Return to work after vocational rehabilitation for sick-listed workers with long-term back,neck and shoulder problems: A follow-up study of factors involved

Abstract

BACKGROUND: Absence from work due to musculoskeletal disorders is a significant problem from a number of perspectives, and there is a great need to identify factors that facilitate return to work (RTW).

OBJECTIVE: To identify factors related to RTW after vocational rehabilitation for sick-listed workers with long-term back, neck and/or shoulder problems, and to compare the results with those from a previous literature review based on studies from 1980 to 2000.

METHODS: A literature review based on studies published 2001–2014 in PubMed, CINAHL, EMBASE, ERIC and the Cochrane Library.

RESULTS: Major risk factors for not returning to work are: higher age, factors related to pain such as higher levels of pain and pain related fear, avoidance of activity, high distress and depression. Facilitating factors for RTW are: lower functional disability, gaining control over one’s own condition, believing in RTW and work-related factors such as occupational training, and having a job coach or an RTW coordinator. Compared with the results from the previous review, the present review shows less of a focus on sociodemographic factors and more on psychological factors. Both studies highlight risk factors such as older age, higher levels of pain, depression and less internal locus of control.

CONCLUSION: Psychosocial and work-related factors are important and should be included in interventions for the RTW of people with long-term back, neck and/or shoulder problems.

Keywords

Musculoskeletal disorders sick leave re-entry working life facilitating factors

1 Introduction

1.1 Long-term sick leave and musculoskeletal disorders

Sickness absenteeism is a great problem in Sweden and in many other countries, and musculoskeletal disorders (MSDs) constitute one of the most common work-related problems [1]. Long-term sick leave, i.e. absence from work 60 days or more due to sickness, illness and/or disability has negative consequences for the employer, the employee, and for society as a whole. For the employer long-term sick leave leads to administrative, financial and practical challenges, and for the individual it is often associated with poor health [2], a worse financial situation, and social stigmatization [3]. For society, the consequences are primarily economic [3], and in Sweden, the cost of MSDs is around 50 billion Swedish kronor (∼7.5 billion USD) a year [4]. During the coming decades, the demographic change, involving new compositions on the labor market and an ageing workforce, will influence western countries in many ways [5, 6]. Work-related health problems are more common among older workers, and around 40% of working men and women in the EU who are over 45 years of age report musculoskeletal and/or psychosocial problems related to work [6, 7].

1.2 Theoretical perspective on disability

There are different models for understanding disability. Traditionally, healthcare for musculoskeletal conditions is based on a medical model [8], which views disability as a feature of the person, where the disability requires medical care, treatment or some other intervention so that the problem can be corrected [9]. Another model is the social model, which sees disability as a problem created by society and not as an attribute of the individual [9]. Both these models are partly valid, but since disability is a complex phenomenon and both these models seem to lack enough personal and psychological factors, the biopsychosocial model has been constructed [8]. Even if symptoms and illness derive from a special health condition, there is now considerable evidence to show that the development of chronicity often depends on psychosocial factors, and that non-specific MSDs can only be understood and dealt with according to this model [8]. The International Classification of Functioning, Disability and Health (ICF) is based on the individual-centered biopsychosocial model, which takes biological, psychological and social factors into consideration, and where function and disability is seen in terms of a dynamic interaction between those factors [8, 9].

The past two decades have provided explanations for predicting factors leading to work disability [10]. Rather than trying to cure peoples’ disabilities, one important step was to see disability in terms of the above-mentioned biopsychosocial model and, in coherence with this paradigm, make efforts to develop and evaluate interventions aimed at returning people to work [10]. Another was that the disability process has been divided into the acute, sub-acute and chronic phases, depending on the duration of absence from work [10].

The sub-acute phase of low-back pain, 4–12 weeks post-initial injury, has been highlighted as the important phase in which to start rehabilitation (both medically-oriented and more workplace-based), in order to prevent the disability problem from becoming chronic [11]. Most people who are sick-listed due to an MSD also return to work within a few weeks [12], and by starting an intervention in the sub-acute phase, unnecessary interventions can be avoided [10]. Different predictors seem to have different impacts in the RTW-process in the acute, sub-acute and chronic phases [13]. Psychological factors seem to be strong predictors of long-term disability, and cognitive and behavioral factors such as pain-related fear, distress and avoidance of activity influence the transition from acute to chronic pain [14].

1.3 Vocational rehabilitation

Among the various stakeholders in the RTW process, i.e. sick worker, the doctor, the physiotherapist, the employer, there is a broad consensus that as soon as the health of the sick workers and the workers with disabilities permits, they should return to work [2]. Here, vocational rehabilitation (VR) has a central role [15]. According to the International Labor Organization (ILO) in 1955, the objective of VR is to enable a person with disability to secure and retain suitable employment [16]. In Sweden, the concept of VR was implemented in the social insurance system in 1992, in order to reduce sickness absence through early and active rehabilitation initiatives [17]. The aim of VR in Sweden is to maintain the so-called “arbetslinjen” (directly translated as “the work line”), so that as many people as possible can provide for themselves through their own gainful work [18]. Waddell, Burton and Kendall defined vocational rehabilitation as follows:

“Vocational rehabilitation is whatever helps someone with a health problem to stay at, return to and remain in work. It is an idea and an approach as much as an intervention or a service.” [15, page 39]

This is a broad definition, considering that for instance in Sweden, VR does not normally include medical interventions [15]. In the review presented here, the definition of VR is influenced by Waddell et al.’s definition, and vocational rehabilitation is defined as all interventions – medical, psychological, social, and occupational – that can bring a sick-listed person with previous work history back to his/her own work or another available job. This definition is also in line with the definition used by Selander et al. [3].

The process of the way back to work through VR for a worker with disability can be seen in Fig. 1.

