Wilcoxon signed-ranks test. \*: *p* < 0.05, \*\*: *p* < 0.01, \*\*\*: *p* < 0.001.

**Table 5.** Univariate and multivariate regression model studying factors affecting the change in the number of body parts with WRMSs over time (T1−T0).


+ All factors refer to the changes over time (T1−T0) in Table 4, except gender, depressive symptoms, self-rated health status, and anxiety (found significant in T0 in Tables 1–3) and age. # Mutually adjusted for each other. \*: *p* < 0.05, \*\*: *p* < 0.01, \*\*\*: *p* < 0.001.

#### **4. Discussion**

Our study is one of the first to develop a WRMS-prevention multidisciplinary program with stretching and muscle-strengthening exercises for low-skilled workers in a low-income community setting. Our results have provided promising evidence of the 2-month program's effectiveness in reducing the number of body parts with WRMSs, job stress, working through pain, improved spine flexibility, and handgrip strength. The factors significantly affecting the reduction in the number of body parts with WRMSs were the change in the workstyle of working through pain, and self-rated health status. In addition, our analyses demonstrated a medium effect size (0.44) with a good statistical power of 0.98.

Our results have also strengthened the evidence that the community could be the setting for WRMSs prevention programs, apart from workplaces. Consistent with Lee et al.'s study [7] in a Korean community, our program not only improved workers' flexibility but also reduced WRMSs and other outcomes. The success of our program could be due to its partnership design, multidisciplinary and multidimensional approaches, and appropriate length (right dose). Studies have shown the ineffectiveness of single interventions (such as ergonomics training or stress management); rather, multidimensional interventions are recommended to tackle the multifactorial nature of WRMSs [4–6,8]. The element of TCM also addressed the cultural needs of the Chinese participants. Furthermore, our exercise duration (15.5–25 min) was longer than that of the Korean study (6 min) [7]. Thus, the community center could serve as a vehicle to reach low-skilled, disadvantaged workers, to allow them to access the WRMSs prevention programs, and to bring them together synergistically [27].

The Workstyle model [21] and the concept of presenteeism (continuing to work despite not feeling well) [28] could be used to explain our results further. Our program might have been able to improve low-skilled workers' musculoskeletal health literacy. For instance, they became more aware that WRMSs are not a necessary consequence of their work, that they are human beings not instruments or machines in the workplace, and that they should take an active role in improving their musculoskeletal health. They might have changed from passivity to becoming active in improving their own WRMSs. Using the Workstyle model, adverse workstyles have been identified as factors associated with WRMSs in office workers [21,29,30] and nursing assistants in nursing homes [31]. Also, working through pain [20,32], self-rated health [20], and self-rated mental health [32] have been found to be associated with WRMSs. Workers with WRMSs tended to have higher presenteeism (such as working through pain) and vice versa [33]. Furthermore, more WRMSs and higher presenteeism have been found to be related to low-skilled workers, low education, low job resources, and longer working hours, and to be more common in female workers [33]. In addition, high presenteeism has been associated with low self-rated health [28,33], and workers with low self-rated health tended to have more WRMSs [33]. Moreover, a population-based study (N = 1615) found that workers with long working hours, working without contracts, and being highly dependent on their wages to contribute to the total household income were associated with higher presenteeism [34]. Reasons given for presenteeism included not wanting to burden colleagues, not being able to afford to be absent for economic reasons, and worrying about being laid off [34]. The factors associated with presenteeism identified in the literature match with the general characteristics of low-skilled workers. Based on the Workstyle model, working with pain is a behavior resulting from limited job resources, financial need to support families, and fear of losing the job (cognitive reasoning and psychological considerations) [21,29]. Additionally, the East Asian hierarchical work structure weakens employees' power to question their seniors or those at management level [32]. Chinese culture also emphasizes self-discipline and the individual's responsibility to others and to society [35]. Because low-skilled workers are at the bottom of the organizational structure [36], they have low bargaining power and are expected to adhere to their job duties and be loyal to the organization. The promising results of our program reflect that the elements of the workshops might have changed the participants' cognitive and psychological appraisals (part of the Workstyle model), with an understanding that presenteeism (working while suffering from pain) could aggravate their WRMSs further.

This study was limited by its single-group pre-and-post study design, with the majority of our participants being female. This might temper the generalizability of our findings. However, we attempted to enhance the representativeness of the study population by including seven different community centers as recruitment sites. Nevertheless, this study provides preliminary evidence for the effectiveness of the program and gives rise to further investigations. A clustered random controlled trial with a robust design, such as a larger sample size with well-matched interventions and a control group, is needed to build up generalizable and representative evidence for the effectiveness of this multidisciplinary exercise intervention program.

There are implications for practice: The findings of our study can help to inform policymakers, employers, and occupational health and safety stakeholders to pay attention to WRMSs, particularly among low-skilled workers. Indeed, the promising results of our study suggest that more resources from policymakers should be allocated to community centers (e.g., non-governmental organizations) to conduct WRMS prevention and management programs. Further studies can explore the concept of the settings approach to improve low-skilled workers' musculoskeletal health literacy. The four core principles of the settings approach, namely, community participation, partnership, empowerment and equity [37], can be used as a framework to guide community-based interventions. Empowerment refers to "the process by which people gain control over the factors and decisions that shape their lives" [38]. Through community participation and partnership with different occupational health and safety stakeholders and employers, community-based programs can empower workers to take control of their own musculoskeletal health. Self-help groups, or even musculoskeletal health ambassadors, could be established in community centers. Those groups or ambassadors could provide further social support to their peers using social media (such as WhatsApp, Line, or WeChat). Addressing musculoskeletal health successfully in low-skilled workers could give them more equitable access to health promotion and to lobby for more resources allocated to communities. The equity approach would enable every low-skilled worker to have the opportunity to access musculoskeletal health and services.

#### **5. Conclusions**

Community could be the alternative setting for WRMS prevention programs for low-skilled workers. Our study has demonstrated that a community-based multidisciplinary program can reduce the number of body parts affected by WRMSs in low-skilled workers in low-income communities. Further studies should be conducted to test the program. Healthcare professions and policymakers should explore the concept of the settings approach to allocate resources to community centers in order to empower low-skilled workers to lobby their right for musculoskeletal health.

**Author Contributions:** K.C.: Planned the study, coordinated the study, drafted, and revised the manuscript. M.M.Y.T.: Planned the study, revised the discussion section, and proofread the manuscript. C.K.W.: Planned the study, coordinated the study, and proofread the manuscript. K.W.M.: Coordinated the intervention, and proofread the manuscript. S.K.L.: Planned the study, coordinated the data collection training, and proofread the manuscript. K.Y.M.: Coordinated the data collection training, drafted the methods section, and proofread the manuscript. K.T.S.T.: Performed data analysis, drafted results section, and proofread the manuscript. E.P.W.L.: Planned the study, and proofread the manuscript. All authors read and approved the final manuscript.

**Funding:** The study was funded by the Community Chest of Hong Kong and Social Welfare Development Fund.

**Acknowledgments:** The authors are grateful to all the low-skilled workers for their participation in the study and the funding provided by the Community Chest of Hong Kong and Social Welfare Development Fund.

**Conflicts of Interest:** The authors declare no conflict of interest.
