Impact of return-to-work pathway on vocational outcomes following severe acquired brain injury: A secondary analysis of the VIP1.0 trial

Abstract

Background

Most research into vocational re-integration following severe acquired brain injury (ABI) has focused on returning to pre-injury employment. While there is significant demand for new employment post-ABI, much less is known about this return-to-work (RTW) pathway.

Objective

To compare and contrast the clinical profiles and outcomes achieved through vocational rehabilitation (VR) following severe ABI according to RTW pathway.

Methods

Secondary data analysis of the Vocational Intervention Program efficacy trial (VIP1.0). Participants with severe ABI undertaking a VR program were re-classified into two RTW pathway groups: return to pre-injury employment (Fast-Track; FT) or obtaining new employment (New-Track; NT). Competitive employment status (Yes/No) and clinician ratings of disability and participation were collected pre- and post-intervention and at 3-month follow-up.

Results

47 (46%) participants undertook a FT pathway and 55 (54%) a NT pathway. Compared to FT, NT participants were less educated, more likely to be single, further post-injury, already attempted to RTW unsuccessfully post-injury, and less likely to complete the VR intervention. For both pathways, employment rates improved from pre- to post-intervention and remained stable at follow-up. While FT was associated with high RTW success (post-intervention: 89%; follow-up: 88%), a significant minority of NT participants also achieved competitive employment (post-intervention: 34%; follow-up: 45%). FT participants were also rated as demonstrating greater improvements in work-related participation from pre- to post-intervention than NT.

Conclusions

Despite its challenges, VR interventions targeting new employment can lead to beneficial RTW outcomes for many individuals unable to return to their previous employer following severe ABI.

Keywords

acquired brain injury clinical trial employment return to work traumatic brain injury vocational rehabilitation

Introduction

Acquired brain injuries (ABI) including traumatic brain injury (TBI) are catastrophic neurologic insults associated with significant impacts across many aspects of motor-sensory, cognitive and behavioural domains.^1,2 Flow-on effects from such injuries can adversely impact on everyday functioning as well as broader participation including employment.³ Most adults who sustain an ABI are employed at the time of injury, with rates ranging from 51–80%.^4,5 Return-to-work (RTW) is consistently identified by patients as an important rehabilitation goal post-ABI.^6,7,8 RTW success is quite high after mild injuries (typically greater than 80%),⁹ but much lower following moderate-to-severe injuries (32–55%),^4,10,11,12 flagging the importance of research within this more severely injured group.

Vocational rehabilitation (VR) is an effective mode of coordinating and delivering interventions to improve employment outcomes following severe ABI.¹³ Accessing competitive employment (i.e., salaried work in a mainstream setting) is the most desirable outcome, comprising either return to pre-injury employment or obtaining new employment. Historically, the majority of research examining RTW after ABI has focused on return to pre-injury employment.¹⁴ The advantages associated with this pathway include clients typically returning to work in a familiar environment for which they have preserved knowledge and skills, along with a pre-existing relationship with their employer.¹⁵ Although there is a significant demand for new employment post-ABI,^5,16–18 much less is known about this RTW pathway. This situation has arisen partly because VR trials and broader descriptive studies of outcomes from VR programs typically aggregate return to pre-injury employment and new employment into a single outcome.¹⁹

The handful of studies that have directly compared these pathways suggest there are important differences in relation to the profile of participants, VR program trajectories and final vocational outcomes. People seeking new employment (vs. return to pre-injury employment) were more likely to be younger, single, and less educated,^17,20 have more severe brain injuries,^17,18,21 higher levels of challenging behaviours and psychosocial stressors,¹⁷ and concomitant injuries.¹⁸ Regarding VR program trajectories, it took longer for participants to find new employment than to return to pre-injury employment.^17,22 In relation to outcomes, participants in new employment were less well paid than their peers who returned to their previous employment,²¹ had poorer employment stability,^17,21,22 and were more likely to be working part-time and in unskilled employment.^17,22 The findings suggest that those embarking on a new employment pathway are likely to require a greater degree of work preparation and more intensive on-job support which would need to be sustained over a longer timeframe.¹⁷

