Job-Related Determinants of Musculoskeletal Disorders Across U.S. Occupations: Moving Beyond the Industry Sector

1. Introduction

Musculoskeletal disorders (MSDs) represent not merely a pervasive health crisis for the American workforce, but a persistent and debilitating economic burden that fundamentally challenges public health and occupational safety initiatives. These disorders, which affect the muscles, nerves, tendons, joints, cartilage, and spinal discs, are broadly categorized as either non-work-related (resulting from external factors like sports injuries) or Work-Related Musculoskeletal Disorders (WMSDs). WMSDs are specifically defined as injuries and disorders caused or significantly aggravated by occupational tasks, most commonly arising from cumulative exposure to repetitive motions, prolonged physical exertion, heavy lifting, and awkward or constrained postures [1]. Given that this study focuses exclusively on the incidence and analysis of occupational risk factors, WMSDs will be referred to simply as MSDs for the remainder of this paper.

The scale of this occupational health crisis is immense, imposing staggering costs on employees, employers, and the healthcare system. According to the U.S. Bureau of Labor Statistics, MSDs consistently account for the largest proportion of non-fatal workplace injuries requiring days away from work, representing a substantial 39% of all such injuries in 2022 [2]. These injuries, primarily sprains, strains, and soreness, impose a significant and escalating economic burden. A recent report by the National Safety Council indicated that workplace MSDs resulted in direct costs exceeding $50 billion [3], with the average compensation cost for a single strain injury reaching $37,185 [4]. This latter figure only accounts for direct costs, such as medical expenses and workers’ compensation claims. When factoring in indirect costs—including productivity losses, administrative overhead, recruitment and training of replacement workers, and reduced morale—the financial strain is estimated to be substantially higher. According to the Liberty Mutual Insurance Workplace Safety Index, U.S. businesses spent more than $1 billion per week for a staggering total of $58 billion per year [5].

Musculoskeletal disorders (MSDs) represent a significant public health challenge in the United States, with societal implications that extend well beyond their immense financial cost. MSDs are a substantial burden on the U.S. workforce, contributing to numerous injury cases where work-related incidents result in a median of 14 days away from work—four days more than the average for all work injuries combined [2,6]. The Centers for Disease Control and Prevention has consistently underscored that the combined effect of these losses severely strains employers and insurers alike, emphasizing the urgency of effective risk mitigation [7]. Compounding this issue is the staggering cumulative loss of human capital, highlighted by the fact that low back pain was the leading cause of years lived with disability in the U.S. in 2021, with other musculoskeletal disorders, neck pain, osteoarthritis, and rheumatoid arthritis following closely behind [8].

1.1. MSD Prevalence Across Diverse U.S. Sectors

The scope of the MSD challenge is comprehensive, touching every industrial sector of the economy. This research considers the incidence of MSDs across all major U.S. industry sectors to capture the full breadth of occupational risk exposure. While MSDs affect every sector, their acute concentration in physically demanding industries serves as a critical example of the need for targeted analysis.

Crucially, the risk is not limited to these heavy trades. MSDs affect a wide range of occupations across multiple sectors, including manufacturing, healthcare, and transportation. Laborers, helpers, and mechanics, for instance, report high rates of MSDs, with injuries frequently affecting the lower back, shoulders, knees, and neck due to heavy lifting, twisting, and prolonged standing or kneeling [9]. Furthermore, the transportation sector is also significantly impacted, particularly among professional drivers. Out of 22 studies included in a meta-analysis, for example, the overall prevalence of MSDs among bus drivers was 73.9% with occurrence by body region as follows: lower back (50.2%), neck (39.9%), shoulder (38.7%), upper back (32.4%), knee (31.7%), foot (28.3%), hip/thigh (14.9%), hand (14.7%), and elbow (9.46%) [10].

1.2. The Challenge of Generalized Intervention Across Diverse Sectors

Despite the clear correlation between physical demands and injury, traditional ergonomic interventions—which often rely on generalized training, static risk assessments, and blanket process changes—have shown limited effectiveness in mitigating the rising MSD incidence across the economy. This shortfall is largely attributable to the vast diversity and lack of standardization across industry worksites and job roles. An intervention effective in a controlled manufacturing environment, for example, is often irrelevant or impossible to implement in a decentralized service industry or a dynamic construction site. Furthermore, the interplay of psychosocial factors unique to each industry, such as job stress, high perceived job demands, and organizational constraints, acts as a significant contributing factor that compounds physical exposures [11].

