In the last few decades, there has been an increasing number of compensation claims for WRUED (Baldwin & Butler, 2006). Similar to LBP, only a minority of workers with WRUED are absent from work for lengthy amounts of time, possibly 5 %, but this fraction of workers is associated with the greatest share of indemnity costs (Baldwin & Butler, 2006). Approximately one-third of workers with WRUED have been found to be at risk for prolonged work instability (Baldwin & Butler, 2006). WRUED, like LBP, is also characterized by recurrence in work absences (Baldwin & Butler, 2006). According to the results of a study by Baldwin and Butler (2006), most workers with WRUED RTW at least once, but they are even more susceptible to multiple work absences than workers with LBP. Among the group of workers who were off work at least once, 26 % of those presenting WRUED reported a second absence, while this percentage was 18 % for workers with LBP (Baldwin & Butler, 2006). Reviewing the literature shows that occupational disability and delays in RTW for WRUED are problematic, but a majority of workers eventually RTW (although they may present recurrent absences).

On a theoretical and conceptual level, different models found in the literature show the potential influence of a great number of factors, spanning from the individual, to the workplace, to societal contexts, on RTW and occupational disability (Schultz et al., 2007). As an example, Loisel et al. visually presented what they call the “arena in work disability prevention,” a figure that highlights the multiple influences and interrelationships between aspects related to the workplace system, the legislative and insurance system, the healthcare system, and the worker’s personal system and coping (Loisel et al., 2005). The work disability problem is viewed from an individual and public health viewpoint. Work disability and failure to RTW are further regarded as multicausal (Briand, Durand, St-Arnaud, & Corbiere, 2008; Pransky et al., 2011), although the actual causal factors are yet to be identified.

On an empirical level, a limited number of studies have been conducted on predictors of RTW in workers with WRUED (Bot et al., 2007; Clay, Newstead, & McClure, 2010). Investigated factors have been classified in different ways. For example, Ijzelenberg et al. identified individual factors, work-related physical factors, as well as work-related psychosocial risk factors (Ijzelenberg, Molenaar, & Burdorf, 2004). Feuerstein et al. distinguished individual psychosocial variables, including job stress and satisfaction, from organizational psychosocial variables, encompassing co-worker support, for instance (Feuerstein, Shaw, Nicholas, & Huang, 2004). In a narrative review, Pomerance classified factors that delay or prevent RTW of workers with WRUED more generally as intrinsic or extrinsic to the worker (Pomerance, 2009). Inspired by these various conceptualizations and classifications, we present here findings regarding predictors of RTW in cases of WRUED on the individual, organizational, and system levels. However, we acknowledge that these categories are not mutually exclusive and are interrelated.

I ndividual–Level Factors: Individual-level factors include sociodemographic, injury-related, and psychosocial variables. Based on findings from Pomerance’s (2009) review, women RTW up to 50 % later than men, while age has an unclear effect on RTW (Pomerance, 2009). Clay et al. recently published a systematic review on early prognostic determinants of RTW after acute orthopedic trauma resulting in upper or lower extremity injuries, some of which occurred during work (Clay, Newstead, & McClure, 2010). In this review, women were also at higher risk of prolonged work disability (Clay, Newstead, & McClure, 2010). The authors found that younger individuals, those with higher than average self-efficacy, and those with higher education are off work for a shorter time (Clay, Newstead, & McClure, 2010). Blue-collar workers and severely injured individuals were found to be at higher risk of prolonged work disability (Clay, Newstead, & McClure, 2010). Other individual-level factors that have been found to be associated with RTW include duration of work experience (Baldwin & Butler, 2006), pain attitudes (Clay, Newstead, Watson, & McClure, 2010), pain intensity (Feuerstein, Shaw, Lincoln, Miller, & Wood, 2003; Pomerance, 2009), behavioral factors (e.g., doctor shopping, recurrent absence from work), stress, coping skills, job satisfaction, and expectations (Clay, Newstead, Watson, et al., 2010; Pomerance, 2009). The role of individual factors in explaining RTW for workers with WRUED, especially individual psychosocial factors, however merits further study (Bongers et al., 2006).