The non-randomized controlled trial of the Vocational Intervention Program (VIP1.0;²³) provided an opportunity to replicate these findings in a prospective study. The trial evaluated the efficacy of two different models of VR service delivery in facilitating return to competitive employment post-severe ABI: (i) through partnerships between community-based brain injury rehabilitation services and local private VR providers (VIP model; n = 75); (ii) directly through an established ABI specialist VR service embedded within a hospital-based brain injury rehabilitation service (hospital-based VR (H-VR) model; n = 33).¹⁵ The two service delivery models were compared against treatment-as-usual (TAU) matched controls (n = 40) as well as against one another. The primary outcome of the trial was return to competitive employment (Yes vs. No). Secondary outcomes were assessed using validated measures of disability and participation. All outcome variables were measured at three time-points (baseline, close of intervention, 3-month follow-up).

In the trial, the two VR service delivery models achieved significantly higher rates of RTW compared to the TAU controls (51% VIP, 69% H-VR vs. 17% TAU; increase from pre- to post-intervention), with these significant differences maintained at the follow-up time-point.²³ Furthermore, the trial found evidence for improved work-related participation and independent living skills (secondary outcomes) from pre- to post-intervention in the intervention arms (VIP and H-VR) compared to the matched controls (TAU). Importantly, there were no significant differences in RTW rates or secondary outcomes between the two VR service delivery models (i.e., VIP vs. H-VR).

As the two VR service delivery models (VIP and H-VR) provided programs targeting either return to pre-injury employment or seeking new employment, the trial provided the opportunity to aggregate the participants from the intervention arms according to these two RTW pathways. Within the VIP trial, the return to pre-injury employment pathway was labelled Fast Track (FT), while the new employment pathway was labelled New Track (NT). To help expand understanding of seeking new employment after severe ABI, a secondary analysis was conducted to test three hypotheses generated from the existing literature, proposing: (i) participants seeking new employment (NT) would have greater adverse psychosocial and injury characteristics compared to participants returning to pre-injury employment (FT); (ii) the RTW trajectory would be significantly longer for NT than for FT; and (iii) the primary and secondary outcomes would be inferior for NT compared to FT, as indicated by significant main effects for group and time, without significant interaction effects.

Methods

Design

VIP1.0 was a prospective, multi-center, three-armed non-randomized controlled trial evaluating the efficacy of two VR service delivery models (VIP, H-VR) compared to TAU in enhancing return to competitive employment following severe traumatic and non-traumatic ABI. It was conducted from July 2015 to September 2018 and involved all 12 adult sites of the New South Wales (NSW) Brain Injury Rehabilitation Program (BIRP). The BIRP spans both metropolitan and rural areas of the state with all sites sharing standardized network referral criteria, resulting in a high degree of homogeneity in client profiles across the program.²⁴ Individually-tailored VR programs were delivered through the VIP model at six of the BIRP sites and the H-VR model at one site, with the remaining five sites serving as TAU controls. The trial was registered with ANZCTR (ACTRN12622000769785). More detail about the design and setting of the trial can be found in Simpson et al.²³

The current study involves a re-categorization and re-analysis of participants in the two VR service delivery model arms. In other words, within each of these two arms, a subset of participants undertook a return to pre-injury employment (FT) pathway while the rest undertook a new employment (NT) pathway. Thus, for the current secondary analysis, VIP and H-VR participants were pooled together and then divided into two new groups according to their chosen RTW pathway (i.e., FT vs. NT).

Participants from the TAU control arm were not included in the current analysis. In the original trial, the TAU individuals were matched to participants in the VIP arm on six variables (sex, age, time post-injury, injury severity, pre-injury vocational status, and pre-injury education) but not by pathway. Therefore, it was not possible to partial the TAU group into controls for the planned analyses.

This study was conducted in line with the principles of the Declaration of Helsinki. Ethical approval for the VIP1.0 trial was granted by the Northern Sydney Local Health District Human Research Ethics Committee (RESP/15/161 & RESP/15/188). All participants in the VIP arm provided written informed consent prior to enrolment. A second ethical approval was obtained to waive consent from H-VR participants before collecting personal information (in line with the conditions specified in the Australian National Statement on Ethical Conduct in Human Research 2023), as there was no direct patient contact beyond routine clinical care.