The process, described earlier by Selander et al. [3], starts with an employee who is at work (1), and then at a certain point falls ill or gets injured (2). The individual receives medical care from the healthcare system (3) and, as mentioned earlier, most people with an MSD will return to work within a few weeks [12]. However, for those individuals who have received healthcare but for whom disability remains, rehabilitation (4) may be considered. In the first phase, the individual might be given medical rehabilitation (4a) at a primary care centre or at a rehabilitation unit. In the next phase, vocational rehabilitation (4b) might be initiated for an individual of working age. VR may include both medical and non-medical measures (see Fig. 1) [3], and it is often those who are long-term sick-listed who get some form of VR [1], even though the time frame for receiving it may differ from one country to another. Stakeholders from different professions, organizations and sectors of society are often involved in VR [19]. Collaboration between these different stakeholders is dependent on factors such as communication, commitment and trust, but also to a common ground for collaboration, leadership, and rules and regulations, and can in many situations be seen as both facilitators and barriers [19]. Furthermore, their different values, goals and ways of working might have consequences for effective collaboration, which might make the rehabilitation harder for the individual [20]. Due to differences between countries regarding social insurance- and social security systems, economic environment and social culture [21], different factors in the process might have different impacts in different countries. Following VR, a decision (5) is made (in Sweden by the social insurance office). In the best case, the individual can return to work and in the worst case, where working capacity is permanently reduced, the person has the right to temporary or regular disability pension (6). When the person with disability can return to work or to the labor market (i.e. actively looking for a job), VR is considered successful. Therefore it is of great interest to find out what factors are related to return to work after VR.

In a literature review from 2002 [3], with an aim similar to that of the present review but based on studies that were published between 1980 and 2000, Selander et al. present many factors associated with RTW following VR. The result from that review shows that several demographic, medical, psychological and social factors, in addition to factors related to the workplace, rehabilitation and sickness benefits can be associated with each other and the final outcome of VR. In short, the following factors facilitate with RTW after VR: when compared to the rest of the sample in each study; being young, native, married, having high work seniority, a steady job, well educated and a long working history, a caring employer, a good working environment and a high income [3]. Further, factors such as having a good social network, being self-confident, not being depressed, having less pain and low disease severity also facilitate RTW [3].

Against the background presented above, it is the authors’ belief that there is a need for a follow-up that can summarize potential new knowledge in this field.

2 Purpose

The main purpose of this literature review is to identify factors related to return to work after vocational rehabilitation for sick-listed workers with long-term back, neck and/or shoulder problems. A secondary purpose is to compare the results from this review with the results from an earlier literature review by Selander et al. [3]. Our aim is not to discuss and analyze all the possible factors associated with a return to work, even if this might be desirable, but rather to provide an overview, which can increase understanding in this area.

3 Methods

3.1 Search strategy

To enable a comparison with the prior review by Selander et al. [3], the search strategy in this current literature review was inspired by the strategy used in that study. After minor modifications, due to an increase in literature and databases, the final search for relevant literature was carried out using two different keyword combinations: (1) vocational rehabilitation AND return to work AND musculoskeletal disorders, and (2) vocational rehabilitation AND return to work AND predictors. In February 2014, a keyword search was performed in the PubMed, CINAHL, EMBASE, ERIC databases and the Cochrane Library.

3.2 Inclusion and exclusion criteria

Inclusion criteria: (1) focus on return to work, and (2) inclusion of vocational rehabilitation. The disorder had to be (3) musculoskeletal, primarily in the back, neck and/or shoulders. Studies that also included “other work-related disorders” were then marked with an asterisk (*) as shown in Appendix A.The MSD had to (4) last for 60 days or more, and the study population (5) had to be of working age (in general 16–65 years). And finally (6) the articles had to be published in English.

Exclusion criteria: (1) mixed diagnoses and (2) accidents with limb amputations and spinal cord injuries were excluded, as were (3) review studies. However, the references in the reviews were searched for relevant articles.

3.3 Quality assessment of the articles

The quality assessment was performed using CASP [22] and STROBE Statement [23]. CASP (Critical Appraisal Skills Programme) is helpful in finding and checking research for trustworthiness, results and relevance [24], and STROBE Statement (STrengthening the Reporting of OBservational studies in Epidemiology) is an international collaborative initiative of epidemiologists, methodologists, statisticians, researchers and journal editors with the common aim of strengthening the reporting of observational studies in epidemiology [25]. When assessing quality, the articles were judged from the overall impression, including risk for bias. Quality was also assessed in relation to the purpose of this review. The quality of the studies was judged as satisfactory or unsatisfactory. The questions used in the quality assessment are presented in Appendix B.As a further step in quality assessment, the two authors separately assessed the quality of all included articles and discussed the articles together until agreement was reached. Unsatisfactory studies were excluded and are not presented in this literaturereview.

3.4 Results of the search

The keyword search carried out in February 2014 yielded a total of 231 articles. After reading the titles and excluding duplicate references that were found in more than one database, the search generated a total of 56 articles. The abstracts of these were read, and the number of relevant articles was further narrowed down to 21. After scanning the reference lists of these articles and the reference lists of the reviews that appeared in the search, 20 more articles were found, 11 of which were found to be relevant after scanning the abstracts. The quality of the 32 articles was separately assessed by the authors of this article, which led to three articles being excluded. Finally, a total of 29 articles were included in this literature review.

4 Results

4.1 Different kinds of interventions and outcome

The studies included in this literature review consisted of many different kinds of interventions, and in approximately one third of the studies the interventions were compared with other interventions or control groups. More comprehensive interventions with a multidisciplinary approach produced good results [12 , 26–30] compared with single-mode treatments [26, 28] or traditional primary care [31, 28] in returning people that are sick-listed to work. The interventions where there was a high rate of return to work were often workplace-based [29, 32], including graded work exposure [26, 30], or occupational training [12]. Having a job coach [29], an RTW coordinator [33], or a case manager with a more proactive role [32] also gave a more favorable result. Furthermore, studies that compared single-mode treatments such as physical group training with cognitive interventions [34], or a cognitive-behavioral program with traditional primary care [13], found no significant differences in outcomes. Two studies showed that disability pensioners were very hard to rehabilitate with brief vocational interventions [35, 36]. All the included studies, with type of intervention and total results, are listed in chronological order in Appendix A.

4.2 Key findings

The identified factors related to RTW after VR have been divided into five different categories: sociodemographic factors, biological/medical factors, psychological factors, work-related/social factors, and systemic factors. The findings are presented and explained in Table 1.

5 Discussion

5.1 Methodological discussion

Conducting a literature review includes making decisions about which studies to include and exclude [49]. The aim of the current study was to investigate RTW following VR after long-term back, neck and/or shoulder problems. All of the articles in this review focused on these disorders, but some also included other musculoskeletal disorders, such as upper or lower limbs. Since these disorders are in many ways similar to back, neck and shoulder problems, we do not think that the results are affected. In the included articles, RTW was defined in many different ways; e.g., returning to part-time or full-time work, being involved in job retraining or different kinds of education programs [27], and this might have had an impact on the outcome. The studies included also had a wide range of study designs.

