Introduction

Work-related musculoskeletal disorders (WRMDs) are a major cause of disability worldwide [1, 2]. WRMDs involving the spine are considered one of the prevalent and costly non-malignant health conditions affecting the working-age population [1, 3, 4]. In Quebec, work-related musculoskeletal injuries (WRMIs) account for approximately 50% of all time-loss claims [5]. Although, the majority of individuals who sustain a WRMI recover without complication, many will remain work-disabled for prolonged periods of time [6–8]. Of interest in this area of research has been to identify the factors or variables that distinguish between injured workers who will recover and those who will remain occupationally disabled.

Numerous studies have reported significant sex differences in recovery outcomes in individuals who have sustained disabling injuries [9–11]. Recent investigations have reported that women with WRMIs have a significantly longer duration of work absences than men with WRMIs [12]. Research has also indicated that women, compared to men, are more likely to experience multisite pain following musculoskeletal injury and show slower recovery than men from musculoskeletal injuries [13]. Keogh et al. [14] reported that women showed less clinical improvement than men following participation in a multidisciplinary pain management program. Additionally, women were less likely than men to maintain treatment gains when re-assessed at 3-month follow-up. Several investigations have reported that women show less physical improvement than men following participation in rehabilitation interventions for musculoskeletal injuries, and are less likely than men to return to work (RTW) [9, 10].

Research has also revealed significant sex differences in post-traumatic stress symptoms (PTSS) following exposure to traumatic events. The term PTSS is used to refer to the presence of symptoms of post-traumatic stress disorder (PTSD) that do not meet all diagnostic criteria for PTSD [15]. In response to a wide range of traumatic events, women have been shown to be more susceptible to developing PTSS/PTSD than men, with the prevalence of PTSD being twice as high in women compared to men [16, 17]. Though odds ratios vary, sex differences in PTSD prevalence rates have been shown to be robust across differences in trauma type, diagnostic criteria, and study methodology [18].

Recent research points to higher rates of PTSS in women in response to workplace stresses and/or injury. Workplace exposure to trauma can occur in the context of high stress occupations such as health care workers, first responders, or military members. Research suggests that across several high stress occupations, the prevalence and severity of PTSS is higher in women than in men [19]. For example, in a sample of health care professionals in emergency departments, Carmassi et al. [20] observed significantly higher levels of PTSS in women than in men. In a recent study, Pavilanis et al. [21] reported that women scored higher than men on a measure of PTSS following a workplace musculoskeletal injury.

It is possible that PTSS may act as a mediator of sex differences in recovery outcomes following musculoskeletal injury. In light of the research showing that PTSS are more prevalent in women following workplace injury, PTSS might add to the burden of musculoskeletal injury, and in turn, impact negatively of recovery potential [22]. Research examining the processes underlying sex differences in recovery outcomes might help identify key targets for psychosocial interventions designed to improve recovery trajectories following work injury.

The primary objective of the present study was to examine whether PTSS mediate sex differences in recovery outcomes in individuals receiving treatment for WRMIs. Participants completed measures of pain, disability and PTSS at admission to a physical therapy intervention. Pain and disability were re-assessed at treatment termination and RTW was assessed 12-months following treatment termination.

Materials and methods

Results

PTSS as mediator of sex differences in recovery outcomes

Mediational analyses were conducted for treatment-related disability reduction (T1 PDI – T3 PDI) and RTW. Results of the first mediational analysis confirmed the hypothesis that PTSS mediated the relation between sex and treatment-related disability reduction. The total effect for the model was significant (R² = 0.18, P < 0.001). As shown in Figure 1, sex was significantly associated with PTSS (Path a; β = –8.5, P < 0.01) and PTSS was significantly associated with treatment-related disability reduction (Path b; β = –0.17, P < 0.001). The bias-corrected bootstrap 95% CI for the indirect effect did not include zero (Path c; β = 1.5, 95% CI = 0.18, 3.2), confirming that PTSS mediates the influence of sex on disability reduction. The results indicated that the direct effect of sex on disability reduction remained significant (Path c’; β = 5.4, P < 0.01) when controlling for PTSS, suggesting partial mediation.

Display full size

Figure 1. 

