Occupational therapy and return to work: a systematic literature review

Figure 1 shows the selection process of articles for full-text analysis (n = 26). Inclusion criteria were:

Vocational rehabilitation studies that did not explicitly describe or mention OT involvement were excluded (see Figure 1).

A preliminary pilot study used to determine how to direct the literature search, showed that OT was often part of multidisciplinary teamwork or program described under "vocational rehabilitation". Therefore, "vocational rehabilitation" was added to the search terms "occupational therapy" and "return to work" (see Figure 1).

By screening the titles, abstracts, and keywords for the terms "occupational therapy" and "return to work" potential studies were identified. Studies published in German [11‐13] were also included. Additionally, using the names of researchers authoring relevant studies, a "snowball search" was used by screening the reference lists of selected articles for pertinent references. Additional research material suggested by OT experts was screened too.

Data extraction from the included papers was performed by one researcher (H.D.) and checked for accuracy by the other authors. Disagreement in data extraction was resolved by consensus. As the structure and content of the OT programs remained vague, except for the study of Lambeek et al., the authors were invited by e-mail between October and December 2010 to provide more information. From 3 studies additional information was provided [14‐16].

Quality assessment was performed by evaluating methodological quality of the studies [17‐19]. Internal validity, study methodology, and external validity were assessed.

Six papers [14‐16, 20‐22] finally met the quality criteria. These six studies included 899 patients older than 18 years (active age) that participated in rehabilitation programs aimed at RTW. All studies included patients - suffering from differing problems - who had jobs at the time of the research project in which they participated and patients were supported by social security system and/or private insurances. The patients were evaluated after the program (1 week to 42 months after discharge). Three of the studies were RCTs; the remaining three were cohort studies. Methodological characteristics and quality are presented in Table 1.

One of the selected studies was of high quality (A score) for all the items. Two selected studies showed good internal validity (score B), and 3 showed moderate internal validity (score C). Four studies were of moderate methodological quality (score C), and only two studies--Vanderploeg et al. [16] and Lambeek [14] --were of high quality (score A). Three studies--Jousset et al. [20], Schene et al. [15] and Lambeek et al. [14] and Vanderploeg et al. [16] did have limited external validity (score C) because they focussed on very specific target groups. Potential bias from selecting patients in insurance-paid programs was recognised by Vanderploeg et al.[16]and Lambeek et al. [14], but not specifically mentioned by the other studies.

Although all studies in this review focussed on RTW and the role of OT in that process, analysis showed many differences which hindered comparison of the studies and their results. All studies showed an effect on RTW in a program in which OT is involved, although large heterogeneity is found. Studies differed in type of intervention, patient type involved, operational definition of the notion RTW and in follow-up period.

All of the selected studies denoted RTW as an outcome measure but their definition of RTW and what RTW involves varied widely (Table 2).

Concluding whether a given intervention has (successful) effects regarding its goal (RTW) depends on the definition of "successful RTW" (see table 2). In the studies reviewed here, both the definition of successful RTW, which ranged from part-time to full-time employment, and the time of follow-up, which ranged from 1 week to 42 months, differed substantially. In the study of Joy et al. [21], successful RTW was measured in terms of the percentage of RTW compared to the situation before participants entered the program. Although the other selected studies also compared different forms of treatment including OT, they did not demonstrate precisely how each professional discipline contributed to RTW.

All OTIs affected RTW. Jousset et al. [20] found significantly lower mean numbers of self-reported sick-leave days in the functional restoration group who took part in OT. Joy et al. [21] suggested that multidisciplinary programs (including OT) such as work-hardening and functional restoration may be of benefit in helping the patients identify and resolve issues that often contribute to disability exaggeration leading to greater RTW success independent of any changes in a patient's overall pain level. This parallels the findings of Lambeek et al. [14], who also concluded that disability decreased although improvement of pain did not differ between both groups. Schene et al. [15] found that adding OT to the usual treatment increased and accelerated work resumption of people suffering from depression. OT, however, did not accelerate recovery from depression. Results of the work of Sullivan et al. [22] revealed that a risk-factor-targeted intervention administered by physical therapists and occupational therapists can have a meaningful impact on RTW following whiplash injuries. The impact of their program was most pronounced for the subgroup of subjects who scored in the risk range on all psychosocial variables targeted by the program. Vanderploeg et al. [16] added (measured by in-person evaluations and structured telephone interviews at 1 year after the programs) the amount of help that the participating patients with traumatic brain injury received and details on any vocational activity over the year since completing the study protocol to their RTW measure. Their study found no difference in RTW between patients that received cognitive didactic and those who received functional experiential approaches during traumatic brain injury rehabilitation.