Performing a review of the literature includes conducting a comprehensive unbiased search based on keywords and search terms [49]. The search terms used in this overview could be discussed in terms of being too few or too narrow for searching this field. This is due to the fact that different concepts are used in connection with VR [19]. Considering this, the search could have been made with an extended number of search terms, which might have resulted in additional relevant articles that could have been included.

5.2 Discussion of results

The main purpose of this literature review was to identify factors related to RTW after VR for sick-listed workers with long-term back, neck and/or shoulder problems. A secondary purpose was to follow up and compare the results we obtained with those from Selander et al.’s previous literature review [3].

5.2.1 Vocational rehabilitation

The results indicate that psychological factors are most important in determining the outcomes of VR. An observation was that, when compared to the measures from 1980-2000, more recent interventions are more comprehensive, that disability is viewed from a biopsychosocial perspective and that the different phases (acute, sub-acute and chronic) of injury are associated with different factors predicting RTW outcomes. It is therefore not surprising that more comprehensive multidisciplinary or interdisciplinary interventions had better results than single-mode treatments or primary care in returning sick workers and workers with disabilities back to work. A further pattern was that interventions with a high rate of RTW were in many cases occupationally oriented and workplace-based [41]. Having a job coach, RTW coordinator or case manager was also predictive of RTW. This was expected, since this person coordinated RTW [33], helped to integrate the needs of the employee with available resources [32], and facilitated contacts with the employer, all of which minimized the psychosocial problems associated with the workplace [29]. The results also showed that there were no significant differences in outcomes between different single-mode treatments, or between a cognitive-behavioral program and traditional primary care; this supports the notion that more comprehensive work-related interventions should be given.

5.2.2 Sociodemographic factors

Many of the studies indicated that younger age is a strong predictor of RTW, but there are also studies [33, 42] indicating that age is a non-significant factor. Younger age as a predictive factor is expected, since younger people often have better health and are often more attractive on the labor market. A more surprising finding was that women were more likely than men to return to work [28 , 33]. This might have been due to the more multidisciplinary/interdisciplinary character of the interventions, with a greater focus on psychological factors related to RTW, or the interventions being more work-related with an RTW coordinator.

As expected, a lower level of education and being less active during leisure time did not facilitate RTW. A more surprising finding was that occupation [42], lack of employment status [38], high education [13, 33] or having a co-habitant [13] neither predicted nor hindered RTW. One explanation for this might be that the interventions were more effective and focused on the right areas. Higher professional ranking did not predict RTW in one study [37]; this was unexpected but was explained by the fact that having higher professional ranking sometimes involves a higher level of stress and a higher degree of demands and responsibilities, which can be negative for a RTW.

5.2.3 Biological/medical factors

Many factors related to RTW after vocational rehabilitation seem to be different sides of the same coin. It comes as no surprise that good general health, lower functional disability, increased physical functioning and status and less pain during interventions facilitated RTW, and that the opposite hindered RTW.

Higher levels of pain, pain-related disability and difficulties in managing pain seem to be among the major risk factors for not returning to work. According to Cutler et al. [44], different psychological factors show strong relationships at different times, but indicate that pain level is the most important determinant of functional capacity and outcome.

5.2.4 Psychological factors

Two studies indicated that RTW has less to do with physical functioning and more to do with psychosocial factors [27], mental health and/or pain symptoms [50]. It seems as if psychological and work-related factors are more common than sociodemographic and clinical factors during the chronic phase [30], and that psychological risk factors are powerful predictors of long-term disability [14].

Powerful risk factors are pain-related fear, distress and avoidance of activity [14]. Fear of pain and injury might be more disabling than pain itself [34]. If a positive outcome is to be achieved, it is therefore important to gain control over one’s own condition (i.e. pain, fears or mood). People with chronic disabilities might need to re-learn that it is safe to move and be physically active despite having pain. Individual perceptions of the RTW process and of its potential outcome are also important, and seem to be an influential factor for RTW [47]. Having a high degree of internal locus of control (LOC) is therefore important.

Workers who did not return to work often had low expectations of returning to work [36 , 47], low self-esteem [36], low self-judgment of work ability [36] and higher levels of depression [27 , 48] and anxiety [14]. They also maintained their initial representation of pain as “abnormal” and something dangerous, or used avoidance behaviors [46].

Somewhat surprising results presented by Marois & Durand [30] are that higher psychological distress and the fear that returning to work will cause an aggravation of symptoms, actually promote RTW. Marois & Durand’s conclusion is that in the chronic phase, the relationship between several factors and RTW appears to be reversed when people participate in an interdisciplinary rehabilitation program with prior screening for those factors [30]. The screening at the time of admission to a program makes it possible to select appropriate activities during the intervention [30].

5.2.5 Work-related / social factors

Interventions where there was a high rate of RTW were often workplace based and occupationally-oriented with someone coordinating the RTW [33], integrating the needs of the employee with available resources and facilitating contacts with the employer [29 , 33]. Other factors that facilitated RTW were a low level of previous sick-listing [13] and development of competent work behaviors [41].

The social factor is also important in the RTW process. Lack of support and understanding from colleagues and supervisors hinders RTW. When examining how sick-listed workers perceived their chances of returning to work, Svajger et al. [47] found that when they experienced exclusion from decision-making regarding work-related ability it led to doubt and fear about future opportunities to participate in work. This in turn determined how they perceived their chances of RTW more than the MSD-related problems.

5.2.6 Systemic factors

As expected, early rehabilitation and early meetings at the employee’s workplace, with all parties involved, facilitated RTW. However, when a worker on long-term sick leave did not share the same vision as the medical stakeholders, or when he/she had a perception of a therapeutic failure, it did not promote RTW at all. What facilitated RTW was close collaboration between the worker, his/her employer, the National Health Insurance Office, the healthcare system and other parties involved. As emphasized by Andersson et al. [19], collaboration can make the rehabilitation process easier, but it can also cause difficulties, depending on the different values, goals, leadership, limitations, rules and regulations of the various stakeholders.

