag, 2019, pp.51–71.

69.

Imai

Keele

Yamamoto

. Identification, inference and sensitivity analysis for causal mediation effects. Stat Sci 2010; 25: 51–71.

70.

Richiardi

Bellocco

Zugna

. Mediation analysis in epidemiology: methods, interpretation and bias. Int J Epidemiol 2013; 42: 1511–1519.

71.

Hotta

Alaiti

Ribeiro

, et al. Causal mechanisms of a scapular stabilization intervention for patients with subacromial pain syndrome: a secondary analysis of a randomized controlled trial. Arch Physiother 2022; 12: 13.

72.

D’hondt

Pool

JJM

Kiers

, et al. Validity of clinical measurement instruments assessing scapular function: insufficient evidence to recommend any instrument for assessing scapular posture, movement, and dysfunction—A systematic review. J Orthop Sport Phys Ther 2020; 50: 632–641.

73.

Hogan

Corbett

J-A

Ashton

, et al. Scapular dyskinesis is not an isolated risk factor for shoulder injury in athletes: a systematic review and meta-analysis. Am J Sports Med 2021; 49: 2843–2853.

74.

Nodehi Moghadam

Rahnama

Noorizadeh Dehkordi

, et al. Exercise therapy may affect scapular position and motion in individuals with scapular dyskinesis: a systematic review of clinical trials. J Shoulder Elbow Surg 2020; 29: e29–e36.

75.

Hodges

Tucker

. Moving differently in pain: a new theory to explain the adaptation to pain. Pain 2011; 152: S90–S98.

76.

Ratcliffe

Pickering

McLean

, et al. Is there a relationship between subacromial impingement syndrome and scapular orientation? A systematic review. Br J Sports Med 2014; 48: 1251–1256.

77.

Hickey

Solvig

Cavalheri

, et al. Scapular dyskinesis increases the risk of future shoulder pain by 43% in asymptomatic athletes: a systematic review and meta-analysis. Br J Sports Med 2018; 52: 102–110.

78.

Jafarian Tangrood

Sole

Cury Ribeiro

. Association between changes in pain or function scores and changes in scapular rotations in patients with subacromial shoulder pain: a prospective cohort study. Arch Physiother 2022; 12: 18.

79.

Jafarian Tangrood

Sole

Ribeiro

. Is there an association between changes in pain or function with changes in scapular dyskinesis: a prospective cohort study. Musculoskelet Sci Pract 2020; 48: 102172.

80.

Bury

West

Chamorro-Moriana

, et al. Effectiveness of scapula-focused approaches in patients with rotator cuff related shoulder pain: a systematic review and meta-analysis. Man Ther 2016; 25: 35–42.

81.

Reijneveld

EAE

Noten

Michener

, et al. Clinical outcomes of a scapular-focused treatment in patients with subacromial pain syndrome: a systematic review. Br J Sports Med 2017; 51: 436–441.

82.

Barcia

Makovicka

Spenciner

, et al. Scapular motion in the presence of rotator cuff tears: a systematic review. J Shoulder Elbow Surg 2021; 30: 1679–1692.

83.

Timmons

Thigpen

Seitz

, et al. Scapular kinematics and subacromial-impingement syndrome: a meta-analysis. J Sport Rehabil 2012; 21: 354–370.

84.

Struyf

Cagnie

Cools

, et al. Scapulothoracic muscle activity and recruitment timing in patients with shoulder impingement symptoms and glenohumeral instability. J Electromyogr Kinesiol 2014; 24: 277–284.

85.

Neumann

Camargo

. Kinesiologic considerations for targeting activation of scapulothoracic muscles - part 1: serratus anterior. Braz J Phys Ther 2019; 23: 459–466.

86.

Phadke

Camargo

Ludewig

. Scapular and rotator cuff muscle activity during arm elevation: a review of normal function and alterations with shoulder impingement. Rev Bras Fisioter 2009; 13: 1–9.

87.

Haik

Alburquerque-Sendín

Fernandes

RAS

, et al. Biopsychosocial aspects in individuals with acute and chronic rotator cuff related shoulder pain: classification based on a decision tree analysis. Diagnostics 2020; 10: 928.

88.

Hanel

Owen

Held

, et al. Effects of exercise training on fear-avoidance in pain and pain-free populations: systematic review and meta-analysis. Sport Med 2020; 50: 2193–2207.

89.

Mintken

Cleland

Whitman

, et al. Psychometric properties of the Fear-Avoidance Beliefs Questionnaire and Tampa Scale of Kinesiophobia in patients with shoulder pain. Arch Phys Med Rehabil 2010; 91: 1128–1136.

90.

Pulles

ANTD

Köke

AJA

Strackke

, et al. The responsiveness and interpretability of psychosocial patient-reported outcome measures in chronic musculoskeletal pain rehabilitation. Eur J Pain 2020; 24: 134–144.

91.

Inrig

Amey

Borthwick

, et al. Validity and reliability of the Fear-Avoidance Beliefs Questionnaire (FABQ) in workers with upper extremity injuries. J Occup Rehabil 2012; 22: 59–70.

92.

Hallman

Rasmussen

CDN

Jørgensen

, et al. Time course of neck-shoulder pain among workers: a longitudinal latent class growth analysis. Scand J Work Environ Health 2018; 44: 47–57.

93.

Kooijman

Barten

D-JA

Swinkels

, et al. Pain intensity, neck pain and longer duration of complaints predict poorer outcome in patients with shoulder pain – a systematic review. BMC Musculoskelet Disord 2015; 16: 288.

94.

Parr

Borsa

Fillingim

, et al. Psychological influences predict recovery following exercise induced shoulder pain. Int J Sports Med 2013; 35: 232–237.

95.

VanderWeele

. Invited Commentary: frontiers of power assessment in mediation analysis. Am J Epidemiol 2020; 189: 1568–1570.

96.

Holden

Matthews

Rathleff

, et al. How do hip exercises improve pain in individuals with patellofemoral pain? Secondary mediation analysis of strength and psychological factors as mechanisms. J Orthop Sport Phys Ther 2021; 51: 602–610.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.10 MB

0.00 MB