The failure of generalized approaches has highlighted a critical research gap: the need for a more personalized, granular understanding of how specific job content and worker characteristics interact with physical risk exposures across the entire economic spectrum. The literature suggests that moving beyond broad, anecdotal risk assessments to utilizing objective, specific occupational descriptors is essential for developing effective prevention strategies [12]. Specifically, researchers are now advocating for leveraging digital and objective metrics to tailor interventions [13]. This includes the need to analyze specific characteristics that describe the conditions under which a job is performed—such as work styles, worker values, cognitive demands, and required training—rather than just the physical aspects of the job. Therefore, the inherent variability of MSDs across the major industry sectors demands intervention based on a comprehensive analysis of the complete occupational profile, rather than generalized, one-size-fits-all ergonomic training [14]. This shift in perspective, from viewing the problem as purely physical to understanding it as a consequence of misaligned work conditions and job demands, defines the core objective of this study.

1.3. Research Aims and Contribution

Given the pervasive impact of MSDs across the major U.S. industry sectors—and the identified limitations of current, generalized intervention strategies—understanding the relationship between detailed occupational characteristics and MSD incidence is essential for reducing their prevalence universally. This study utilizes granular data on occupations within the major U.S. industry sectors to examine how their multifaceted occupational characteristics—including work activities, job context, work styles, values, and educational and training requirements—influence the occurrence of MSD events. By integrating detailed occupational profiling data with injury outcomes across the entire economy, this research aims to move beyond traditional risk factor analysis. The findings will inform policy decisions, enhance the specificity of workplace interventions, and contribute to the development of more effective, personalized safety protocols, ultimately aiming to enhance worker safety and well-being by identifying key, modifiable determinants of MSD risk.

2. Materials and Methods

This study employed a quantitative, observational research design using secondary data analysis to investigate the extent to which occupational characteristics (e.g., activities, context, styles, job educational/training requirements) associated with select occupations across U.S. sectors influenced the prevalence of musculoskeletal disorders (MSDs). To do so, this study leveraged publicly available data collections from federal agencies that maintain ongoing, annual updates on occupations in the U.S., and the prevalence of MSDs occurring from tasks performed by workers in the workplace. For this study, the research Null Hypothesis (H₀) was as follows:

There is no statistically significant relationship between occupational descriptors (e.g., activities, context, styles, values) or moderating variables (e.g., MSD outcome, MSD symptoms) and the frequency of musculoskeletal disorder (MSD) events in select U.S. sector occupations.

For testing the hypothesis, this section outlines the data sources, measures, research design, and analytical procedures used in carrying out the proposed study.

2.1. Data Sources

The MSD data for this study were sourced from the U.S. Bureau of Labor Statistics’ Survey of Occupational Injuries and Illnesses (SOII), which provided counts of MSD ergonomic-related reported injuries [15]. These counts were organized at various industry levels based on the North American Industry Classification System (NAICS), a standard used by federal statistical agencies to categorize business establishments for data collection and analysis. From the data, industry groups identified by four-digit NAICS codes were selected because their narrative descriptors aligned most closely with occupational-level analysis.

Another data source for this study was the U.S. Department of Labor’s Occupational Information Network [16]. O*NET data included detailed narrative descriptors for each U.S. occupation (e.g., tasks and responsibilities), a categorical variable (Jobzone) indicating the education level required for the occupation, and 43 descriptors that defined its activities, context, styles, and values [17,18,19,20]. Additionally, the dataset included each occupation’s unique federal 7-digit Standard Occupational Classification (SOC) code, which classifies workers into occupational categories.

A third data source used In this study consisted of annual wages and total employment figures by SOC code and was obtained from the U.S. Bureau of Labor Statistics [21].

2.2. Measures

Using SAS® 9.4 procedures, the SOII and O*NET data were initially matched based on shared keywords in their NAICS/SOC narrative descriptors. Duplicate matched records with identical NAICS/SOC codes were deleted resulting in a dataset of 1,016 unique observations. Each matched SOII/O*NET observation was further processed by unpacking and then recombining the data to create a file in which each record included a variable indicating the count of MSD reported injuries (MSD_events). During this process, two moderating variables were also developed for the current study based on the SOII survey [15]. Derived from definitions provided by the National Safety Council [22], the first moderating variable (MSD_outcome) was coded 0 if the MSD events resulted in a job transfer or restriction (DJTR); and coded as 1 if the MSD events resulted in days away from work (DAFW). The second moderating variable (MSD_symptoms) was coded 0 if the MSD events represented injuries resulting in soreness or pain; and coded as 1 if the MSD events represented injuries resulting from sprains, strains, or tears. Based on this process, a final data set was produced consisting of 4,064 observations for analysis. The reference group—where MSD_outcome and MSD_symptoms both equaled 0—consisted of MSD events that resulted in a job transfer or restriction due to soreness or pain from work-related injuries.