Participants

Participants were 102 clients of the 7 participating BIRP sites delivering VR programs. All had sustained a severe ABI and were undergoing a multidisciplinary brain injury rehabilitation program. Most of the study eligibility criteria were identical across the two RTW pathways, namely: having an active rehabilitation goal of RTW; primary diagnosis of severe ABI (i.e., severe to extremely severe TBI (PTA duration >24 h) or severe non-traumatic brain injury); living in stable accommodation and independent in the community; sufficient cognitive capacity to participate in a RTW program; and no current substance use or mental health disorder (according to the primary BIRP clinician's judgement and informed by relevant items of the Health of the Nation Outcome Scale – Acquired Brain Injury²⁵).

Additional eligibility criteria specific to FT were: employed at time of injury and willing to RTW with pre-injury employer; and employer able to provide a graded RTW program. Additional eligibility criteria specific to NT were: no identified employer; requiring a work trial opportunity as an initial step towards new employment; and agree to undertake an unpaid work trial of at least eight hours per week for up to 12 weeks.

Intervention

The general VR program (which included separate FT and NT pathways) was originally developed at the Head2Work VR service at Liverpool Hospital Brain Injury Unit.¹⁵ The VIP1.0 trial evaluated the delivery of the program through a novel context, namely a service partnership model, linking BIRP community brain injury rehabilitation teams with local private VR providers. These providers worked with a range of disability groups, although some did not have specialist brain injury expertise. They were appointed by the trial sponsors through a tender process and funded on a per-person milestone fee structure.

Activities provided in FT programs were identical across VIP and H-VR arms and included functional and workplace assessments, employer education, development and implementation of a graded RTW plan (considering strategies, graduated hours, equipment prescription, and training, etc.), on-site reviews, and ongoing support to the participant and employer throughout the intervention period.

Services provided in NT programs at the six BIRP sites implementing the VIP model focused on sourcing and managing a work trial as an initial step towards returning to the workforce. The work trial was an unpaid placement in open employment for up to 12 weeks, to trial a new occupation, upgrade work readiness, and position oneself for ongoing paid work, whilst circumventing the formal recruitment process. These programs commenced with vocational assessments to establish goals, progressed to identifying suitable employers for work trials, workplace assessment and negotiation of working conditions, on-site training and support, consideration of equipment and/or training needs, employer education and support, monitoring and upgrading of the work trial, and planning for onward VR. For NT programs implemented through the H-VR model, Head2Work staff offered job-seeking services to secure paid work, in addition to work trials.

Measures

Vocational outcomes

The primary study outcome was whether the participant had achieved competitive employment (defined as full-time or part-time work taking place within the integrated, mainstream labour market for which the participant was salaried), operationalized as a binary (Yes vs. No) outcome. Secondary vocational outcomes included work activity status (i.e., whether working full-time or part-time), number of hours worked, and the nature of employment – open employment (full duties or partial duties – i.e., not completing all of the components in the standard job description and/or working at reduced hours) or other non-competitive outcomes including supported employment (i.e., segregated employment where the employee works alongside other people with disability and may receive some form of remuneration), studying, volunteering, or avocational. Pre-injury vocational status was also recorded.

Clinician-rated measures

Dysexecutive behaviour, disability status, and participation levels were rated by each participant's primary BIRP rehabilitation clinician using measures with well-established psychometric properties for the ABI population:^26,27

The Dysexecutive Questionnaire (DEX)²⁸ is a 20-item measure of behavioural changes and everyday problems commonly encountered by individuals with frontal lobe injuries measured on a 5-point scale, resulting in a total score from 0 to 80. Psychometric studies of the DEX in ABI populations have reported excellent internal consistency (α>0.90), good inter-rater reliability (r = 0.79), and adequate construct validity.^29–31 The informant version was completed by the clinician in this study.