Effects of OTIs in rehabilitation programs regarding RTW are recognisable, but large differences in settings, design, in- and exclusion criteria, disciplines concerned in the study-programs, and in outcomes made it difficult to determine the extent to which OT contributions to these interventions affected RTW (see table 3). The different OTIs, integrated in the multidisciplinary intervention are:

• Jousset et al. mention "work simulation" as part of the Functional Restoration Program but do not specify what exactly the content of that part of the program was, in which settings it was performed or what the specific approach of the occupational therapist was [20].

• Lambeek et al.[14] mention the contribution of occupational therapists in the description of the study, but in the specifications of different types of therapeutic services, provided in primary and secondary care, the description "occupational therapist" is not used. Nevertheless, the additional description of the protocol of the "integrated care " used in the study, point outs very clearly in what way occupational therapy was used. The main part of the work of the occupational therapist included in the study of Lambeek et al.[14] is to provide a workplace intervention. Being a member of the multidisciplinary team, the OT takes part in gathering patient information. In the additional report [23], a detailed OT protocol is included and supported by an "occupational therapist flow chart", thus indicating the OTI time-span. Every four weeks (telephone) conferences with the clinical occupational physician, physical therapist and medical specialist need to take place. The protocol mentions the issues that need to be discussed and the timeframe for the OTI.

• The content of a therapeutic program item like "work hardening" is mentioned in the work of Joy et al., but is not clear what the therapeutic actions are, what type of approach is used, what activities are performed [21].

• The program of Schene et al.[15] provides three individuals visits in the last phase. The information, separately published in the intervention protocol, clarifies precisely the content of the program [24].

• The OT part of the program of Sullivan et al contains "increasing activity involvement" but it is not specified in the study-report what precisely the therapeutic actions of the occupational therapist were [22].

• Vanderploeg et al.[16] mention that the OT was part of the multi-disciplinary team, but do not give further details on the content of the input of OT. The additionally provided information [25] gives more specific information on the research protocol, however, no specification of the precise content of the occupational therapy part of the program was indicated [25]. The protocol clarifies how the whole team had to collaborate but does not offer a detailed description of the specific actions of each discipline involved.

Whether OT had a meaningful role in the outcome of the different programs, is not only a result of the OT contribution itself, but also of the composition of the services offered by the multidisciplinary team concerned in the program (see tables 3 and 4).

Remarkable in the selected studies was the mixture of terms used to describe multidisciplinary teams. The disciplines concerned in "multidisciplinary rehabilitation" across the six studies show that a great variety of disciplines is involved in RTW programs. Moreover, many of the authors failed to differentiate between occupational therapy, physiotherapy, and physical therapy. Lambeek et al. [14] however, describe precisely the contribution of each discipline in the multidisciplinary team, including a flow chart of the process in which each of those team members was involved in the integrated care protocol.

Recognizing the role that OT plays in the overall therapeutic effort (by using the WFOT definition of the profession [1]), is not obvious, but for experienced OT's nevertheless very well recognisable in the papers by Jousset [20], Joy, Sullivan [22] and Vanderploeg [16, 25]. This finding supports the statement of Lee and Kielhofner that specific evidence of OTIs is lacking.

Schene et al.'s [15] conclusions show that a holistic approach (e.g. psychosocial intervention combined with physical therapy) is useful for preventing loss of capacities (and thereby loss of the ability to work). Moreover, Sullivan et al. [22] suggested that a holistic approach can increase successful RTW by 25%. Vanderploeg et al. [16, 25] determined the contribution of OT (cognitive-didactic versus functional-experiential approaches) during different stages of the therapeutic process. Referring to the definition of OT used in this review, the cognitive-didactic approach can be considered as the OTI in the Vanderploeg et al. study [16]. Although they did not find significant differences after one year of rehabilitation, they found that participants in the cognitive-didactic program showed better post-treatment cognitive performance.