Each observation record in the recombined data file also included measures on a progressive scale ranging from zero (e.g., no occurrence) to 100 (e.g., continual occurrence) for each of the 43 descriptors [23] that defined an occupation’s activities, context, styles, and values. Additionally, each record included a variable (Jobzone_Adjusted), whose original values from 1-5 were adjusted for the regression model with a corresponding numeric value from 0 to 4, indicating the education level required for the occupation: 0 = no high school diploma, 1 = high school diploma or GED, 2 = associate degree, 3 = bachelor’s degree, and 4 = master’s degree or higher [24].

2.3. Research Design

This study used the following regression model where the dependent variable was the natural logarithm of MSD counts (i.e., MSD_events), which corrected for skewness in the count distribution.

ln(MSD_events) = β₀ + β₁X_1…+…β_nX_n + β_n+1X_n+1 +

c₁MSD_Outcome + c₂MSD_Symptoms + c₃Jobzone_Adjusted + e

The intercept term, β₀, defined the regression constant. The X predictor variables consisted of the 43 descriptor measures that defined an occupation’s activities context, styles, and values. For a select descriptor predictor variable, its regression estimate b multiplied by 100 defined the approximate percentage change in MSD events in natural form for a one-unit change in the predictor, after holding all other covariates constant. Thus, if a regression parameter estimate b was positive, the percentage resulted in an increase in MSD events; if negative the percentage resulted in a decline of MSD events. Random errors were captured in the term e [25].

The c₁ and c₂ estimates for the MSD_outcome and MSD_symptoms dichotomous variables, respectively, represent the effect of each categorical condition relative to the reference group, while holding all other predictors constant. As defined earlier, the reference group—where MSD_outcome and MSD_symptoms both equal 0 (zero)—consisted of MSD events that resulted in a job transfer or restriction due to soreness or pain from work-related injuries. In the regression model, these estimates reflected the percentage change in MSD events as a shift in the intercept (β₀) caused by belonging to a non-zero coded dichotomous category relative to the reference group, after controlling for effects of other predictors.

When applied to the regression results, the MSD reference group’s predicted y equaled exp(β₀), i.e., the exponentiated value of the regression intercept. When the MSD_outcome variable changed from 0 to 1, the percentage change in the predicted y value equaled [exp(c₁) – 1] x 100. Similarly, when the MSD_symptoms variable changed from 0 to 1, the percentage change in the predicted y value equaled [exp (c₂) – 1] x 100. When both MSD_outcome and MSD_symptoms changed to values of one, the percentage change in the predicted y value equaled [exp (c₁ + c₂) – 1] x 100.

Similarly, the c₃ estimate for the Jobzone_Adjusted categorical variable represented the percentage change in MSD events as a shift in the intercept (β₀) caused by belonging to a non-zero coded adjusted category relative to the reference group (where Jobzone_Adjusted = 0), after controlling for effects of other predictors. To estimate the percentage change in the predicted y value from an adjusted Jobzone value of 0 to a higher category level, t, the calculation equals [exp (c3 x t) – 1] x 100, where t equals the education levels 1, 2, 3 or 4.

2.4. Analytic Procedures

Given the large number of descriptor variables available for analysis, the first analytical procedure consisted of running the research model as a stepwise regression. Stepwise regression is a systematic method for building a regression model by iteratively selecting the most statistically significant predictor variables. In so doing, the result is a subset of explanatory variables that best predict the dependent variable thereby balancing model complexity with predictive power. For this study, the process first involved permanently entering the MSD_symptoms and MSD_outcome dichotomous variables as model moderators. Subsequently, a stepwise selection method was applied to the adjusted Jobzone categorical variable and the 43 descriptor variables, utilizing an entry criterion of p ≤ .25 and a retention criterion of p ≤ .05. At each step, the model re-evaluated the results to ensure that only variables that contributed significantly to the predictive power remained, after controlling for the moderator variables. The second analytical procedure consisted of running the regression model with explanatory variables identified as statistically significant in the stepwise regression, along with the moderator variables. In addition to these regression models, the current study used descriptive statistical procedures to analyze the data.

3. Results

Given the large number of descriptor variables available for analysis, the first analytical procedure consisted of running a stepwise regression. In stepwise regression, the F-value statistic determines whether a predictor should be added to or removed from the model based on its contribution to explaining the variance in the dependent variable. In doing so, it measures the strength of the relationship between a predictor and the dependent variable, accounting for the other predictors already in the model. A large F-value suggests the predictor explained a significant amount of variance in the dependent variable, relative to the error variance. A small F-value indicated that the predictor contributed little to the model and may not be statistically significant. Here, the Pr > F statistic refers to the p-value associated with the statistical significance of the F-statistic for a predictor variable. In this manner, the Pr > F statistic represents the probability of observing an F-value as extreme as (or more extreme than) the one calculated, assuming a true null hypothesis.