5.2.7 Comparison of the results

A comparison of the present review with the one by Selander et al., based on data between 1980 and 2000, shows that the current focus (data from 2001-2014) is less on sociodemographic factors and more on psychological ones. This might be a consequence of looking at disability from a biopsychosocial perspective. In both reviews, multidisciplinary treatment showed better results compared with single-mode treatment [3].

Furthermore, both reviews indicate that older age is a major obstacle for returning to work. More studies in the present review indicated that women returned to work more often than men, and contrary to Selander et al.’s study [3], occupation and employment status was not associated with RTW. In both studies it was reported that less pain and lower functional disability facilitated RTW.

In the present review, psychological factors are emphasized as powerful predictors. As in Selander et al.’s review [3], depression and less LOC hindered RTW, while good self-confidence, better experienced health and a positive expectancy of RTW facilitated RTW. Other powerful cognitive and behavioral risk factors identified in the current study, such as pain-related fear, distress and avoidance of activity, were not found in the previous study.

Both reviews emphasized the importance of early rehabilitation, and of the individual being able to influence his/her own rehabilitation. An “understanding of the workplace” by rehabilitation providers is shown to facilitate RTW in Selander et al.’s review [3], which can be compared with the importance of having early meetings with all stakeholders at the employee’s workplace, so that everyone involved can gain an understanding of the situation at the same time.

5.2.8 Where do we go from here?

Although many of the comprehensive interventions led to RTW, many injured workers still reported disabilities, pain, high mental distress and persisting fears. In many cases there seemed to be no difference in how they ranked their health. What factors influenced their decision to return to work? Why did they return to work? Is it as simple as returning to Antonowsky’s “sense of coherence” [51] with the three components: comprehensibility, manageability and meaningfulness? Moreover, in vocational rehabilitation and its different interventions it seems wise to focus on factors that can be improved. Factors such as age and gender cannot be changed, while factors such as fear of pain, physical activity, and mood can. It is also important to address the MSD from a biopsychosocial point of view. As Waddell [8, page 64] states: “Biopsychosocial problems need biopsychosocial solutions.”

The strength of this study is that it offers broad knowledge, based on available evidence from 2001–2014, about factors related to RTW after vocational rehabilitation and different interventions, but it lacks a deeper understanding of the factors involved. We did not discuss and analyze every single factor, which was not the aim, however desirable it might have been. The aim was rather to provide an overview in order to increase a broad understanding in this area. The authors hope that this overview can be of value for decision-makers, and contribute to new practices to get long-term sick people back to work.

6 Conclusion

Factors associated with a RTW after vocational rehabilitation for workers who are long-term absent from work due to long-term back, neck and/or shoulder problems have been identified. The results of this literature review indicate that psychosocial and work-related factors are important in the RTW process. Major risk factors for not returning to work are: higher age, high levels of pain, difficulties of managing pain and several powerful psychological factors such as high distress, pain-related fear, avoidance of activity and depression. Facilitating factors for returning to work are: good general health and physical functioning, less pain, lower functional disability, gaining control over one’s own condition and believing in a return to work.

In comparison with the previous literature review by Selander et al. [3], the focus is now less on sociodemographic factors and more on psychological factors. However, in both studies, interventions with a multidisciplinary approach showed better results. The results of both reviews also indicate that e.g. older age, depression and less LOC are risk factors for not returning to work. Meanwhile factors such as a positive expectancy of return to work, less pain, lower functional disability, the importance of early rehabilitation, and being able to influence one’s own rehabilitation facilitated a return to work.

Conflict of interest

The authors report no declarations of interest.