The Disability Rating Scale (DRS)³² is a clinician-rated 8-item measure of functional ability. Items are rated on a scale from 0–3 or 0–5, producing a total score ranging from 0–29. In the VIP1.0 trial, item 8 (the “employability” item) was excluded from calculation of the total score, making 26 the maximum possible score from the modified total. The DRS has demonstrated robust psychometric properties in relation to TBI, including high internal consistency (α=0.83–0.84), inter-rater reliability (ICC > 0.90), and test-retest reliability (r = 0.95), as well as adequate construct and predictive validity.³³

The Sydney Psychosocial Reintegration Scale-2 (SPRS-2)³⁴ is a 12-item measure designed to evaluate community participation following ABI. Items are scored from 0 to 4 with a total score ranging from 0 to 48. SPRS-2 subscales are: Occupational Activity for Work and Leisure, Interpersonal Relationships, and Independent Living Skills, each containing four items and scored from 0 to 16. The clinician version of Form B was used in this study. The SPRS-2 Form B has demonstrated excellent internal consistency (α=0.93), inter-rater reliability (ICC = 0.84), and test-retest reliability (r = 0.90), along with adequate construct validity.³⁵

Screening measures

The Health of the Nation Outcome Scale – Acquired Brain Injury (HoNOS-ABI)²⁵ is a 12-item clinician-rated measure of common neuropsychiatric manifestations of ABI. Each item assesses a unique function or outcome using a 5-point scale (0 = no problem, 4 = severe problem). Reported psychometrics include adequate inter-rater reliability (ICC = .58-.97)³⁶ and concurrent validity (including r = .62 with post-injury employment status and r = .75 with psychosocial outcome (Portland Adaptability Index)).²⁵ In the current study only a subset of HoNOS-ABI items were rated, predominantly to screen for severe comorbid substance use and other mental health issues. These were items 1 (disturbance of social behaviour), 3 (problem drinking or drug use), 5 (physical illness or disability problems), 6 (psychosis/confabulation), 7 (depression), and 8 (other mental and behavioural problems).

Procedure

After providing informed consent, participants in the VIP arm were referred to the assigned private VR provider by their local BIRP site. Participants in the H-VR arm (based at Liverpool Hospital Brain Injury Rehabilitation Unit) were referred directly to the Head2Work VR service following the usual procedures for clinician referral of clients with active RTW goals.

For FT participants, VR providers contacted their employer to confirm their support with RTW planning, which took place over the next 4 weeks. Once the participant had commenced working, their RTW progress was monitored, with gradual upgrading of the RTW program for up to 6 months. At the end of this phase, the VR provider oversaw a progress review with the participant and their employer, and onward referral for further support was made if needed. The participant's file was then closed by the VR provider, marking the end of the VR intervention.

For NT participants, VR providers worked with the participant over the first 4 weeks to identify suitable options for a work trial via a vocational assessment, with input from the assigned BIRP case manager, the participant's family, and other stakeholders. The VR provider then sourced a host employer and conducted a workplace assessment to confirm suitability. An unpaid work trial of up to 12 weeks was then negotiated with a host employer. At the conclusion of the work trial, the VR provider oversaw a progress review with the participant and host employer. Although completion of the 12 weeks constituted the end of the VR intervention, in some cases the VR provider continued to work with the participant and host employer, or alternatively, made arrangements for onward referrals as needed.

Study outcomes were collected at three time points: pre-intervention, post-intervention, and 3-month follow-up. The first time point (pre-intervention) occurred at the commencement of VR activities. The second time point (post-intervention) occurred at file closure, and the third time point (follow-up) three months after file closure. Clinician ratings were not able to be obtained at follow-up for H-VR participants due to limitations in available staff resources. Demographic and injury-related variables were also collected pre-intervention for all participants.

Statistical analysis

Data were analyzed according to intention-to-treat (ITT) principles, with the eligible study sample consisting of all VIP1.0 trial participants who commenced a FT or NT pathway, regardless of whether or not they eventually completed the VR program. Descriptive statistics were generated to characterize the two RTW pathways, comparing demographic and injury-related variables with chi-square or Mann-Whitney U tests. Between-group comparisons were primarily conducted using JMP v14 (SAS^® Institute Inc.; Cary, NC). p-values < .05 were considered statistically significant.