The Jobzone (adjusted) categorical variable and a total of 19 MSD descriptor variables did not meet the retention threshold (p < .05) during the stepwise regression procedure. These non-significant predictors are shown in Table 1, alphabetized by variable name for easier reference.

In comparison, Table 2 provides descriptions, means and standard deviations for the remaining 24 MSD predictors—alphabetized by their variable name and assigned a letter for easier reference. After 31 iterative steps, these predictors’ F-values resulted in statistically significant Pr > F (p < .05) values and thereby, qualified for further regression analysis.

Table 3 shows the results from this subsequent regression analysis for the 24 MSD predictors that met the p<.05 threshold in the stepwise regression procedure. The table shows the predictors organized in two groups: MSD Risk Amplifiers and MSD Risk Mitigators. For MSD Risk Amplifiers, the positive predictor estimates were sorted in descending value to highlight those occupational characteristics with the most to the least effect on increasing MSD occurrences.

For MSD Risk Mitigators, the negative predictor estimates were sorted in ascending order to highlight those occupational characteristics with the most to the least effect in decreasing MSD occurrences. The predictor’s reference letter (e.g., a, b, c, etc.) associated with its description in Table 2 displays in the left-most table column.

Predictors under MSD Risk Amplifiers had positive estimate signs, which indicated that a one-unit change in the predictor resulted in an approximate b percentage increase in MSD events in natural form, after holding all other covariates constant. In comparison, predictors under MSD Risk Mitigators had negative estimate signs, which indicated that a one-unit change in the predictor resulted in an approximate b percentage decrease in MSD events in natural form, after holding all other covariates constant.

Note that all 24 predictors were statistically significant (p<.05 or less). The right-most column shows the estimates in exponentiated form, which when multiplied by 100 results in the percentage change per unit increase (e.g., accuracy_import, 0.0225 x 100 = 2.25%). For the MSD reference group where both MSD_outcome and MSD_symptoms equaled 0, the predicted y was approximately 29.1 MSD events, i.e., the exponentiated value of the regression intercept (3.3697). When the MSD_outcome variable changed from 0 to 1, the percentage change in the predicted y value was about 38.5 percent, i.e., the exponentiated value of the c₁ estimate (0.3258) minus 1 times 100. Similarly, when MSD_symptoms variable changed from 0 to 1, the percentage change in the predicted y value was 115.6 percent, i.e., the exponentiated value of the c₂ estimate (0.7682) minus 1 times 100. When both MSD_outcome and MSD_symptoms changed from 0 to 1, the percentage change in the predicted y value calculated to 198.6 percent, i.e., the exponentiated value of c₁ and c₂ estimates combined (1.0940 = 0.3258 + 0.7682) minus 1 times 100. Given the predictor and dichotomous variables, the regression model’s adjusted R-square explained only about one-quarter (0.2370) of the proportion of the variance in the MSD events dependent variable, adjusting for the number of predictors.

Although the Jobzone_Adjusted category variable was not statistically significant (p = .4776) in the stepwise regression for inclusion in the final model, Table 4 presents the distribution of MSD events by Jobzone level, stratified by MSD outcome and symptom category, across the 1,016 unique SOC/NAICS observations. Three key findings emerged from these results. First, of the 1,016 occupations, nearly two-thirds (n=651) were classified as requiring only a high school diploma or GED. Second, of the 5,289,670 MSD injuries, approximately 56.6% (n=2,995,300) were associated with this jobzone level. Third, events resulting from sprains, strains, or tears accounted for nearly two-thirds of the MSD events (67.3%), resulted in either days away from work (35.8%) or job transfers/restrictions (31.5%, see Event Totals row). Conversely, the remaining MSD events (32.6%), attributed to soreness or pain, resulted in a smaller proportion of days away from work (21.3%) and job transfers/restrictions (11.3%).

Table 5 presents the MSD risk per 100 employed across select industrial sectors in the data. For context, the NAICS column indicates the number of associated NAICS codes included in each sector. The MSD risk per 100 employed for an industrial sector was calculated by first dividing the number of MSD events in that sector by its total employed workforce (BLS, 2024). This estimate was then multiplied by 100 to normalize the risk across all industrial sectors.

While the Life, Physical, and Social Science (#19) sector had the highest individual risk estimate (70.5 MSD events per 100), four specific sectors were responsible for the vast majority of the total burden, accounting for 86.4 percent of the total MSD cases (n=5,289,670) and 50.7% of the total employed workforce (n=22,760,390) represented in the data. These high-volume MSD industrial sectors include:

▪

Healthcare Support (#31): 42.0 MSD events per 100 employed

▪

Construction and Extraction (#47): 32.0 MSD events per 100 employed

▪

Production (#51): 36.9 MSD events per 100 employed

▪

Transportation and Material Moving (#53): 55.4 MSD events per 100 employed

The influence of these four sectors underscores the significant challenge of MSDs across U.S. industries, as 80.3 percent (n=816) of the 1,016 NAICS codes included in the data were associated with these select industrial groups. Collectively, these four sectors represented a combined risk estimate of 39.6 MSD events per 100 employees—a staggering magnitude of over six times the risk of 6.4 MSD events per 100 employees of all other industrial sectors combined in the table.