Footnotes

Appendix A

Studies included listed in chronological order

Reference	Kind of intervention/ study design	Definition of disorder	Results and factors involved in the process.
			+Facilitating factors related to RTW.
			–Hindering factors related to RTW.
			o Factors related to RTW that are neither facilitating nor hindering.
Durand &Loisel, [26].	Evaluation of a Therapeutic Return to Work (TRW) program combining graded work exposure with functional restoration therapy. Compared with only functional restoration (FR) therapy, community services without any rehabilitation intervention (CS) &the group denied TRW and FR.	Low back pain (LBP)	In the TRW group, 93% of the participants returned to work.
	A two-year follow up study with controls. 127 participants.		RTW rate significantly higher only when comparing with the group denied TRW (33%) and the FR (73%) group.
			The FR group did not return more participants to work than the CS group that had no rehabilitation treatment.
			The denied participants had the worst RTW rate, scoring lowest on QoL, had higher level of pain, fear and avoidance beliefs, physical disability and litigation.
			+The TRW group had significantly less pain intensity than the other groups.
Arnetz et al., [32].	A more proactive role for the insurance case manager, together with workplace-based ergonomic interventions vs. traditional case management.	Back, neck and shoulders^*	Odds ratio of RTW in the intervention group v/s the reference group: 2.5. Number of sick days significantly less, but no significant difference between groups when ranking health.
	A prospective controlled intervention study with 137 participants.		+The case manager’s more proactive role in integrating the needs of the employee with available resources.
			+Close collaboration between the National Health Insurance agency (FK), employer and the healthcare system.
			+Early meeting at the employee’s workplace involving the employee, employer, the occupational therapist/ergonomist and the FK.
			+Patient rates the role of the FK as more supportive and important than the reference group.
Cutler et al., [44].	Multidisciplinary pain treatment center.	LBP	Functional capacity measures on admission and different psychological factors (at different times) are strong predictors of RTW outcome. The only psychological factor independently predicting long-term employment outcome is trait anxiety.
	Long term follow up study with 188 participants.
			–Higher levels of pain, depression and receiving worker compensation.
Koopman et al., [12].	Multidisciplinary Occupational Training Program.	LBP	85% of participants returned to work one year after (>60% full-time).
	A prospective cohort study with 51 participants.		+Increase in VO2 max and trunk flexibility.
			+Reduction in generalized fear, psychoneuroticism and depression, less catastrophizing and lower functional disability during the program.
			+Male as gender and younger age.
			+Using the coping strategy of reinterpreting pain sensations more frequently.
Staal et al., [31].	Behavior-oriented graded activity program vs. usual care.	LBP	The behavior-oriented graded activity program was more effective than usual care in reducing the number of days of absence and in improving functional status and pain.
	(Ensuring the disabled worker that despite pain, it is safe to move and to be physically active).		+Graded activity in an occupational health setting.
	A randomized controlled trial with 134 participants.
Watson et al., [48].	Occupationally oriented pain management program for long-term unemployed people. Including psychological, physiotherapeutic and vocational focusing.	LBP	At six-month follow-up, 38.4% had returned to work (none to their previous employer).
	Six months follow up study with 86 participants.		–People with longer duration of unemployment (significant), higher initial scores of somatic anxiety and depression.
			o Type of benefit did not seem to influence outcome – contradiction to other studies.
			o Being away from work more or less than two years.
Watson &Main, [52].	Occupationally oriented pain management rehabilitation program for unemployed people.	LBP	No significant difference between the two groups regarding pre-program benefit type for employment outcome or positive progress. Both groups equally likely to return to work.
	Aim: Influence of type of benefit on RTW, incapacity benefit vs. job seekers’ allowance.		No significant differences between groups regarding: self-reported disability, pain, physical performance, fear avoidance beliefs or psychological distress.
	Prospective study with controls. 86 participants.
Vowles et al., [27].	Interdisciplinary treatment based on a sports medicine approach.	Back, neck^*	70.9% back at work within six months or involved in some type of retraining.
	A six month follow up study with 138 participants.		–Higher levels of depression, pain intensity and pain-related disability post-treatment.
			–Higher age. (Age mean value; Not working 42.8 and working: 39.1.)
			–People able to lift lower amounts of weight from floor to waist level at the end of treatment.
			Measures of pain, psychosocial distress &physical functioning generally related. General emotional distress could be the most important predictor of work status post-treatment.
Jensen et al., [28].	Behavioral medicine rehabilitation program (BM) including behavioral-oriented physiotherapy (PT) and cognitive behavioral therapy (CBT) vs. only PT, only CBT or	LBP, neck.	Better results for BM than the three other programs. Strongest effect in women. Women in the BM group returned to work faster and showed better health compared with CG.
	treatment as usual (CG).		+Combination of CBT and PT.
	Randomized controlled trial with 214 participants.		+Increasing the activity and at the same time eliminating the fear of pain and movement.
Loisel et al., [45].	The values underlying team decision-making in a work rehabilitation facility in a hospital. Weekly discussions.	MSDs^*	10 common decision values:
	PREVICAP program:		1. Team agreement and reliability.
	1. Disability diagnostic step.		2. Collaboration with stakeholders.
	2. A progressive RTW process combining a therapeutic RTW and an ergonomic intervention centered in the workplace.		3. Worker’s internal motivation.
	A single case observational study (qualitative approach) with 22 participants.		4. Workers’ devotion to the program.
			5. Workers’ reactivation (start training, moving without fear).
			6. Single message (same goal of actions taken towards the worker).
			7. Reassurance (to worker, team &stakeholders).
			8. Graded intervention.
			9. Pain management
			10. Return to work as a therapy.
			These were sorted into team-related values, stakeholder-related values, worker-related values and general values that influence the intervention.
Lydell et al., [37].	Rehabilitation program.	Back^*	+Higher levels of ability to undertake activities, QoL and fitness on exercise are important independent factors.
	A descriptive quantitative study with 385 participants.		+Early rehabilitation.
			–More pain and earlier sickness certification periods.
			–Higher profession scores in the “sickness presence group” (Opposite relation to profession).
			Not significant: No job to return to.
Storheim et al., [34].	Physical group training intervention or cognitive intervention.	LBP	No significant difference between interventions.
	A prospective study with a randomized controlled trial with 93 participants.		–A high degree of fear-avoidance beliefs for work, increased disability and lower cardiovascular fitness (best predictors for not returning to work).
			–Less educated and less active during leisure time.
			–Higher workload and more negative job characteristics.
Sullivan et al., [42].	Pain disability prevention–program. Community-based psychosocial intervention.	Back, neck^*	63.7% RTW within four weeks post-treatment.
	Targeted risk factors:		+Reductions in all four targeted risk factors.
	^*Fear of movement/re-injury.		+Reductions in pain catastrophizing.
	^*Catastrophizing.		+Pain reduction was a significant predictor of outcome when considered alone, but not in combination with other psychological risk factors.
	^*Perceived disability.		–Elevated pretreatment scores on fear of movement and re-injury and pain severity.
	^*Depression.		–Time off work.
	A prospective follow up study with 215 participants.		RTW not related to age, gender or occupation.
Ross et al., [50].	Standard care vs standard care+Worker-Based Outcomes Assessment System (WBOAS).	Back^*	Standard care+WBOS did not improve RTW or success in staying at work. Improved physical functioning and new injury/re-injury avoidance, but did not improve mental health and pain symptoms.
	WBOAS: Data of patient-reported functional health status, pain symptom and work role performance outcomes given to PT/OTs and their patients.
	A non-randomized controlled trial where 158 participants completed the study.
Cheng &Hung, [29].	Workplace-based (WWH) vs clinic-based (CWH) work hardening.	Rotator Cuff Disorder	WWH more effective than CWH in returning people to work (74.4% vs.37%).
	WWH: WWH training with rehabilitative principles, biomechanics and specific job activities. Job coach.		WWH in comparison with CWH:
	CWH: Traditional generic work hardening training.		+Significant decrease in self-reported shoulder problems, greater improvement in functional work capabilities and a decrease in perceived pain and disability.
	A randomized controlled trial with 103 participants.		+Job coach ⟶ contact with the employer, which minimizes the psychosocial problems associated with the workplace.