O rganizational–Level Factors : Organizational factors associated with RTW in workers with WRUED include physical, as well as psychosocial factors. Bot et al. conducted a prospective longitudinal study specifically looking at work-related physical and psychological workplace factors associated with sick leave in workers who consult their general practitioners for neck or upper extremity complaints (Bot et al., 2007). They found that heavy physical work increased the risk of sick leave, while prolonged sitting reduced sick leave in the subgroup of workers who worried a lot about their pain. Psychosocial workplace factors were not found to be related to sick leave in this study. According to the results of Burton et al.’s review, available studies indicate that there is strong evidence that workplace psychosocial factors, including beliefs, perceptions, and work organization, are associated with upper extremity disorders for numerous outcomes including symptom development and work absence (Burton et al., 2009). In Clay et al.’s systematic review (Clay, Newstead, & McClure, 2010), none of the included studies were found to have considered factors related to the work organization. The role of psychosocial work conditions, such as employer or co-worker support, in explaining RTW for workers with upper extremity disorders has also been given relatively little attention (Bongers et al., 2006).

System–Level Factors : System-level factors associated with RTW are related to healthcare and insurance systems, for instance. Although they may also be viewed as individual-level factors, compensation status and active litigation were found to be associated with delayed RTW in Pomerance’s (2009) review. In Clay et al.’s review (Clay, Newstead, & McClure, 2010), workers not receiving compensation were also off work for a shorter time. The healthcare response (e.g., physician approach, stakeholder collaboration) is another example of system-level factors influencing RTW in workers with WRUED (Pomerance, 2009). Because they did not view compensation as a system-level factor, Clay et al. stated they did not find any studies on policy/system factors. Still, this indicates that further study is required (Clay, Newstead, & McClure, 2010).

Ergonomics is probably the form of workplace intervention most often cited and studied in the literature specifically on WRUED. Nonetheless, there is great variety and confusion regarding the nature of ergonomics, terms used to refer to ergonomics, as well as content of ergonomics-related interventions (Loisel et al., 2005). For example, according to Dick et al., ergonomic training includes education sessions for workers and also more action-based training, both varying in length and intensity (Dick et al., 2011). As for ergonomic interventions, they include actions such as workstation assessments and adjustments (Dick et al., 2011). There has also been a call for ergonomics to go beyond posture and forces in order to consider aspects such as work styles, gender, and weight (Feuerstein & Harrington, 2006). In a recent systematic review, Dick et al. examined the effectiveness of workplace interventions for WRUED, mainly ergonomics, to prevent or reduce sickness absence, retain one’s normal job, or prevent ill-health/retirement (Dick et al., 2011). In this review, upper limb disorders included CTS, nonspecific arm pain, extensor tenosynovitis, and lateral epicondylitis but excluded neck/shoulder pain. Based on their detailed analyses of 28 papers, the authors found limited evidence for the use of alternative keyboards to reduce CTS and tenosynovitis and lacking evidence for workplace interventions for epicondylitis (Dick et al., 2011). Similar findings were obtained in a systematic review on the use of specific keyboards with alternative force displacement of the keys or alternative geometry for workers with CTS (Verhagen et al., 2009). Another review by Burton et al. also found limited evidence for alternative mouse or keyboards to reduce pain in office workers and insufficient evidence for changes in equipment in the manufacturing industry (Burton et al., 2009). Studies on the effectiveness of arm supports showed moderate evidence of positive effects (Kennedy et al., 2010). Dick et al. concluded that there was moderate evidence that ergonomic work redesign targeting equipment or work organization does not reduce the incidence and prevalence of WRUED (Dick et al., 2011), although such interventions may improve worker comfort, which is non-negligible (Burton et al., 2009). Mixed results have indeed been reported for the effectiveness of ergonomic training and interventions (Dick et al., 2011). Also, according to Burton et al., giving too much attention to ergonomic interventions may send the message that work is the major cause of WRUED (Burton et al., 2009), which may be damaging for the worker because the epidemiology of the condition does not support such an affirmation. Furthermore, adopting a purely ergonomic approach to intervention is insufficient (Feuerstein & Harrington, 2006), although these forms of interventions have been by far the most prevalent. For example, Kennedy et al. found strong evidence that workstation adjustments alone are not effective (Kennedy et al., 2010). Integrating interventions addressing ergonomic and psychosocial factors, for instance, could be helpful in promoting return to regular work of workers having suffered an upper extremity injury (Shaw, Feuerstein, Lincoln, Miller, & Wood, 2001).