4. Discussion

The current study offered critical, model-derived evidence on the occupational determinants of Musculoskeletal Disorders (MSDs), emphasizing the direct role of job-related factors in shaping workplace risk profiles. By analyzing secondary data through a stepwise regression approach, the research identified 24 robust predictors that were statistically significant in explaining MSD risk, which we categorized based on their correlation as either MSD Risk Amplifiers (positive association) or MSD Risk Mitigators (negative association). These two categories form the foundation of our proposed dual-focus safety strategy. The 24 work factor descriptors were then strategically grouped into the following three distinct, manageable themes—Biomechanical, Environmental, and Cognitive/Administrative—to facilitate a more structured and comprehensive discussion of the results. This thematic classification, further stratified by the Amplifier/Mitigator effect, allows us to move beyond simple correlation reporting and focus directly on the practical implications of managing distinct categories of workplace hazards for effective prevention.

4.1. Biomechanical Risk Factors: Amplifiers and Mitigators

The analysis of physical demands revealed a clear distinction between factors that amplify MSD risk and those that appear to mitigate it.

4.1.1. Amplifiers of Biomechanical Risk

Findings overwhelmingly confirmed that awkward postures and dynamic loads act as primary MSD amplifiers. For instance, more time spent bending or twisting the body and climbing ladders, scaffolds, or poles were strongly associated with a higher incidence of MSDs. These results are consistent with the established literature, necessitating continued focus on engineering controls to eliminate these static and dynamic postural strains [26]. The positive association with more time spent walking or running suggests that cumulative impact forces on the lower kinetic chain also contribute significantly to overall MSD burden.

Additionally, the analysis revealed a positive association between the more frequent use of common protective or safety equipment (e.g., safety shoes, glasses, gloves) and higher MSD incidence. This unexpected finding suggests that Personal Protective Equipment (PPE) is not always a neutral intervention; poorly designed or ill-fitting equipment, such as heavy safety shoes or cumbersome gloves, can introduce new or amplified ergonomic risks by increasing muscular effort, restricting natural movement, or forcing awkward postures [27].

4.1.2. Mitigators of Biomechanical Risk

Conversely, several biomechanical variables were identified as potential mitigators, correlating negatively with MSD incidence. The finding that more time spent sitting is associated with lower MSD incidence suggests that highly strenuous, active labor roles carry a disproportionately higher risk compared to sedentary tasks. Furthermore, the inverse relationship observed for more time spent keeping or regaining balance may indicate that jobs demanding higher physical conditioning or requiring more deliberate, controlled movements are inherently protective. These results suggest that physical fitness and purposeful movement may be protective factors against general MSD development.

Intriguingly, the study found that more time spent making repetitive motions and more time spent repeating the same tasks were both associated with a lower incidence of MSDs. This result may reflect highly optimized ergonomic setups in repetitive manufacturing environments, where tasks are frequently refined and automated, or an organizational culture that effectively manages work pace and provides mandated micro-breaks [28].

Table 6. Summary of Biomechanical Risk Factors, Categorized by Effect.

Effect	Variable Definition	Estimate Interpretation
Amplifier	Spend Time Bending or Twisting Your Body	More time spent bending or twisting the body is associated with higher MSD incidence.
Amplifier	Spend Time Climbing Ladders, Scaffolds, or Poles	More time spent climbing ladders, scaffolds, or poles is associated with higher MSD incidence.
Amplifier	More Time Spent Walking or Running	More time spent walking or running is associated with higher MSD incidence.
Amplifier	Wear Common Protective or Safety Equipment such as Safety Shoes, Glasses, Gloves, Hearing Protection, Hard Hats, or Life Jackets	More frequent use of common protective or safety equipment (e.g., safety shoes) is associated with higher MSD incidence.
Mitigator	More Time Spent Sitting	More time spent sitting is associated with lower MSD incidence.
Mitigator	Time Spent Keeping or Regaining Balance	More time spent keeping or regaining balance is associated with lower MSD incidence.
Mitigator	Spend Time Making Repetitive Motions	More time spent making repetitive motions is associated with lower MSD incidence.
Mitigator	Importance of Repeating Same Tasks	More time spent repeating the same tasks is associated with lower MSD incidence.

Table 6. Summary of Biomechanical Risk Factors, Categorized by Effect.