Mediation model of sex on disability reduction by PTSS. Beta (β) coefficients are unstandardized. PTSS: post-traumatic stress symptoms; Sex: sex of participant coded as women (0) men (1); Disability Reduction = PDI (T1) – PDI (T3). ^** P < 0.01, ^*** P < 0.001

The results of the second mediational analysis revealed that PTSS also mediated the relation between sex and RTW. The Nagelkerke pseudo R² indicated that the final model accounted for approximately 16% of the total variance, P < 0.001. As shown in Figure 2, sex was significantly associated with PTSS (Path a; β = –8.5, P < 0.001) and PTSS was significantly associated with RTW (Path b; β = –0.02, P < 0.01). The bias-corrected bootstrap 95% CI for the indirect effect did not include zero (Path c; β = 0.20, 95% CI = 0.002, 0.50), confirming that PTSS mediates the influence of sex on RTW. The results indicated that the direct effect of sex on RTW was no longer significant (Path c’; β = 0.69, P < 0.12) when controlling for PTSS, suggesting full mediation.

Display full size

Figure 2. 

Mediation model of sex on RTW by PTSS. Beta (β) coefficients are unstandardized. PTSS: post-traumatic stress symptoms; Sex: sex of participant coded as women (0) men (1); RTW: return to work. ^* P < 0.05, ^** P < 0.01

Discussion

Study findings were consistent with previous research showing that women report more severe pain than men, show more modest reductions in disability, and are less likely than men to RTW following treatment [14, 21]. Numerous investigations have revealed that women, compared to men, experience more severe pain, more frequent pain, and have a poorer prognosis for recovery [40–43]. Previous research also shows that women make fewer gains than men following participation in rehabilitation interventions and are less likely than men to maintain treatment gains [9, 14, 44]. Significant sex differences in disability outcomes have been reported even when controlling for pain severity and depression [8, 10, 41]. In other words, although women report more severe pain and depression than men following musculoskeletal injury, available research suggests that pain and depression are not the factors that account for sex differences in disability outcomes.

Previous research has also revealed significant sex differences in PTSS following injury or trauma [45, 46]. In a recent longitudinal study of patients admitted to a trauma centre, women were more likely than men to screen positive for PTSD and showed more pronounced long term functional limitations [47]. Pavilanis et al. [21] found that women reported more severe PTSS than men in the weeks immediately following a WRMI. The findings of the present study also showed that women obtained significantly higher scores than men on a measure of PTSS. On the basis of recommended cut-scores for the IES-R, 18% of women, compared to 9% of men in the present study, scored above threshold for clinically significant PTSS. These data suggest that there is a high prevalence of clinically significant PTSS in the early weeks post-injury, particularly in women.

The findings join a growing body of literature suggesting that debilitating injuries or illnesses can give rise to clinically significant levels of PTSS, even though the triggering event might not satisfy DSM-5 Criterion A (defining features of a trauma) for a diagnosis of PTSD [48–50]. In the DSM-5, the “event” criterion for a diagnosis of PTSD is quite broad, encompassing “exposure to actual or threatened death or serious injury” [51]. It is a challenging task to define “actual or threatened serious injury” in objective terms since degree of “threat” is a subjective appraisal. A musculoskeletal injury resulting in pain and limitations sufficient to comprise individuals’ ability to perform their occupational activities could be appraised by many as a serious injury.

Recent studies have yielded findings suggesting that health conditions associated with persistent pain are particularly likely to contribute to elevated levels of PTSS, even though the onset of the conditions would not necessarily be considered “traumatic” [52]. For example, significant relations between pain and PTSS has been reported in several groups including individuals with musculoskeletal pain [48], chronic pain [53, 54], arthritis [55], fibromyalgia [49], cancer [50], and burns [56]. The experience of persistent pain may act as a vulnerability factor for PTSS following occupational injury [57].

Bootstrap mediation analyses were conducted for treatment-related reductions in disability and RTW. Mediation was not assessed for treatment-related reductions in pain severity given that the sex × time interaction for treatment-related changes in pain severity was not significant. Mediation was supported for the role of PTSS in accounting for sex differences in disability reduction and RTW. These results suggest that heightened levels of PTSS in women with WRMIs partially explain sex differences in recovery outcomes following disabling work injury.

Research points to a number of factors that might be implicated in the mediational role of PTSS in the relation between sex and recovery outcomes. As noted earlier, the findings of numerous studies show that women are more likely than men to develop PTSS following exposure to trauma [18]. In the context of occupational injury, women’s smaller physical size and strength might contribute to heightened threat appraisals of the possible negative consequences of physical injury. Women typically score higher than men on psychosocial correlates of PTSS such as catastrophic thinking or fear of re-injury, raising the possibility that these variables might also account for the relation between sex and recovery outcomes [58]. There are also important sex differences in role demands where women are more likely than men to assume a disproportionate share of household and caregiving responsibilities [59, 60]. The prospect of disruption in role functioning might also foster heightened threat appraisals of the consequences of injury. Women have also been shown to be more likely to engage in rumination in stressful situations, which could in turn, heighten susceptibility to the development of PTSS [61].