A series of multilevel models with piecewise slopes were conducted using PROC GLIMMIX and MIXED in SAS^® Studio v3.8 (SAS^® Institute Inc.; Cary, NC) to analyse the impact of the two RTW pathways on each outcome over time. Repeated measurements (i.e., three time points) were nested at level 1 within each participant at level 2, such that variability between participants could be modelled using random effects. Initially, unconditional multilevel models (with fixed and random intercepts only) were defined. Next, two piecewise variables representing time from: 1) pre-intervention to post-intervention (“pre-post time slope”) and 2) post-intervention to follow-up (“post-follow-up time slope”); were added as level 1 variables. Random slopes corresponding to the piecewise time variables were also considered and included if they improved the overall model fit. Over successive steps, a level 2 group variable (representing RTW pathway) was added, followed by cross-level interaction variables between group and each of the piecewise time slopes. Pre-injury years of education and time since injury were also considered as covariates at level 2 as they were found to differ between the two RTW pathways. Given the longitudinal nature of the data, we also considered whether using an alternate R-side covariance structure (e.g., autoregressive-1) improved the overall model fit.

As the primary study outcome, competitive employment, was a binary variable, a generalized linear mixed model was conducted to predict the probability of being employed, using a logit link function and binomial conditional outcome distribution, with an inverse link function used to provide model-predicted proportions (between 0 and 1) and their standard errors.

The maximum likelihood marginal expansions pseudo-likelihood (MMPL) method was used to obtain final parameter estimates, as it provided a closer approximation to the (raw) obtained fixed parameters than comparable models estimated using maximum likelihood (ML) with either adaptive Gauss-Hermit quadrature or Laplace approximation to quasi-likelihood. Empirical sandwich estimates were used to improve the robustness of the standard errors of the fixed effects in the final model.³⁷

The study secondary outcomes, relating to clinician ratings of disability and participation, were continuous variables. Each score (and its standard error) was estimated by a linear mixed model using restricted maximum likelihood (REML) estimation, following the same model-building process outlined above.

Results

Of the 102 participants meeting eligibility criteria for the current study, 47 (46%) enrolled in a FT pathway and 55 (54%) in a NT pathway. Figure 1 presents a breakdown of the sample by pathway, with available vocational and clinician-rated outcomes at each time point.

Figure 1.

Participant flow diagram.

Hypothesis 1: clinical profile

Sample demographic and injury characteristics are provided in Table 1. Compared to FT, NT participants were less educated, more likely to be single, a longer time post-injury, and comprised a higher proportion of non-traumatic ABI cases (although both groups were majority TBI). They did not differ with regards to other demographic or injury variables examined. By definition, all FT participants were employed at the time of injury, whereas only 35/55 (65%) NT participants were. One FT participant (2%) had unsuccessfully attempted to return to their pre-injury employer since their injury, while 17/55 NT participants (31%) had done so (χ²= 14.45; p = .0001).

Table 1.

Sample demographic and injury characteristics.

	FT (n = 47)	NT (n = 55)	Test
Sex (n, %)
Male	38 (81)	46 (84)	χ²= 0.14
Female	9 (19)	9 (16)	p = .71
Age at Referral (Years)
Med (IQR)	42.0 (29.0–51.0)	35.0 (25.0–48.0)	U = 1045.00
			p = .10
Marital Status		(n = 54)
Married/De Facto	29 (62)	20 (37)	χ²= 6.75
Single	17 (36)	30 (56)	p = .04
Divorced/Separated	1 (2)	4 (7)
Years of Pre-Injury Education	(n = 41)	(n = 44)
Med (IQR)	14.0 (12.0–16.0)	12.0 (10.0–14.0)	U = 572.00
			p = .003
Highest Education Achieved (n, %)		(n = 46)
Year 10 or less	3 (6)	15 (33)	χ²= 11.15
Year 12	10 (21)	8 (17)	p = .01
TAFE	18 (38)	15 (33)
University	16 (34)	8 (17)
Country of Birth (n, %)
Australia	38 (81)	49 (89)	χ²= 1.37
Other	9 (19) ^a	6 (11) ^b	p = .24
Time Post-Injury (Months)
Med (IQR)	4.0 (3.0–6.0)		U = 247.50
			p < .0001
Type of Injury (n, %)
TBI	40 (85)	37 (67)	χ²= 4.36
ABI	7 (15)	18 (33)	p = .04
PTA Duration (Days) (TBI only)	(n = 38)	(n = 34)
Med (IQR)	21.5 (9.8–39.0)	29.0 (11.8–62.3)	U = 535.00
			p = .21
TBI Severity (PTA) (n, %)	(n = 38)	(n = 34)
Severe (1–7 days)	5 (13)	6 (18)	χ²= 2.30
Very Severe (8–28 days)	19 (50)	11 (32)	p = .32
Extremely Severe (>28 days)	14 (37)	17 (50)