Magnussen, Nilsen et al., [36].	Interview with disability pensioners, examining the perceived barriers against returning to work.	LBP	–Earlier negative experiences in their workplaces.
	(This project was part of a brief vocational-oriented intervention for disability pensioners.)		–Lack of support and understanding from colleagues and supervisor.
	Focus groups interviews with 17 participants.		–Lack of support from social security authorities.
			–Unsuitable economic arrangements.
			–Low self-judgement of work ability (main barrier).
			–Negative expectations of RTW.
			–Low self-esteem (preventing women from telling their needs and rights at the workplace).
Magnussen, Strand et al., [35].	Brief vocational-oriented intervention vs control group for disability pensioners.	LBP	No statistically significant effect on RTW or having entered a RTW process at one-year follow-up. Although, twice as many in the intervention group had entered an RTW process compared with the control group.
	Intervention: education, reassurance, motivation and vocational counselling.		Very small differences related to the RTW process (secondary outcome measures):
	Randomized controlled trial with 89 participants.		+Positive expectancy, better physical performance and less pain.
Moliner et al., [43].	Comparing individualized subjective quality of life, ISQoL, according to work status and reference values following interdisciplinary work rehabilitation program (IWR).	Back, neck^*	Global ISQoL gap is not related to work status.
	A cross-sectional study with 71 participants.		Emotional distress does not differ with regard to work status but is the main explanatory variable of ISQoL.
			Related to ISQoL are: Pain, PCS (summary of physical functioning, role physical, bodily pain and general health), MCS (summary of vitality, social functioning role, emotional and mental health), perceived disability and distress (38%).
			Factors related to emotional well-being might be the most challenging for the chronic MSD population.
			Sociodemographic variables not associated with global ISQoL.
			–More time before starting an IWR, reporting more health problems, perceiving pain intensity as greater, having worse PCS score and perceiving greater disability in other activities than work.
Selander et al.,[38].	Rehabilitation program consisting of individual and group activities preparing the individual for the demands of daily life and work situation.	LBP	+Good general health. Perception of own health in positive terms.
	A six months follow up study with 347 participants.		+Higher degrees of vitality.
			+High degrees of Internal Locus of Control (LOC).
			–Higher age.
			No findings that men are more likely to return to work, or that people with no employment status and previous long periods of sickness absence are more difficult to rehabilitate.
Sjöström et al., [14].	Multidisciplinary rehabilitation program.	Back, neck	Full-time sick leave decreased gradually during the follow-up period (more in women).
	Individually adopted.		+Increased health-related QoL.
	A prospective intervention design study with 60 participants.		+Participants still physically active after two years.
			+Decreased anxiety (both sexes) and depression (only women).
			+Decreased self-experienced stress.
			Cooperation between participant, employer, social insurance office and other parties involved was important in the rehabilitation program.
Marois &Durand, [30].	Previcap programme.	MSDs^*	High RTW rate: 73%.
	Interdisciplinary work rehabilitation program. TRW+gradual exposure to the real work environment.		More women than men return to work (82.4% vs 66.4%).
	A descriptive correlational study with 222 participants.		In the chronic phase, the relationship between several factors and RTW seems to be reversed when combined with participation in an interdisciplinary rehabilitation program that includes prior screening for these factors and a workplace intervention.
			General prediction model:
			+Presence of awkward postures and prolonged static work postures and light duties unavailable. Diagnostic labelling, fear of intensifying symptoms if returning to normal activity and longer participation in the program.
			–Shorter tenure and worker’s higher perception of disability and of a therapeutic failure.
			Men’s model:
			+Absence of MSD history requiring an absence from work, longer duration of absence from work, unavailability of light duties, diagnostic labelling and longer participation in the program.
			–Worker’s higher perception of disability, shorter tenure, worker’s perception of a therapeutic failure and worker’s perception of having a serious injury.
			Women’s model:
			+Fear of intensifying symptoms if returning to normal activity, the presence of work equipment judged unsuitable, higher psychological distress and presence of recent personal events.
			–Perception of a therapeutic failure, increased pain level stopped the attempt to return to work and presence of a favorable representation to a RTW.
Svajger et al., [47].	How sick-listed people perceive their chances of returning to work.	Back^*	In developing perceptions, the work environment and health and disability systems were experienced as the most influential.
	A qualitative study with 6 participants.		–Experiencing exclusion from decision-making regarding their work-related ability ⟶ doubt and fear about their future chances of participating in work. The exclusion determined how they perceived their chances of returning to work more than the MSD-related problems themselves.
Luk et al., [21].	Multidisciplinary rehabilitation program,	LBP	+Increase in pain control.
	including physical conditioning, work conditioning and work readiness.		+Greater improvement in physical functioning and self-perceived disability.
	A prospective study with 65 participants.		–Older age.
			–Absent from work longer.
			–Having quit their jobs instead of taking sick leave.
Coutu et al., [46].	Intensive interdisciplinary rehabilitation program, including providing reassurance, reducing fears &the avoidance of pain and movement. (A progressive RTW and collaboration with stakeholders.)	Back^*	+Gaining control over their own condition (pain, fears or mood).
	A longitudinal, exploratory descriptive study with 16 participants.		+Finding a new “normal pain”. Probably the most important impact of the work rehabilitation program.
			+Redefining their life priorities.
			–Maintained initial representation of their abnormal pain.
			–Avoidance behaviors (avoiding the idea of getting better, avoidance as proof of the pain being real).
Lindell et al., [13].	Cognitive-behavioral program vs traditional primary care.	Back, Neck	Treatment was non-predictive.
	Randomized controlled trial with 125 participants.		+Low level of previous sick-listing (including all diagnoses). Strongest predictor in two follow-ups (18 &24 months).
			+Young age (max 44 years). Second strongest in one follow-up (18 months).
			Subjective factor:
			+High self-prediction (own belief in RTW). Strongest in one follow-up (12 months).
			o Non-severe functional impairment, health-related QoL and state of health.
			o Being male, high education, good social support, comfortable work postures, high physical workload, co-habitant and non-unemployed.
Coutu et al., [40].	How health and illness representations of MSD workers affect RTW.	Back^*	+Transitioning of worker from a less mechanistic to a more functional view of health (individual responsibility for own health).
	Interviews during an intensive interdisciplinary work rehabilitation program.		+Better physical and mental health at the end of program and at one-month follow-up.
	A quality study – longitudinal, exploratory, descriptive study with 16 participants.		+More active, less pain or they had learned how to control pain.
			–People not returning to work used the term “handicapped” or “sick” of themselves throughout the process.
Durand et al., [41].	Therapeutic return to work (TRW).	MSDs^*	+Interventions in the workplace.
	Comprehensive rehabilitation process centralized in the workplace. A worker’s progressive RTW.		+Development of competent work behaviors.
	A multiple-case study with 20 participants.		+Decrease in fears about pain and movement, improvement in health status, perception of self-efficacy regarding work, and maintenance of physical performance.
			–No shared vision with medical stakeholders, a negative view of the workplace, difficulty in managing pain.
			–For some individuals, viewpoint discrepancy between the interdisciplinary team and the physician attending hindered RTW.
Hamer et al., [33].	Interdisciplinary chronic pain disorder (CPD) treatment program with presence of an RTW coordinator (RTWC).	MSDs^*	+Shorter time from injury to referral (<15 months).
	A retrospective cohort study with 1002 patients.		+Presence of an RTW coordinator had the greatest effect.
			If there was no RTW coordinator, increased time since injury had a negative effect on chances of returning to work, whereas time since injury did not seem to influence the likelihood of RTW when an RTWC was involved.
			+Female gender.
			+Born in the country of intervention: Canada.
			o Age, level of education, PCS (Pain Catastrophizing Scale) score on admission, pre-injury job, depression and level of physical demands.
Jensen et al., [39].	Case manager guidance within a multidisciplinary setting.	LBP	Predictors of unsuccessful RTW after one year:
	A cohort study based on a randomized clinical study with 325 participants.		Interplay between clinical- and psychosocial risk factors.
			Clinical predictors:
			–Pain score (back and leg pain) &side-flexion.
			Psychosocial predictors:
			–Bodily distress, fear avoidance, low expectations of RTW, blaming work for pain, no home ownership &drinking alcohol less than once a month.

