Work and Employment

Interventions for Enhancing Employment

Access to vocational counselling, educational or job training has often been mentioned as a key issue in enabling return to work after SCI (Jang et al. 2005; Jongbloed et al. 2007; Lidal et al. 2007). However very few studies have empirically tested strategies to increase job opportunities and most reports we found were either case series or observational studies. Various strategies were described through case studies that had successful return to work and job retention. These individualized strategies addressed activities of daily living and mobility needs, job accommodation including workplace support, and employers’ needs and concerns. In this section we reviewed intervention studies examining strategies which lead to return to work.

Table 6. Interventions for Enhancing Employment Post-SCI

Author Year; Country Score Research Design Total Sample Size	Methods	Outcome
Ottomanelli et al. 2013; USA PEDro=5 RCT Level 2 N=157	Population: Veterans with SCI between the ages of 18 and 65 who received health care services in the SCI Centers at one of six participating Veterans Affairs Medical Centers. Experimental: n=81 (mean age 48.7) Control: n=76 (mean age 49.8). Treatment: Experimental group members received Supported Employment (SE) services by a vocational rehabilitation counsellor who was trained in the Individual Placement and Support Model, and integrated as provider among the SCI interdisciplinary care team in the SCI Center. Control groups: group members received Treatment as usual and received referrals to vocational rehabilitation services outside the SCI Centre. Data was collected for 12-months. Outcome measure: Competitive employment in the community (paying job earning at least minimum wage).	Employment: Among the 157 participants, 33 participants (21.0%) accounted for 88 total jobs. 24 participants in the SE group accounted for 60/88 jobs (68.2%). The rate of employment for SE participants was significantly greater (29.6%) than the control (11.8%). SE participants accounted for 50 of 72 (69.4%) jobs (competitive employment) and were significantly more likely to achieve employment (25.9%) compared to control (10.5%). SE participants worked significantly more hours per week (22.0 vs. 17.0), averaged significantly fewer wages ($233.9 vs. $267.3), and missed fewer hours per week (0.3 vs. 1.8).
Ottomanelli et al. 2012; USA PEDro=5 RCT Level 2 N=201	Population: 201 veterans with SCI (192M 9F) between the ages of 18 and 65 who received medical and/or rehabilitation care at 1 of 6 participating centers. Experimental: n=81 (mean age 48.7) Control: n=76 at intervention site (mean age 49.8); n=44 at observational site (mean age 45.1) Treatment: Experimental group members received a supported employment (SE) intervention based on an Individual Placement and Support (IPS) model. There were two control groups: one at the intervention sites through which individuals were randomly assigned to the control group – treatment as usual – intervention site (TAU-IS) and 1 at sites were the SE intervention was not available. All individuals at these observational sites received treatment as usual – TAU (TAU-OS). Data was collected for 12-months. Outcome measure: Competitive employment in the community (paying job earning at least minimum wage).	Individuals in the SE group were 2.5 times more likely than individuals receiving TAU-IS and 11.4 times more likely than individuals receiving TAU-OS to obtain competitive employment. The rate of employment for SE participants was significantly greater than that of either the TAU-IS group or the TAU-OS group. Intent to treat analysis found that participants in the SE group earned significantly more per week than the TAU-OS group. Participants in the SE group earned significantly more per week than participants in both the TAU-IS and TAU-OS groups.
Allen and Blascovich 1996; USA PEDro=6 RCT Level 1 N=48	Population: All individuals were classified as having severe ambulatory disabilities. Experimental: n=24 (SCI: n = 11, 7M 4F) Control n=24 (SCI= 11, 7M 4F) Treatment: Experimental group members received trained service dogs 1 month after the study began. Wait-list control group received dogs in month 13. Participants included individuals who had expressed interest in a service dog and who required substantial personal assistance. Data was collected for 2 years Outcome measure: Spheres of Control Scale (to assess internal locus of control), Rosenberg Self-esteem Scale, Affect Balance Scale (to assess psychological wellbeing), Community Integration Questionnaire, and data regarding the number of received paid and unpaid assistance.	The experimental group had significant improvements on all psychosocial status tests at months 6 and 12 when compared to the control group. The experimental group had a significant decrease in hours of assistance needed at months 6 and 12 when compared to the control group. After receiving a service dog, there were no significant differences between the groups at the same relative data points (months 0, 6 and 12 for the control group, months 12, 18 and 24 for the wait-list control groups. After 12 months, the presence of the service dog was associated with a decrease of 68% of biweekly paid assistance hours. After receiving a service dog, all participants reported substantial increases in terms of school attendance, part time employment, increased levels of social interaction and use of public transportation.
