Early vocational rehabilitation after spinal cord injury: A survey of service users

1.1 The New Zealand Spinal Trust Vocational Rehabilitation Service

The New Zealand Spinal Trust (NZST) Vocational Rehabilitation Service provides early vocational services while a person is still in their initial rehabilitation phase at one of the two spinal rehabilitation centres in NZ (Kelly, 2017). The service is dedicated to supporting people to return to meaningful and sustainable employment and has a ‘client-centric’ focus designed to promote empowerment. The service aims to reduce the fear of work and remove identified barriers. Spinal Units in both New South Wales and Victoria, Australia, have successfully implemented this model (Bloom et al., 2017; Hilton et al., 2017; Middleton et al., 2015). The NZST program undertakes vocational rehabilitation with people very early in their rehabilitation, with 80% of clients engaged within three weeks following their SCI. Specialist Vocational Consultants provide individual career coaching with a strong emphasis on clients doing as much of the work as practical, and where possible, assisting the client maintain engagement with their employer. They focus on reducing risk for developing attitudinal and psychological barriers such as negative recovery expectations and workplace factors, while also fostering hope and optimism (Kelly, 2014). The benefits of access to early vocational services integrated into specialised rehabilitation have been noted (Fadyl & McPherson, 2010) and can include building positive perceptions and confidence in thinking about returning to work.

While the theoretical benefits of early intervention vocational programmes have been discussed in the SCI vocational literature, there is a lack of data supporting their efficacy and outcomes (Middleton et al., 2015). Our preliminary work has shown that the innovative early intervention and targeted support provided by the NZST programme can improve RTW rates for people with new SCI (Kelly, 2017; Kelly & Bourke, 2017). However, it is not known who best benefits from early vocational input, what mechanisms of intervention effect can facilitate optimal RTW outcomes and when these should be introduced.

1.2 Study objectives

This study’s objective was to survey people with SCI who had been discharged from the two NZ spinal units within the previous five years and had accessed early vocational rehabilitation from NZST. We aimed to describe the demographic and clinical variables characterising this population, including their RTW outcomes. We also aimed to identify the components and resources, as well as the causal processes (that is, mechanisms of effect) between the NZST programme and RTW outcomes that could be discerned from participant experiences.

2.1 Study design

A cross-sectional online survey recruiting participants by email invitation was used to capture the views of people with SCI who received early intervention vocational rehabilitation from the NZST Vocational Rehabilitation Service.

The survey study sits within a broader study called the early vocational rehabilitation in neurological conditions (EVOCS) study which is a realist mixed methods study examining the mechanisms of effect and contexts that impact on RTW outcomes among people with neurological impairments. The survey component of the EVOCS study was intended to provide information about the perspectives of people with SCI who had received early vocational rehabilitation from NZST, particularly focusing on intervention mechanisms of effect derived from earlier phases of the study.

2.2 Survey development and piloting

The survey was developed by team discussion and from emerging theory from earlier phases of the EVOCS study (see Dunn et al., 2021). The aim was to gather demographic, clinical and employment outcomes to provide relevant background information for the wider EVOCS study. The survey comprised 82 questions across six broad sections and we used branching logic to minimise redundant questions as participants completed the survey.

The first section of the survey focused on demographic questions including age, gender, ethnicity, region and education. The second section included questions about SCI history such as year of injury, injury type and cause, funder of rehabilitation and mobility status. The third section captured information about participants’ social and living situation, including relationship status. The next two sections focused on pre- (at time of SCI) and post-injury employment such as employment category and role, job satisfaction and RTW outcomes after SCI. The final section covered the mechanisms of effect regarding the NZST vocational service derived from the realist literature review and qualitative interviews conducted during the first phases of the EVOCS study.

The final survey was piloted with six people with SCI and providers of vocational services. Minor modifications were made to wording and layout to improve clarity following pilot feedback. Questions included dichotomous yes/no, multiple choice and free text written options. Based on the pilot study feedback, the time taken to complete the survey was approximately 20 minutes.

2.3 Participant recruitment and data collection

The participants of this study were adults with SCI aged 18 or older who:

Sustained an SCI and discharged from either one of the two NZ spinal centres (Burwood Spinal Unit or Auckland Spinal Rehabilitation Unit) between 1 Jan 2015–31 December 2019

The NZST Vocational Rehabilitation Service delivers rehabilitation to approximately 85 people with newly acquired SCI each year admitted to either of the two NZ spinal centres. All people of working age or who are currently in work are offered EIVR, except if clinical staff explicitly state the EIVR team should not engage. Potential participants meeting the inclusion criteria were identified from the NZST database and invited by email by NZST staff to participate in the study. The email included a REDCap (Research Electronic Data Capture; Harris et al., 2009) hyperlink to the survey. Participants could also complete the survey by requesting a hard copy be posted to them. Only one email reminder was permitted to be sent by the NZST and this was sent two weeks after the first email invitations. The survey was live between 4 November 2020 and 7 December 2020. To protect the confidentiality of the study participants, no identifying information was requested during the survey.