Effect	Variable Definition	Estimate Interpretation
Amplifier	Spend Time Bending or Twisting Your Body	More time spent bending or twisting the body is associated with higher MSD incidence.
Amplifier	Spend Time Climbing Ladders, Scaffolds, or Poles	More time spent climbing ladders, scaffolds, or poles is associated with higher MSD incidence.
Amplifier	More Time Spent Walking or Running	More time spent walking or running is associated with higher MSD incidence.
Amplifier	Wear Common Protective or Safety Equipment such as Safety Shoes, Glasses, Gloves, Hearing Protection, Hard Hats, or Life Jackets	More frequent use of common protective or safety equipment (e.g., safety shoes) is associated with higher MSD incidence.
Mitigator	More Time Spent Sitting	More time spent sitting is associated with lower MSD incidence.
Mitigator	Time Spent Keeping or Regaining Balance	More time spent keeping or regaining balance is associated with lower MSD incidence.
Mitigator	Spend Time Making Repetitive Motions	More time spent making repetitive motions is associated with lower MSD incidence.
Mitigator	Importance of Repeating Same Tasks	More time spent repeating the same tasks is associated with lower MSD incidence.

4.2. Environmental Exposure: Amplifiers and Mitigators

The study of environmental variables provided nuanced insight into how the surrounding workplace conditions, both physical and administrative, influence MSD risk.

4.2.1. Amplifiers of Environmental Risk

Several factors were identified as potent amplifiers of MSD incidence. Exposure to contaminants (e.g., chemicals, dust) and disease or infections were both positively correlated with higher MSD rates. This suggests that environmental stressors that induce inflammation or stress on the body may act synergistically with physical demands to promote MSD development. Notably, exposure to indoor, environmentally controlled workspaces also emerged as an amplifier, likely reflecting the pervasive issue of poor static workstation ergonomics and the hazards of prolonged sedentary behavior within typical office settings. Finally, whole body vibration (a known physical risk) and working outdoors, exposed to all weather conditions confirmed their role as amplifiers, stressing the need for engineering controls (vibration reduction) and administrative controls (weather breaks/PPE).

In addition, a higher frequency of exposure to minor burns, cuts, bites, or stings was positively correlated with higher MSD incidence. This finding does not imply a direct physiological link but rather suggests a failure in hazard control systems. A workplace with a high incidence of minor injuries likely lacks a robust safety culture, poor housekeeping, and inadequate task-specific safety training, all of which indirectly contribute to a higher overall MSD risk profile [29].

4.2.2. Mitigators of Environmental Risk

Conversely, three factors were identified as mitigators, suggesting that while the risks they represent are severe, their presence correlates with effective safety systems. Higher exposure to radiation and distracting or uncomfortable noise levels were both associated with lower MSD incidence. This paradoxical finding is attributed to the compliance effect: work environments involving these high-consequence hazards are typically subject to strict regulation, robust engineering controls (e.g., shielding, noise dampening), and mandated personal monitoring. These rigorous safety protocols appear to have a broad, protective ergonomic benefit. Additionally, more work performed outdoors under cover was found to be a mitigator, highlighting the protective role of simple environmental shielding (e.g., from direct sun, rain) in reducing the environmental stress and fatigue that can lead to MSDs.

Table 7. Summary of Select Environmental Factors, Categorized by Effect.

Effect	Variable Definition	Estimate Interpretation
Amplifier	Exposed to Contaminants	Higher frequency of exposure to contaminants is associated with higher MSD incidence.
Amplifier	Exposed to Disease or Infections	Higher frequency of exposure to disease or infections is associated with higher MSD incidence.
Amplifier	Indoors, Environmentally Controlled	Higher exposure to indoor, environmentally controlled workspaces is associated with higher MSD incidence.
Amplifier	Outdoors, Exposed to All Weather Conditions	Higher exposure to outdoor, all-weather conditions is associated with higher MSD incidence.
Amplifier	Exposed to Whole Body Vibration	Higher exposure to vibration is associated with higher MSD incidence.
Amplifier	Exposed to Minor Burns, Cuts, Bites, or Stings	Higher exposure to minor burns, cuts, bites, or stings is associated with higher MSD incidence.
Mitigator	Exposed to Sounds, Noise Levels that are Distracting or Uncomfortable	Higher exposure to distracting or uncomfortable noise levels is associated with lower MSD incidence.
Mitigator	Outdoors, Under Cover	More work performed outdoors under cover is associated with lower MSD incidence.
Mitigator	Exposed to Radiation	Higher exposure to radiation is associated with lower MSD incidence.

Table 7. Summary of Select Environmental Factors, Categorized by Effect.