Shem et al. 2011; USA Longitudinal Level 2 N=39	Population: 39 participants with SCI (28M 11F); age 16–26 years. Average(SD) age of mentees was 19.8(3.0) years. 17 employed mentors. In total, 29 participants were matched with mentors, and 10 participants (34%) completed the program Treatment: Each mentee with SCI was matched with a community-based mentor, with or without a disability. The mentoring relationship was planned for 2 years. Participants were evaluated with standardized questionnaires at intake, 3 months after entry, every 3 months thereafter, at the time of post-secondary education or employment entry and 4 months post entry. Outcome measures: return to school, return to work.	7 (24%) participants returned to school; 2 (6.9%) participants returned to work 1 (3.4%) participant returned to school part-time. For mentees who successfully completed the program, there was a trend for improvement in cognitive independence and occupation measures of Craig Handicap Assessment and Reporting Technique, and statistically significant improvements were found with Participation Index of the Mayo-Portland Adaptability Inventory-Version 4, Disability Rating Scale and Supervision Rating Scale, but not with the Satisfaction with Life Scale.
Ottomanelli et al. 2015; USA Case-control Level 3 N=81	Population: 81 military veterans with SCI, average age (SD) 48.7 years (9.8), average time since injury (SD) 11.7 years (11.2), AIS Level of injury – A: 32.5%; B: 13.8%; C: 22.5%; D: 31.3%. Treatment: This study was part of a larger 3-year randomized control trial comparing EBSE to TAU provided for 12 months each to unemployed Veterans with SCI who were 18 to 65 years of age and receiving medical and/or rehabilitation health care services at 1 of 6 VHA SCI Centers. All participants received standardized evidence-based supported employment (EBSE) with activities including integrated vocational and medical rehabilitation treatment, rapid engagement in job finding, competitive employment, inclusion regardless of severity or type of disability, ongoing job support, and focus on participant preferences. Outcome measures: The IPS Fidelity Scale was used to measure the distribution of vocational services and time of those services delivered by vocational counselors. Mean time reflects average time per documented activity. Comparisons were made between groups that gained competitive employment (CE) and those that did not.	Competitive employment (CE) rates during 1 year of evidence-based supported employment for persons with spinal cord injury (N = 81) was 25.9%. There was a statistically significant difference observed between groups; participants obtaining CE were more likely to receive job development (26.6% vs 20.7%), job placement (1.3% vs 0.3%), and employment follow-up (8.4% vs. 2.2%) and less likely to receive vocational counseling (15.3% vs 28.4%).
Sinnott et al. 2014; USA Case series Level 4 N=1578	Population: 157 participants with SCI; average age = 48.7±9.8yo; time since injury 10.7±11.3y Treatment/Methods: A vocational rehabilitation program of Supported Employment (SE) for veterans with SCI; participants were randomly assigned to the intervention of SE (n=81) or treatment as usual TAU (n=76). Outcome measures: Costs and quality-adjusted life years were estimated from the Veterans Rand 36-Item Health Survey and extrapolated to Veterans Rand 6 Dimension utilities.	Average cost for the SE intervention was $1,821. In 1 year of follow-up, and produced fewer QALYs (n.s.), suggesting that SCI-VIP was not cost-effective compared with usual care. An intensive program of SE for veterans with SCI was more effective in achieving competitive employment but was not cost effective after 1 year of follow-up. Longer follow-up and a larger study sample will be necessary to determine whether SE yields benefits and is cost-effective in the long run for a population with SCI.
Dorstyn et al. 2019; Australia Pre-post Level 4 N=5	Population: 5 people with SCI; mean age 46.4 +/-10.2yo; 4 females initially reviewed Work and SCI; Twenty-four with SCI/D subsequently enrolled, of whom 16 (mean age 46.4 years, SD = 11.1; 7 female), completed the intervention. Intervention: Intervention participants accessed the email-based information package (Work and SCI) over a 4-week period. Outcome Measures: My Vocational Situation Scale, Job Procurement Self-Efficacy Scale, Patient Health Questionnaire-9, and Life Orientation Test-Revised	Reliable change in pre-post scores across outcomes were reported by 38% (n = 6) of participants. Favorable comments on the (Work and SCI) resource were provided in addition to suggestions for improvement. Preliminary data suggest that (Work and SCI) may help to establish vocational interests among jobseekers with a SCI/D, however further work is needed to enhance participant compliance. This might include moderator support to promote and maintain participation.