The results of previous studies suggest that PTSS adds to the burden of disability associated with injury [21, 52, 62]. PTSS can interfere with full participation in family, social, recreational, and occupational activities [21, 63, 64]. Avoidance is a central feature of the experience of PTSS where individuals are motivated to avoid situations or activities that might trigger memories of the trauma [18, 20]. When trauma exposure has occurred in the workplace, the workplace might become a key object of avoidance and contribute to prolonged work absence. It is also possible that elements of PTSS, such as avoidance, might compromise full engagement in physical therapy interventions, and ultimately reducing the benefit the individual might derive from treatment.

Research has accumulated indicating that individuals with musculoskeletal conditions that have co-morbid mental health problems are likely to face a more prolonged and problematic course of recovery [65, 66]. Pain-related musculoskeletal conditions with co-morbid mental health problems are associated with greater health care costs and longer periods of work absence than musculoskeletal conditions without co-morbid mental health problems [65, 67]. Investigators have called for greater attention to assessing current and previous mental health in individuals who have sustained occupational injuries [68].

Screening measures for PTSS are not routinely administered in primary care, particularly in cases of musculoskeletal injury since the context of the injury might not be considered traumatic. Consequently, PTSS following musculoskeletal injury are likely to go undetected and untreated [69]. In light of research findings pointing to the potential mutual maintenance of symptoms of pain and PTSD left untreated, PTSS may persist over extended periods of time, compromising recovery of the WRMI [70, 71].

Screening tools such as the IES-R or PCL-5 might be usefully incorporated into the intake assessment protocols of primary care physical therapy clinics. Early identification of injured workers experiencing elevated PTSS could permit instituting a pathway of care addressing multiple elements of the client’s problem profile. Although several empirically-supported treatments for PTSD have been described in the literature [72, 73], issues related to duration of treatment (e.g., 12 weeks) and timely access (e.g., long waitlists) limit the degree to which they can be usefully integrated within primary care services for WRMIs. As well, third party payors will likely be reluctant to consider referrals for lengthy psychological services for injured workers who do not satisfy all diagnostic criteria for PTSD.

There are recent reports supporting the effectiveness of brief treatments for PTSD using telehealth delivery platforms although these have yet to be tested with injured workers with elevated PTSS [74, 75]. Sullivan et al. [76] reported promising RTW outcomes following participation in a risk-targeted behavioral activation intervention for PTSD. Brief interventions aimed at reducing PTSS that can be easily integrated within the structure of care offered within primary care might be the most viable option for promoting more complete recovery following WRMI.

Several limitations invite caution in the interpretation of the study findings. First, all participants were employed full-time prior to their injury, and all were receiving a salary indemnity from a work-injury insurer. Participants were recruited from a network of primary care physical therapy clinics to which they had been referred by the work injury insurer. These selection criteria necessarily have implications for the generalizability of the study findings. Another limitation was that PTSS was assessed by self-report questionnaire. As a result, the degree to which participants who scored in the clinical range of the IES-R met diagnostic criteria for PTSD remains uncertain. The binary nature of the sex variable used in the present did not permit a more in-depth exploration of how gender identity might have influenced the pattern of findings.

It is not feasible within the limits of one research project to examine an exhaustive list of all variables that might mediate sex differences in recovery outcomes following work injury. It is possible that psychosocial correlates of PTSD/PTSS such as catastrophic thinking [77, 78], fear of injury [13] may also have a mediating influence on the relation between sex and recovery outcomes. Whether the explanatory value of PTSS is independent of the contributions of other potential mediators of the relation between sex and recovery outcomes following work injury remains to be clarified by future research.

The results of the present study suggest that the experience of PTSS might be one of the factors that explains sex differences in recovery outcomes following a WRMI. The findings suggest that PTSS are prevalent in the weeks following a WRMI, particularly in women. The experience of clinically significant PTSS following WRMI might add to the burden of disability, in turn, impeding timely recovery and RTW. The findings suggest that early screening for PTSS and the development of brief interventions aimed at reducing the severity of PTSS should be considered as potential avenues for fostering timely recovery and RTW in injured workers with elevated levels of PTSS.