Effect	Variable Definition	Estimate Interpretation
Amplifier	Exposed to Contaminants	Higher frequency of exposure to contaminants is associated with higher MSD incidence.
Amplifier	Exposed to Disease or Infections	Higher frequency of exposure to disease or infections is associated with higher MSD incidence.
Amplifier	Indoors, Environmentally Controlled	Higher exposure to indoor, environmentally controlled workspaces is associated with higher MSD incidence.
Amplifier	Outdoors, Exposed to All Weather Conditions	Higher exposure to outdoor, all-weather conditions is associated with higher MSD incidence.
Amplifier	Exposed to Whole Body Vibration	Higher exposure to vibration is associated with higher MSD incidence.
Amplifier	Exposed to Minor Burns, Cuts, Bites, or Stings	Higher exposure to minor burns, cuts, bites, or stings is associated with higher MSD incidence.
Mitigator	Exposed to Sounds, Noise Levels that are Distracting or Uncomfortable	Higher exposure to distracting or uncomfortable noise levels is associated with lower MSD incidence.
Mitigator	Outdoors, Under Cover	More work performed outdoors under cover is associated with lower MSD incidence.
Mitigator	Exposed to Radiation	Higher exposure to radiation is associated with lower MSD incidence.

4.3. Cognitive and Administrative Factors: Amplifiers and Mitigators

The study reveals that organizational and cognitive factors exert a profound influence on MSD risk, demonstrating that the design of the work process and social environment is as critical as physical ergonomics.

4.3.1. Amplifiers of Cognitive and Administrative Risk

The factors identified as amplifiers underscore the link between mental stress, cognitive load, and static physical tension. Higher importance of being exact or accurate and higher frequency of decision-making were both positively correlated with MSD incidence. This suggests that the intense mental focus required for these tasks leads to prolonged static muscle loading in the neck, shoulders, and back, transforming cognitive strain into a physical risk factor. Additionally, high time pressure acts as an amplifier by encouraging faster, more forceful movements and limiting the opportunity for self-initiated recovery breaks [30]. Finally, greater physical proximity to other people in the workspace emerged as a subtle but significant amplifier, potentially increasing psychological stress, restricting natural movement, or forcing awkward postures to avoid collision [31].

4.3.2. Mitigators of Cognitive and Administrative Risk

The mitigators in this category present the most complex, yet potentially valuable, findings, often reflecting a "compliance paradox". The association between a longer work week and lower MSD incidence, and a similar inverse relationship for exposure to hazardous equipment, suggests that these roles are characterized by compensating controls. Jobs with overtly hazardous equipment (Mitigator) often enforce stringent safety protocols, superior engineering controls, and robust training, which inadvertently reduces overall ergonomic risk [32]. Similarly, jobs with longer work weeks may offer higher worker autonomy, lower physical intensity, or less organizational pressure, acting as protective administrative factors [33].

The inverse correlation between the higher consequence of errors and lower MSD incidence further supports this compliance paradox. In roles where an error has severe financial or safety outcomes (e.g., surgical technician, control room operator), organizations mandate a culture of meticulousness, strict protocol adherence, and slow, deliberate movements [34].

This focus on "doing it right" translates into a protective ergonomic environment, as it limits the fast, forceful, and risky movements often associated with high MSD incidence. These findings suggest that implementing the rigorous administrative controls and safety culture inherent in high-hazard roles can serve as a model for holistic MSD prevention across all job types [35].

Table 8. Summary of Cognitive and Administrative Factors, Categorized by Effect.

Effect	Variable Definition	Estimate Interpretation
Amplifier	Importance of Being Exact or Accurate	Higher importance of being exact or accurate is associated with higher MSD incidence.
Amplifier	Frequency of Decision Making	Higher frequency of decision-making is associated with higher MSD incidence.
Amplifier	Time Pressure	Higher time pressure is associated with higher MSD incidence.
Amplifier	Physical Proximity	Greater physical proximity to other people in the workspace is associated with higher MSD incidence.
Mitigator	Exposed to Hazardous Equipment	Higher exposure to hazardous equipment is associated with lower MSD incidence.
Mitigator	Duration of Typical Work Week	A longer work week is associated with lower MSD incidence.
Mitigator	Consequence of Error	Higher consequence of errors is associated with lower MSD incidence.

Table 8. Summary of Cognitive and Administrative Factors, Categorized by Effect.

Effect	Variable Definition	Estimate Interpretation
Amplifier	Importance of Being Exact or Accurate	Higher importance of being exact or accurate is associated with higher MSD incidence.
Amplifier	Frequency of Decision Making	Higher frequency of decision-making is associated with higher MSD incidence.
Amplifier	Time Pressure	Higher time pressure is associated with higher MSD incidence.
Amplifier	Physical Proximity	Greater physical proximity to other people in the workspace is associated with higher MSD incidence.
Mitigator	Exposed to Hazardous Equipment	Higher exposure to hazardous equipment is associated with lower MSD incidence.
Mitigator	Duration of Typical Work Week	A longer work week is associated with lower MSD incidence.
Mitigator	Consequence of Error	Higher consequence of errors is associated with lower MSD incidence.