Phillips et al. 2012; United States Case Series Level 4 N=111	Population: Newly injured individuals at an Atlanta rehabilitation. Mean(SD) age: 35(11.8) years; 78% male; 76% white. Treatment: Video-based telerehabilitation intervention (9 weeks); telephone-based telerehabilitation intervention (9 weeks); standard follow-up care. Outcome Measures: Time to productive activities (attending school, VR, working as a homemaker, volunteering) from injury. Time to employment from injury date among individuals employed prior to injury.	Being in one of the intervention groups (either phone- or video-based telerehabiiltation) trended towards a longer time to return to productive activities. Being in one of the intervention groups did not have a significant impact on the time to return to employment for individuals that were employed prior to injury.
King et al. 2004; USA Case series Level 4 N = 174	Population: 174 participants with SCI up to 12 months post-discharge from inpatient rehabilitation. No other demographics given. Treatment: An enhanced case management program (Marcus Community Bridge Program) assisting people to return to the community and to return to work or educational training. Outcome measure: Rate of return to work or educational training at 1-year post-discharge.	One year after discharge the rate of return to work was 17% (i.e. identical to the rate reported by the U.S. Model Systems) and the rate of return to educational training was 31.6% (compared to 15.3% reported by U.S. Model Systems)
O’Neill et al. 2017; USA Prospective Study (without controls) (Conference Abstract) Level 4 N=54	Population: 54 participants; 75% males 25% females; mean age = 37±13yo; level of injury 37% tetraplegia, 30% paraplegia, 33% non-traumatic SCI Intervention: The intervention consisted of a vocational resource facilitator (VRF) being the single point of contact providing medical/vocational case coordination to inpatients and outpatients to ensure the continuity of vocational rehabilitation services upon discharge and long-term follow-up in the community. Outcome Measures: interest in pursuing employment, return to work	At time of discharge, 48% of participants remained interested in pursuing employment. 81% of these outpatient individuals were referred for state vocational rehabilitation services, with 17 actively engaged in the vocational rehabilitation process. Almost half of all eligible inpatients remained actively engaged in pursuing employment after discharge with some returning to work immediately and others actively working with the state vocational rehabilitation agency to secure competitive employment. 23% outpatient individuals returned to work: 15% to same employer-same job and 8% to same employer-different/modified job. None who returned to work received state sponsored vocational rehabilitation services; although two were referred for services, but were denied due to income restrictions. Preliminary findings indicate considerable interest in employment among newly injured persons with SCI.
Rowell and Connelly, 2010; Australia Observational Level 5 N=181 (SCI n=109)	Population: 181 respondents; 73.5% male; mean age: 44 years; 61% unmarried; mean time since injury: 18 years; 39% in labour force and 26% employed. Treatment: no treatment per se but examines the impact of a publicly funded set of services to enable return to work i.e. Adult Lifestyle Support Packages e.g. support with activities of daily living Spinal Injuries Survey Instrument (SISI) developed and administered, Short Form-36 (SF-36) and modified SF-36 administered. Outcome Measures:Labour market outcomes, exposure to the Adult Lifestyle Support Packages (ALSP), clinical and demographic covariates	No statistically significant effect of either the ALSP or support packages from private insurance sources (i.e. PPSP) on labour market participation was found. A number of other factors are significantly correlated with labour market participation: individuals who undertook education or training post-SCI were more likely to be labour market participants females were less likely to be labour market participants a positive attributional style is associated with a higher likelihood of labour market participation a weak non-linear age effect was detected, which suggests that the probability of labour market participation is decreasing in age The marginal effects for the ALSP are statistically insignificant. Thus, the hypothesis that the ALSP has a zero effect on labour market participation cannot be rejected. The strongest marginal effect is for post-SCI education, which is statistically significant at the 1% level and for which the 95% confidence interval is 0.108–0.503. This suggests that post-SCI training and education has an important effect on labour market participation. The probability of labour market participation is increasing in the ln (Attributional Style index, positive scenario). The higher the individual’s propensity to ‘‘internalize’’ positive employment outcomes to his/her own attributes (or ‘‘capabilities and functionings’’), the more likely he/she is to be a labour market participant.