2.4 Data analysis

Data were analysed using IBM SPSS Statistics for Windows, Version 26.0. Any ineligible responses were removed before exporting the dataset to SPSS. Contingency tables (cross-tabs) were calculated for discrete variables and chi-squared tests determined significance of 2-way associations. For continuous variables, independent-samples t-tests were used. Results are descriptive and reported as frequencies, percentages, means and standard deviations, medians and interquartile ranges, and odds ratios (OR) with 95% confidence intervals (CI), as appropriate.

We used list-wise deletion, the default SPSS approach, to account for missing data. We believed this was appropriate because of the descriptive nature of the data analyses and sample size. Missing data are shown by reporting sample sizes for the various variables examined.

The free text responses at the end of the survey were analysed using qualitative description (Sandelowski, 2000). This is a straight-forward approach when plain descriptions of phenomena are preferred. Ideas emerging from the written responses were categorised by one researcher (DS) and reviewed by the team. First, responses were collated, read and re-read. The responses were then grouped into preliminary categories and then grouped together based on similarities to form a smaller number of categories and then revised and refined again following discussion between the research team. The quantitative survey results and qualitative free text responses were integrated using a triangulation approach (O’Cathain, Murphy & Nicholl, 2010). After component data were separately analysed we looked for agreement, disagreement or silence across data from the respective methods. We particularly looked at integrating quantitative findings related to mechanisms of EIVR with free text responses because this was most relevant to our research questions.

The EVOCS study, including this survey, received ethical approval from the University of Otago Human Ethics Committee (ref H19/170). In line with this approval, all participants provided informed consent before being able to proceed to the survey.

3.1 Description of study sample

Table 1 provides a summary of participant demographic characteristics (n = 37). There were similar numbers of male (49%) and female participants (50%) and the average age of the sample was 50 years (SD 15, range 18–71). The sample was largely NZ European (78%). While we captured participants living in a broad range of regions, the largest proportions, not unexpectedly lived in the two larger urban centres where spinal units are located (Auckland and Christchurch). The sample was also highly educated, with more than half (57%) reporting university-level study.

Table 2 summarises the employment status among participants at the time of their SCI. A majority of participants reported they were in full time paid work at the time of their SCI (68%) and felt satisfied or completely satisfied with their employment situation (77%). In terms of work category, n = 26 described their work type or role. Of these, 14 said they were in management roles, where management role was defined as a senior-level role where the individual was likely to have at least some autonomy and control over their own RTW process.

Table 3 summarises the post-SCI employment characteristics of participants. Of the whole sample, just over half reported they returned to work after their SCI (51%), with a majority of this group stating they returned to their pre-injury employer (63%) and role (58%). A majority also reported feeling completely or somewhat satisfied with their employment situation at the time of survey completion (61%). Of those who had returned to work, just over half said they returned to work within 12 months of their SCI (58%) and most said they stayed working in the same role with the same employer (74%) once they returned to work. A majority reported they were still employed at the time of completing the survey (90%). There was no association between pre-injury job satisfaction and RTW outcome following SCI.

Considering associations between RTW outcomes and demographic and clinical variables, although the numbers are small, those in management roles had higher odds of returning to work within 12 months of injury (OR = 3.8, 95% CI 0.6, 22.3; p = 0.04), to the same employer (OR = 4.2, 95% CI 0.7, 24.5; p = 0.01) and to the same role (OR = 3.3, 95% CI 0.6, 20.17; p = 0.09) compared with those in non-management roles. There were no associations identified between RTW outcomes and demographic or clinical variables such as age, gender, ethnicity, education, SCI type, rehabilitation funder or mobility status. There was a relationship between those reporting their household income was enough or more than enough to meet needs before SCI and having returned to paid work following SCI (OR = 2.43, 95% CI 1.24, 4.75, p = 0.02), compared with those reporting their income was not enough or just enough before their SCI.

In this section of the survey, we asked participants to consider aspects of services provided by the NZST Vocational Rehabilitation Service, including those resources that we had identified in earlier phases of the EVOCS study. These findings are displayed in Table 4. The key finding that emerged from these data was the importance of feeling hopeful about returning to work. There was an association between feeling hopeful about returning to work while still in the spinal unit and RTW outcome (OR = 2.25, 95% CI 1.13, 4.50; p = 0.04). There were non-significant trends to those finding support to talk through options, discuss realistic expectations, explore adaptations, and successfully returning to work.