4.4. Limitations and Assumptions

While this study provides valuable insights into the occupational determinants of MSDs, several limitations and assumptions must be considered to ensure a clear and accurate interpretation of the findings.

A key limitation resides in the reliance on secondary data sources, such as the SOII and the O*NET. While widely used, these data collections rely on self-reported or organizationally reported information, which may lead to inaccuracies due to underreporting or misclassification of MSD events. The integration of SOII and O*NET datasets was based on matching keywords in narrative NAICS descriptors and SOC codes, which assumed uniformity in classification across sources. Despite efforts to minimize such discrepancies, classification errors remain a limitation.

The regression model operated under specific assumptions, including the linearity of relationships between predictor variables and the dependent variable (natural logarithm of MSD events), as well as the independence and normality of errors. Any violations of these assumptions, such as nonlinearity, multicollinearity among predictors, or heteroscedasticity, could affect the reliability of the findings. Additionally, stepwise regression, while useful for variable selection, has been critiqued for its potential to overfit the data and its sensitivity to sample-specific patterns, which may limit the generalizability of the results [36]. Future researchers seeking to mitigate these known limitations of stepwise regression should explore variable selection alternatives like ridge regression or its variant Lasso regression to analyze the NAICS/SOC data and improve model generalizability [37].

The study also assumed that the progressive O*NET descriptor scales (ranging from 0 to 100) accurately reflected the underlying occupational activities, contexts, and demands. Finally, the study assumed that the identified predictors remained stable over time and were not influenced by external factors, including changes in workplace safety regulations, economic conditions, or technological advancements.

5. Conclusions: A Dual-Focus Strategy for MSD Prevention

This comprehensive analysis of 24 distinct work factors reveals that the effective control of Musculoskeletal Disorders (MSDs) requires a fundamental shift toward a dual-focus safety strategy. The traditional ergonomic model, while necessary, is insufficient on its own. Our findings necessitate an integrated approach that simultaneously targets the elimination of risk amplifiers and the reinforcement of protective mitigators.

5.1. Eliminating Risk Amplifiers Through Engineering and Administration

The vast majority of positively correlated factors, i.e., Amplifiers—represent direct causes or compounding stressors of MSDs. These range from established physical hazards like bending or twisting and whole body vibration to newly quantified cognitive and environmental risks, such as high requirements for exactness/accuracy and exposure to contaminants. These risks demand controls based on the hierarchy of controls. Policy must mandate the elimination of these amplifiers at the source. This requires, first, engineering controls, such as redesigning workstations to eliminate awkward postures, automating highly precise manual tasks, and investing in low-vibration tools. Second, it requires administrative controls, including implementing mandatory micro-breaks for high-cognitive-load roles, establishing task rotation schedules to prevent static loading, and enforcing rigorous work-rest cycles for exposure to extreme weather. Failing to eliminate these amplifiers is equivalent to accepting chronic, cumulative trauma as an inevitable part of the job.

5.2. Reinforcing Mitigators Through Culture and Training

Perhaps the most valuable contribution of this study is the identification of key Mitigators—factors that, counterintuitively, correlate with a lower incidence of MSDs. These include exposure to high-consequence hazards like radiation or hazardous equipment, and even frequent repetitive tasks. These inverse relationships reveal a crucial compliance paradox: where risk is perceived as high, the resultant stringent safety protocols—superior training, strict procedure adherence, mandated maintenance, and detailed supervision—provide an accidental, holistic ergonomic benefit. Our primary recommendation is therefore to reinforce and generalize the cultural benefits of these mitigators across all job roles. This involves adopting a high-hazard safety model, which means applying the rigorous safety culture and detailed protocol adherence found in high-consequence industries (e.g., nuclear, heavy manufacturing) to lower-hazard settings (e.g., office, light assembly). Furthermore, it requires enhancing training and wellness by integrating proactive wellness programs that build resilience against MSDs, for example, by reinforcing the benefits derived from factors like time spent keeping or regaining balance through targeted movement and balance training programs.

In summary, the next generation of MSD prevention must move beyond simply addressing the injury after it occurs. By actively eliminating—or at least reducing significantly the Amplifiers and systematically incorporating the protective structure and cultural discipline associated with the Mitigators, organizations can build a resilient, ergonomically sound work environment that supports both physical and cognitive well-being. The specific policy implications and practical steps for this integrated strategy are provided in the Appendix A table.