Hansen, 2007; India Observational Level 5 N= 46	Population: 46 participants with SCI (40M 6F). No other demographics given. Treatment:Participation in the work rehabilitation program with the Center for Rehabilitation of the Paralyzed. Program includes physical conditional, vocational training and work placements. Outcome measure:Vocational status.	23 individuals returned to work: 18 participants were employed in a job similar to their pre-injury job; 5 were employed in a different occupation than what they were doing pre-injury. Of the 23 individuals that returned to work 4 used a wheelchair, and 5 used crutches.
Jongbloed et al. 2007; Canada Observational Level 5 N=357	Population: 357 participants with SCI (243M 114F); 92 with complete tetraplegia, 142 with complete paraplegia, 108 with incomplete SCI, 15 unknown; mean age = 46. Treatment: Report on access to vocational counselling and job retraining. Outcome measure: Mailed questionnaire inquiring about factors influencing employment.	Social, economic and political environmental factors contribute to individuals working less than desired. Personal reasons were the most influential. Vocational counselling and job retraining were the most important factors in obtaining employment. Other factors were access issues, attendant care, willing employers, personal presentation and the chance to prove oneself. The impact of policies of government and third party payers were cited as having both positive and negative effects on reemployment.
Jang et al. 2005; Taiwan Observational Level 5 N=169	Population: 169 participants (147M 22F); 32 participants with incomplete paraplegia, 86 with complete paraplegia, 24 with incomplete tetraplegia, 27 with complete tetraplegia; mean age = 39. Treatment: Report on access to vocational training. Outcome measure: Employment status, vocational training	88% were gainfully employed at time of injury; post-injury 79% were employed full time, 21% part-time, 53% were unemployed, 5% attended school or vocational training 50% of those employed received vocational training compared to only 28% of unemployed. Predictive factors of return to work include greater duration post-injury, higher level of education, being married, independence in use of public and private transportation, higher Barthel Index score, age at injury <25 years, and receiving vocational training after injury.
Wang et al. 2002; Taiwan Observational Level 5 N=91	Population: 36 participants with SCI (29M 7F); 13 participants with tetraplegia, 23 with paraplegia; from the Asylum Center Spinal Cord Injury (ACSCI); age range: 18-49; 11 complete, 25 incomplete. 55 participants with SCI (47M 8F); 21 with tetraplegia, 34 with paraplegia; from the Spinal Cord Injury Association of the Republic of China (SCIAROC); age range 18 – >60; 16 complete, 39 incomplete. Treatment: ACSCI group: training program with 6 months of training including: psychosocial consulting, functional, strengthening exercises, endurance, and vocational training; SCIAROC: no specific training program. Outcome measure: Employment status, self-reported Functional Independence Measure (SRFIM).	All participants in the SCIAROC group had no ACSCI training. All participants with tetraplegia were unemployed; 1 subject with paraplegia was a student, 11 were employed, and 22 were unemployed. Employment rates in the SCIAROC group were related to the level of functional independence and injury level. ACSCI group: all 36 participants were unemployed because they were just completing the ACSCI program. Individuals with tetraplegia in the ACSCI group showed significantly better functional independence than those in the SCIAROC group.
Cotner et al. 2018; USA Qualitative N=82	Population: 82 service providers in the VA gave 130 interviews over the course of the 24 month vocational program. Intervention: Individual placement and Support (IPS). Outcome Measures: Interviews were conducted every 6 months at each site by two or three qualitative researchers using an open-ended, semi-structured interview guide. Interviews were conducted to determine barriers and facilitators to employment and implementation of the IPS program.	Twelve barriers to IPS implementation were identified including: obtaining resources, caseload size and area, veteran-specific factors (e.g., low motivation, fear, lack of transportation, etc.), provider education, hiring, provider turnover and integration of vocational rehab counselors (VRs) into the SCI clinical care team, time management, and lack of leadership/salesperson type. Facilitators included: integration of vocational and clinical team, engagement of SCI providers, fit of IPS model, audit and feedback, and obtaining resources. Some of the named barriers and facilitators were the same, indicating that they could be key components to a program going well or going poorly, or that different parts of implementation were required at different times.