^*Other musculoskeletal disorders included, but mostly back, neck and shoulders.

Appendix B

The quality assessment was based on a number of self-defined questions, bearing in mind the tools of CASP (Critical Appraisal Skills Programme) [22] and STROBE (STrengthening the Reporting of Observational Studies in Epidemiology) Statement [23]. Questions used:

Does the article have a clear purpose?

Does the article address a clearly focused problem/issue?

Is the background well described, e.g. literature?

Does the article include a theoretical aspect?

Are the methods

well described?

appropriate for the study?

Have ethical issues been taken into consideration?

Is there a clear statement of the results?

Are the results in line with other available evidence?

Can the results be applied in other populations? Can they be generalized?

Discussion: Are the key results summarized and the limitations discussed?

References

Försäkringskassan. Socialförsäkringen i siffror 2013 (In Swedish). Stockholm: Danagård LiTHO; 2013.

Waddell

, Burton

. Is work good for your health and well-being? United Kingdom: The Stationary Office (TCO); 2006.

Selander

, Marnecroft

S-U

, Bergroth

, Ekholm

. Return to work following vocational rehabilitation for neck, back and shoulder problems: Risk factors reviewed. Disabil Rehabil 2002;24(14):704–12.

Toomingas

, Mathiassen

, Tornqvist

. Arbetslivsfysiologi (In Swedish) Lund: Studentlitteratur AB; 2008.

Trinczek

. Demographic change and work in Europe, [cited 2014 Mar 20]. Available from: www.eurofound.europa.eu.

Ilmarinen

. The ageing workforce - challenges for occupational health. Occup Medicine 2006;56(6):362–4.

Parent-Thirion

, Vermeylen

, van Houten

, Lyly-Yrjänäinen

, Biletta

, Cabrita

, Niedhammer

. Fifth European Working Conditions Survey. Luxembourg; 2012.

Waddell

. Preventing incapacity in people with musculoskeletal disorders. Br Med Bull 2006;77 and 78:55–68.

WHO. Towards a common language for functioning, disability and health. ICF [Internet]. 2002 [cited 2014 Mar 31]. Available from: www.who.int/classifications/icf/icfbeginnersguide.pdf?ua=1.

10.

Loisel

. Intervention for return to work - what is really effective? Scand J Work Environ Health 2005;31(4):245–7.

11.

Hlobil

, Staal

, Spoelstra

, Ariëns

, Smid

, van Mechelen

. Effectiveness of a return-to-work intervention for subacute low-back pain. Scand J Work Environ Health 2005;31(4):249–57.

12.

Koopman

, Edelaar

, Slikker

, Reynders

, vander Woude

LH V

, Hoozemans

MJM

. Effectiveness of a multidisciplinary occupational training program for chronic low back pain: A prospective cohort study. Am J Phys Med Rehabil 2004;83(2):94–103.

13.

Lindell

, Johansson

S-E

, Strender

L-E

. Predictors of stable return-to-work in non-acute, non-specific spinal pain: Low total prior sick-listing, high self prediction and young age. A two-year prospective cohort study. BMC Fam Pract 2010;11:53.

14.

Sjöström

, Alricsson

, Asplund

. Back to work–evaluation of a multidisciplinary rehabilitation programme with emphasis on musculoskeletal disorders. A two-year follow-up. Disabil Rehabil 2008;30(9):649–55.

15.

Johansson

, Aydin

, Bergendorff

, Granqvist

, Josephson

, Sohlberg

. Arbetslivsinriktad rehabilitering (In Swedish). Underlagsrapport nr 7 till den parlamentariska socialförsäkringsutredningen. Stockholm: Elanders Sverige AB; 2011.

16.

ILO. Definition of vocational rehabilitation [Internet]. [Cited 2014 Apr 14]. Available from: www.ilo.org/public/english/standards/relm/ilc/ilc86/r-iii1b.htm#N_43_.

17.

Sandmark

. Perspektiv på kvinnors hälsa i arbetslivet (In Swedish). Lund: Studentlitteratur AB; 2011.

18.

Larsson

, Ahlin

, Brdarski Jacobson

, Goine

. SOU 2000:78. Bilaga 1. Kommittedirektiv. Den arbetslivsinriktade rehabiliteringen (In Swedish). Dir 1999:44. 2000.

19.

Andersson

, Ahgren

, Bihari-Axelsson

, Eriksson

, Axelsson

. Organizational approaches to collaboration in vocational rehabilitation - an international literature review. Int J Integr Care 2011;11(18):1–10.

20.

Aydin

, Odelmark

, Jeffmar

, Kärrholm

. Samverkan inom socialförsäkringen (In Swedish).En sammanställning av uppföljnings- och utvärderingsrapporter inom verksamheter finansierade med samverkansmedel. Rapport 2010-2 Stockholm; 2010.

21.

Luk

KDK

, Wan

TWM

, Wong

, Cheung

KMC

, Chan

KYK

, Cheng

ACS

, Kwan

MWW

, Law

KKP

, Lee

PWH

, Cheing

GLY

. A multidisciplinary rehabilitation programme for patients with chronic low back pain: A prospective study. J Orthop Surg (Hong Kong) 2010;18(2):131–8.