Other workplace interventions have been discussed in the literature. For example, the potential role of case managers has been recommended to facilitate RTW following upper extremity injury (Burton et al., 2009; Shaw et al., 2001). Case managers can help workers address barriers to their RTW (e.g., by obtaining workplace accommodations and engaging workers in active problem-solving) (Shaw et al., 2001). In their review, Bongers et al. found inconclusive evidence for a positive effect of task rotation, task enrichment, and added rest breaks on upper extremity disorders and, at the most, promising evidence for management engagement (Bongers et al., 2006). The need to individually screen each worker to identify problem areas in order to consequently target the best-fitting interventions has also been suggested (Noonan & Wagner, 2010). Other examples of interventions in the workplace include training programs for stress control, promoting physical activity, and offering wellness programs, but little is known of their effectiveness in WRUED (Pransky et al., 2002). Kennedy et al. found mixed evidence for the effectiveness of exercise programs, and exercise programs combined with ergonomics training (Kennedy et al., 2010). There was moderate evidence that biofeedback and job stress management training taken separately were not effective. The evidence was limited for rest breaks and insufficient for rest breaks combined with exercise. Modified work, taking the form of “light duties,” or gradual RTW through increasing demands in performance and time on the job may allow workers to gradually resume their tasks and reintegrate into the workplace, hopefully reducing stress, anxiety, and other psychosocial factors that may prevent RTW (Noonan & Wagner, 2010). Individual-level and organizational-level interventions may act on worker stress, for example, and consequently help alleviate upper extremity symptoms (Bongers et al., 2006). Studies examining the effectiveness of other work interventions such as work organization are rather scarce. This has notably been justified by the methodological challenges they entail (Bongers et al., 2006).

Preventive efforts need to address modifiable risk factors (Bongers et al., 2006; Staal et al., 2007). However, further work is needed in the area of prevention of WRUED. Overall, as a field, work disability prevention is emerging (Loisel et al., 2005). Compared to LBP, upper extremity disorders have received much less attention in work disability prevention (Loisel et al., 2005). The effectiveness of interventions aiming to prevent the development of WRUED, thought as primary prevention, has been largely understudied (Bongers et al., 2006). For instance, the evidence on the effectiveness of broad injury prevention programs and injury prevention programs with physical therapy was insufficient in a recent systematic review by Kennedy et al. examining the effectiveness of occupational health and safety interventions on upper extremity disorders in terms of symptoms, signs, disorders, injuries, claims, or lost time (mostly in office work) (Kennedy et al., 2010). Briand et al. described RTW interventions as tertiary prevention, aiming to reduce the consequences of work absence, while the goal of secondary prevention interventions is to avoid absence from work (Briand et al., 2008). In the literature addressing specifically the prevention of WRUED, distinctions between primary, secondary, and tertiary prevention measures have not been routinely made. Studies also often address more than one level of action. In Kennedy et al.’s review, some of the studies addressed only primary (n = 9) or secondary prevention (n = 8), 15 were a mix of both, two were a mix of secondary and tertiary prevention, and two others were a mix of primary, secondary, and tertiary prevention (Kennedy et al., 2010). Overall, the authors concluded that the level of evidence was mixed for the effectiveness of occupational health and safety interventions on upper extremity disorders outcomes, with medium- and high-quality studies having provided inconsistent results (Kennedy et al., 2010). Such pooled results do not allow to distinguish the effectiveness of primary, secondary, and tertiary prevention. Future studies related to prevention in cases of WRUED should clearly identify their level(s) of action in order to help develop knowledge in this area and identify the best targets of action to aim for in order to improve effectiveness of preventive efforts. For instance, Staal et al. suggested that because incapacitating WRUED only affect a minority of the working population, preventive efforts should only target high-risk groups or workers already presenting symptoms (Staal et al., 2007).