Note: A 2012 study by Kolakowsky-Hayner et al. was excluded based on the fact that individuals with SCI only constituted 29.8% of the sample population, and there was no specific analysis or coefficients that would enable understanding of the SCI specific subsample. The SCIRE criteria states that over 50% of the sample must be individuals with SCI for inclusion if a subgroup analysis is not performed.

Discussion

This review suggests that there is a profound lack of high-level evidence studies which have a focus on work and employment-related interventions. Three randomized controlled studies were found, with Allen and Blaskovich (1996) examining access to trained service dogs, and suggesting an improvement in psychosocial status including self-esteem, internal locus of control, and overall psychological well-being. Other benefits of having trained service dogs were a decrease in assistance time by either a professional assistant or family and friends, an increase in school attendance and part-time employment, and an increase in social participation and community (Allen & Blascovich 1996).

The second randomized trial by Ottomanelli et al. (2012) found that a supported employment (SE) intervention was more effective at returning veterans to work than treatment as usual (TAU). Those receiving SE were 2.5 times more likely than those receiving TAU at the intervention sites (offering both SE and TAU) and 11.4 times more likely than those receiving TAU at the observation sites (offering only TAU) to achieve employment over the 12 month follow-up period. However, subjects in the SE group earned significantly less per week than those in the TAU intervention site group. This study was followed up by a third randomized trial (Ottomanelli et al. 2013) which reported that the supported employment (SE) intervention participants had a significantly higher rate of employment than the control, worked significantly more hours per week and missed significantly fewer hours of work.

There was one prospective studyby Shem et al. (2010) which found that participants who completed a mentorship program improved their functioning, independence, and participation, which may have contributed to their favorable return to work (or return to school) outcomes. However, it was not clear from the study whether or not participation in the mentorship program was directly related to employment post-SCI; hence existence of evidence is uncertain in this case. Other studies included case series and observational studies. These studies examined employment outcomes of people with SCI who received various vocational rehabilitation services. One study (Inge et al. 1998) suggests that people enrolled in a program using person-centered planning tools to identify needs and to direct the job search might gain employment but the workplace support greatly varied- from minor to intensive support. Another study (King et al. 2004) described a modified case management approach to return people with SCI to work. Comparing their preliminary results with those of the U.S. Model Systems, it appears that the program is successful for increasing return to educational training but not to work. Marini et al. (2008) suggest that people with SCI registered in state vocational rehabilitation agencies and receiving job placement services are likely to have a higher employment rate. Likewise, Jellinek and Harvey (1982) supported the conclusion of higher employment rate in individuals with SCI who had access to on-site professional counsellors for vocational / educational rehabilitation in addition to state vocational rehabilitation agency, compared to the state vocational rehabilitation agency only. They concluded that the vocational or educational placement was as high as 78% among those who had on-site vocational or educational services. In their sample of 169 people (49% engaged in gainful employment), Jang et al. (2005) found that fifty percent of the employed had received vocational training, compared with only 28% of the unemployed. Jongbloed et al. (2007) also found that employment re-training and education were identified as important contributors to success. However, the participants stated that services and information were perceived as difficult to access. Another observational study examined whether a publicly funded set of support services such as help with activities of daily living is associated with labour market participation. The authors found no effect of these services on labour market participation compared to support packages from private insurance sources (Rowell and Connelly 2010). Hence, evidence cannot be ascertained in this case. However, the same study found that an individual’s propensity to internalize positive employment outcomes in relation to his or her capabilities may contribute to returning to work.