22.

CASP. Critical Appraisal Skills Programme. Making sense of evidence. [Internet]. 2014 [cited 2014 Mar 4]. Available from: http://www.casp-uk.net/#!casp-tools-checklists/c18f8.

23.

STROBE Statement. STROBE Statement Strengthening the reporting of observational studies in epidemiology [Internet]. 2009 [cited 2014 Mar 18]. Available from: http://www.strobe-statement.org/pdf/index.php?id=available-checklists.

24.

CASP. Critical Appraisal Skills Programme. Making sense of evidence. [Internet]. 2014 [cited 2014 May 19]. Available from: www.casp-uk.net/.

25.

STROBE Statement. STROBE Statement Strengthening the reporting of observational studies in epidemiology [Internet]. 2009 [cited 2014 May 19]. Available from: http://www.strobe-statement.org.

26.

Durand

M-J

, Loisel

. Therapeutic Return to Work: Rehabilitation in the workplace. Work 2001;17(1):57–63.

27.

Vowles

, Gross

, Sorrell

. Predicting work status following interdisciplinary treatment for chronic pain. Eur J Pain 2004;8(4):351–8.

28.

Jensen

, Bergström

, Ljungquist

, Bodin

. A 3-year follow-up of a multidisciplinary rehabilitation programme for back and neck pain. Pain 2005;115(3):273–83.

29.

Cheng

AS-K

, Hung

L-K

. Randomized controlled trial of workplace-based rehabilitation for work-related rotator cuff disorder. J Occup Rehabil 2007;17(3):487–503.

30.

Marois

, Durand

M-J

. Does participation in interdisciplinary work rehabilitation programme influence return to work obstacles and predictive factors? Disabil Rehabil 2009;31(12):994–1007.

31.

Staal

, Hlobil

, Twisk

, Smid

, Köke

, van Mechelen

. Graded Activity for Low Back Pain in Occupational Health Care. Ann Intern Med 2004;(140):77–84.

32.

Arnetz

, Sjögren

, Rydéhn

, Meisel

. Early workplace intervention for employees with musculoskeletal-related absenteeism: A prospective controlled intervention study. J Occup Environ Med 2003;45(5):499–506.

33.

Hamer

, Gandhi

, Wong

, Mahomed

. Predicting return to work following treatment of chronic pain disorder. Occup Med (Lond) 2013;63(4):253–9.

34.

Storheim

, Brox

, Holm

, Bø

. Predictors of return to work in patients sick listed for sub-acute low back pain: A 12-month follow-up study. J Rehabil Med 2005;37(6):365–71.

35.

Magnussen

, Strand

, Skouen

, Eriksen

. Motivating disability pensioners with back pain to return to work–a randomized controlled trial. J Rehabil Med 2007;39(1):81–7.

36.

Magnussen

, Nilsen

, Råheim

. Barriers against returning to work–as perceived by disability pensioners with back pain: A focus group based qualitative study. Disabil Rehabil 2007;29(3):191–7.

37.

Lydell

, Baigi

, Marklund

, Månsson

. Predictive factors for work capacity in patients with musculoskeletal disorders. J Rehabil Med 2005;37(5):281–5.

38.

Selander

, Marnetoft

S-U

, Asell

. Predictors for successful vocational rehabilitation for clients with back pain problems. Disabil Rehabil 2007;29(3):215–20.

39.

Jensen

, Stengaard-Pedersen

, Jensen

, Nielsen

. Prediction model for unsuccessful return to work after hospital-based intervention in low back pain patients. BMC Musculoskelet Disord. BMC Musculoskeletal Disorders 2013;14(1):140.

40.

Coutu

M-F

, Baril

, Durand

M-J

, Côté

, Cadieux

. Health and illness representations of workers with a musculoskeletal disorder-related work disability during work rehabilitation: A qualitative study. J Occup Rehabil 2011;21(4):591–600.

41.

Durand

M-J

, Berthelette

, Loisel

, Imbeau

. Validation of the programme impact theory for a work rehabilitation programme. Work 2012;42(4):495–505.

42.

Sullivan

MJL

, Ward

, Tripp

, French

, Adams

, Stanish

. Secondary prevention of work disability: Community-based psychosocial intervention for musculoskeletal disorders. J Occup Rehabil 2005;15(3):377–92.

43.

Moliner

, Durand

M-J

, Desrosiers

, Coutu

M-F

. Subjective quality of life according to work status following interdisciplinary work rehabilitation consequent to musculoskeletal disability. J Occup Rehabil 2007;17(4):667–82.

44.

Cutler

, Fishbain

, Steele-Rosomoff

, Rosomoff

. Relationships between functional capacity measures and baseline psychological measures in chronic pain patients. J Occup Rehabil 2003;13(4):249–58.

45.

Loisel

, Falardeau

, Baril

, José-Durand

, Langley

, Sauvé

, Gervais

. The values underlying team decision-making in work rehabilitation for musculoskeletal disorders. Disabil Rehabil 2005;27(10):561–9.

46.

Coutu

M-F

, Baril

, Durand

M-J

, Côté

, Rouleau

, Cadieux

. Transforming the meaning of pain: An important step for the return to work. Work 2010;35(2):209–19.

47.

Svajger

, Winding

. Perceptions of possibilities of returning to work with chronic musculoskeletal disorders. Work 2009;32(4):443–54.

48.

Watson

, Booker

, Moores

, Main

. Returning the chronically unemployed with low back pain to employment. Eur J Pain 2004;8(4):359–69.

49.

Bryman

, Bell

. Business Research Methods. 3rd Edition. New York: Oxford University Press; 2011.

50.

Ross

, Callas

, Sargent

, Amick

, Rooney

. Incorporating injured employee outcomes into physical and occupational therapists’ practice: A controlled trial of the Worker-Based Outcomes Assessment System. J Occup Rehabil 2006;16(4):607–29.

51.

Collingwood

. Your sense of coherence [Internet]. [cited 2014 May 28]. Available from: Psychcentral.com/lib/your-sense-of-coherence/000241.

52.

Watson

, Main

. Influence of benefit type on presenting characteristics and outcome from an occupationally orientated rehabilitation programme for unemployed people with chronic low back pain. Physiotherapy 2004;90(1):4–11.