Of the full sample, 17 people wrote comments in the free text box at the end of the survey (“Is there anything else you would like to tell us”). Reviewing these comments revealed similar ideas to those emerging from the quantitative analyses. We organised the comments into four broad areas. First, participants described challenges around timing and setting expectations for themselves for RTW because they were still learning about the impact of their SCI:

These comments contrasted with responses to closed survey questions that indicated just over half of the sample felt they had been approached at just the right time to discuss return to work (51.4%) and a majority felt hopeful about returning to work while still in the spinal unit (64.9%).

Second, participants described the need to rebuild confidence about their capabilities in order to consider RTW but that this can take time and is a slow process:

Responses to the closed survey questions echoed these sentiments. For example, many participants endorsed that EIVR input helped them feel confident and optimistic about returning to work (59%) and assisted them work through hurdles and options (59%).

Third, participants noted the impact of worrying about whether their employer would keep their job for them and would support and accommodate their RTW:

This aspect was not particularly captured by responses to closed survey questions. A small number endorsed the option of assistance to engage with the employer as helpful (19%). No participants agreed that EIVR input involved liaising with the employer on their behalf, consistent with the stated intentions of the NZST vocational service as supporting their clients to do much of this work themselves.

Finally, while it was positive when expectations matched the process: “The process was good and very well thought out in relation to my expectations of recovery” (ID # 23, male, 60y), confusion about the RTW process was also identified. For example, in relation to conflicting expectations between the individual and various organisations involved, one participant said:

The free text comments supported the responses to closed survey questions around the importance of developing positive expectations and hopefulness about RTW (see Table 4). However what was not captured by closed survey questions was confusion experienced when expectations of others conflicted with the individual’s own expectations of RTW. In addition, free text comments highlighted participant experiences of revising their expectations as they grappled with living with their SCI:

4.1 Clinical implications

Our findings underscored the value participants placed on feeling hopeful about RTW very early following injury and suggested this as being a key foundational mechanism. However the free-text comments remind us that not everyone feels ready to consider RTW goals during their acute rehabilitation, emphasising the importance of individualised input. Associated with this are reflections on loss of confidence in one’s abilities, the need to establish realistic goals and to adjust one’s expectations with experience. This suggests that to sustain RTW hopefulness in terms of agency and pathways, there is a need to be alert to individual facilitators, barriers and continuity of support beyond the acute setting. The NZST team’s input generally ceases at the point of discharge and this may risk losing momentum and RTW hopefulness gained during the acute inpatient phase.

We noted that optimal RTW outcomes in our sample seemed associated with the ability to return to the same employer and role. Hence supporting the person with SCI to stay connected with their pre-injury employment context is also likely to help sustain hope and promote faster RTW. Participants identified loss of connection with one’s employer and fears the employer may not understand and/or want to wait long enough as key concerns that risk undermining hopefulness about RTW. This is another reason to address continuity between acute and post-acute vocational rehabilitation.

4.2 Limitations

This study has a number of limitations. First, the recruitment strategy and associated restrictions resulted in a small sample with a low response rate. Our response metrics however are comparable to those of a recent Australian study surveying vocational outcomes of people with SCI. In this previous study a response rate of 20% and a sample size of n = 34 were reported (Dorsett & McLennan, 2019). It is common among survey studies to use multiple strategies to optimise recruitment, including use of paper-based and online response options, repeat invitations and reminders to ensure those who would like to complete the survey have the opportunity to do so. Additional strategies such as snowballing methods can also promote higher response rates, although they also have their limitations (Ritter & Sue, 2007a, 2007b). Our response rate means the survey findings cannot be assumed to reflect the views of the wider SCI population of adults who may have accessed early vocational interventions. Our study and the study by Dorsett and McLennan, likely reflect the difficulties recruiting from the community of people with disabilities (Bourke et al., 2016).

Second, the characteristics and outcomes described by participants diverge somewhat from previously published work, for example in relation to gender balance and education levels. While such diverging characteristics could reflect changing SCI demographics as noted earlier, participants responding to our survey may have also been those who had better access to technology and/or those with more to say about their RTW experiences. These issues likely reflect selection bias and the self-selecting nature of those responding to the survey. In addition we asked participants to think back to their acute SCI admission, introducing the risk of recall bias, especially for those whose injury was more than 12 months ago.

Third, we used a cross-sectional design across a wide time frame (five years), so we cannot make comments about longitudinal outcomes given diversity among the sample with respect to time post-SCI. Future research following a cohort over time will be better placed to examine longitudinal RTW outcomes.

Fourth, we did not collect information on severity of injury beyond paraplegia vs tetraplegia and mobility status. It is widely reported that completeness of injury is associated with RTW outcomes, with those with more complete injuries have poorer RTW outcomes (Yasuda et al., 2002).

Despite these limitations the study provides insights into the perspectives of people with SCI who have experienced early vocational rehabilitation and draws out information on the components or potential mechanisms that may be associated with outcomes. This provides a platform for more in-depth evaluation of the causal mechanisms and contextual factors facilitating RTW outcomes, including who such programmes work for best and in what circumstances.