Two other studies examine vocational interventions. Wang et al. (2002) compared a group of persons with SCI receiving a multimodal 6-month training course to a group without specific training. They found that individuals with paraplegia had higher employment rate which indicated an association between level of injury and employment. Hansen et al. (2007) interviewed male participants with SCI in a work rehabilitation program which included physical conditioning, vocational training, and work placements. Less than half were employed in a similar or identical job as their previous employment and only about a quarter of those who used a wheelchair returned to work. Overall, the studies included in this review investigated different types of interventions and used different measures to assess the interventions. Although this may limit the generalizability of the outcomes, there is evidence in general supporting the use of interventions to enhance employment post-SCI.

One study evaluated a telerehabilitation intervention and included employment as an outcome (Phillips et al. 2012). The intervention arm included a nine week telephone or video-based telerehabilitation intervention (not focused on employment) compared with care as usual. Return to employment was analyzed in those that were employed pre-injury; with those receiving the intervention not returning to work any faster than those receiving standard care.

Conclusions

There is level 1 evidence (Trenaman et al. 2014) suggesting that supported employment shows the strongest evidence that it can improve employment outcomes amongst individuals with SCI. Service dogs have also been shown to increase employment.

There is level 1 evidence (Kent & Dorstyn et al. 2014) that 3 psychological constructs: affective experiences, quality of life, and life satisfaction could be considered clinically important in their effects on employment.

There is level 1b evidence (Allen and Blascovich 1996) that suggests a service dog improves integration and participation in school and employment and decrease the number of hours of paid assistance after the first year.

There is level 2 evidence from two studies (Ottomanelli et al. 2012; Ottomanelli et al. 2013, Ottomanelli et al. 2015) that suggests that a supported employment intervention improves employment rates compared with treatment as usual over a one-year period, increases the number of hours worked per week and decreases the number of missed hours of work.

There is level 5 (Jellinek and Harvey 1982) and level 4 evidence (Marini et al. 2008) that on-site vocational rehabilitation counselling during inpatient rehabilitation can increase employment rates.

There is level 4 evidence (Marini et al. 2008) that use of job placement services may help individuals with SCI find employment.

There is level 4 evidence (Inge et al. 1998) suggesting that person-centred planning tools facilitate employment.

There is level 4 evidence (King et al. 2004) that case management programs increase return to educational training, but not to work.

There is level 4 evidence (Sinnott et al. 2014) that although the supported employment (SE) program for veterans with SCI was more effective in achieving competitive employment than treatment as usual (TAU), it was not cost effective after 1 year of follow-up.

There is level 4 evidence (Dorstyn et al. 2019) that providing structured information on SCI and employment (Work and SCI) over a 4-week period may help to establish vocational interests among job-seeking persons with SCI.

There is level 5 evidence from 4 studies (Wang et al. 2002; Jang et al. 2005; Jongbloed et al. 2007; Hansen et al. 2007) that receiving vocational training increases the likelihood of employment.

There is level 5 evidence (Rowell and Connelly 2010) that an individual’s propensity to internalize positive employment outcomes in relation to his or her capabilities may contribute to returning to work.

People with SCI may benefit from vocational rehabilitation in the process of job placement and work reintegration.

There is a dearth of high quality research in vocational (re) training. Consequently, conclusions are mostly based on evidence from observational studies or case studies.

Continuous support to the employee and employer before and after vocational placement may lead to a successful return to work and job retention.