Note. ABI = Acquired brain injury; FT = Fast-Track; NT = New Track; PTA = Post-traumatic amnesia; TAFE = Technical and Further Education (NSW); TBI = Traumatic brain injury.

Percentages may not add up to 100 due to rounding error.

New Zealand (n = 2), Austria (n = 1), East Africa (n = 1), Fiji (n = 1), Philippines (n = 1), South Africa (n = 1), Turkey (n = 1), and United Kingdom (n = 1).

Hong Kong (n = 1), Iraq (n = 1), Lebanon (n = 1), Mauritius (n = 1), New Zealand (n = 1), and United Kingdom (n = 1).

Hypothesis 2: trajectories

Significantly more participants referred in the VIP arm undertook the NT pathway (61%) compared to the H-VR arm (33%) (χ²= 6.26; p = .01). There was a marginally significant trend suggesting the duration of the VR program was longer in NT (Med = 9 months, IQR = 6–14.75; n = 32) than FT (Med = 7 months, IQR = 6–8.75; n = 44) (p = .07). This difference reached statistical significance following removal of 2 outliers from FT and 1 from NT (all >25 months; > 3SD) (NT: Med = 9 months, IQR = 6–14 vs. FT: Med = 7 months, IQR = 6–8) (p = .04). NT participants (53%) were less likely than FT (81%) to complete the VR intervention (χ²= 8.89; p = .003).

Hypothesis 3: vocational and clinical outcomes

Competitive employment outcome

Table S1 in Supplementary Materials provides descriptive statistics for (raw) vocational outcome data. Pre-intervention, 2% of FT and 13% of NT participants were competitively employed. Post-intervention, employment rates increased substantially for FT (89%) while there was a smaller increase for NT (34%). At 3-month follow-up, employment remained stable for FT (88%) but continued to increase for NT (45%).

The final generalized linear mixed model parameters and model-estimated group means of the proportions achieving competitive employment at each time point are presented in Table 2 and Figure 2 respectively. Overall tests of the fixed effects of pre-post time slope (p < .0001) and the group*pre-post time slope interaction (p = .0004) were significant, indicating that while RTW rates improved for both groups from pre- to post-intervention, the rate of improvement was significantly higher in FT relative to NT. The between-group difference remained stable from post-intervention to follow-up (p = .52).

Figure 2.

Estimated group means and 95%CIs from the competitive employment (primary) outcome generalized linear mixed model.

Table 2.

Final multilevel model parameters.

	Employment Outcome (GLMM)	DEX Total Score (LMM)	DRS Total Score (LMM)	SPRS-2 Total Score (LMM)	SPRS-2 Work/Leisure (LMM)	SPRS-2 Relationships (LMM)	SPRS-2 Living Skills (LMM)
Fixed Effects (B, SE)
Intercept	−3.83 (1.03)***	31.58 (7.83)***	1.86 (0.59)**	37.09 (1.04)****	9.66 (0.42)****	13.60 (0.45)****	10.41 (1.22)****
Piecewise Time Slope Main Effects
Pre-Post	5.88 (1.12)****	−3.04 (1.36)*	−0.45 (0.15)**	5.22 (0.85)****	3.92 (0.43)****	0.05 (0.33)	1.27 (0.26)****
Post-FU	−0.11 (0.67)	1.32 (1.79)	−0.06 (0.20)	−0.19 (0.74)	−0.33 (0.38)	−0.21 (0.46)	0.27 (0.20)
Group Main Effects
FT	0	0	0	0	0	0	0
NT	1.90 (1.12)	5.93 (2.55)*	0.25 (0.21)	−4.37 (1.43)**	−1.06 (0.58)	−2.46 (0.62)****	−0.22 (0.46)
Group*Time Slope Interaction Effects
FT*Pre-Post	0	0	0	0	0	0	0
NT*Pre-Post	−4.55 (1.25)***	1.01 (2.19)	0.56 (0.24)*	−2.63 (1.30)*	−2.36 (0.64)***	0.39 (0.51)	−0.74 (0.41)
FT*Post-FU	0	0	0	0	0	0	0
NT*Post-FU	0.54 (0.85)	−3.13 (2.72)	−0.25 (0.31)	0.35 (1.08)	0.37 (0.56)	0.55 (0.66)	−0.43 (0.30)
Pre-Injury Education^{a, b}	.	−1.50 (0.56)**	−0.09 (0.04)*	.	.	.	0.25 (0.09)**
Error Variance
Level-1	.^c	39.20 (5.74)****	0.51 (0.07)****	6.11 (1.38)****	1.71 (0.39)****	5.21 (3.39)	0.46 (0.10)****
AR(1)	.	.	.	.	.	0.55 (0.29)	.
Level-2 Intercept	1.96 (0.76)	96.96 (19.19)****	0.40 (0.10)****	44.24 (7.36)****	6.53 (1.25)****	4.13 (3.58)	4.10 (0.70)****
Time Slope (Pre-Post)^d	.	.		19.38 (5.69)***	4.67 (1.47)***	.	2.11 (0.51)****

* p < .05 **p < .01 ***p < .001 ****p < .0001

Note. AR(1)=Autoregressive-1 R-side covariance structure; DEX = Dysexecutive Questionnaire; DRS = Disability Rating Scale; FT = Fast Track; FU = Follow-up; GLMM = Generalized linear mixed model; LMM = Linear mixed model; NT = New Track; SPRS-2 = Sydney Psychosocial Reintegration Scale-2.

^aFor nine participants, highest education achieved was known but the exact years of education was missing. Median imputation (based on the corresponding educational attainment category) was used to estimate years of education in these participants to maximise the available sample.

^bTime since injury was also considered as a fixed covariate at level-2. However, it did not reach statistical significance nor lead to a substantial improvement in overall model fit. Consequently, it was not included in any of the final models.

^cFor the Bernoulli (binomial) distribution, there is no residual variance to estimate – the variance is redundant to the mean.

^dIncluding random piecewise time slopes representing Post-FU did not improve model fit in any analyses. Consequently, they were not retained in any of the final models.

Other secondary vocational outcomes

There were substantial differences between the RTW pathways with respect to other employment outcomes (see Table S1). Specifically, FT participants overall secured almost twice as many hours of work as NT (Med = 35 vs. 17.5 h per week). Similar proportions of FT participants returned to full-time employment (43%) and partial duties (45%) (the remaining 11% were avocational). In contrast, the nature of employment attained by NT participants was far more varied. Just under half (44%) were avocational. Those working in open employment were more likely to be on partial duties (22%) than full duties (9%), while an additional 3% were in supported employment, 6% were studying, and 16% were volunteering.

Clinician-rated outcomes

Descriptive statistics for the clinician-rated outcomes at each time point are presented in Table S2 in Supplementary Materials. The final parameters and model-estimated group mean scores of each linear mixed model conducted in the following analyses are provided in Table 2 and Figure 3 respectively.

Figure 3.

Estimated group means and 95%CIs for clinician-rated outcome linear mixed models.

Disability measures

DEX: Overall tests of the fixed effects of pre-post time slope (p = .02), group (p = .02), and pre-injury education (p = .009) were significant. At pre-intervention, DEX scores were generally low in both groups, albeit a mean of 7.81 points lower (indicating less executive dysfunction) in FT (p = .002). Scores decreased to a similar extent from pre- to post-intervention in both groups and remained relatively constant from post-intervention to follow-up (p = .35).

DRS: Significant overall tests were found for the fixed effects of pre-post time slope (p = .004), group*pre-post time slope interaction (p = .02) and pre-injury education (p = .04). DRS ratings were very low (near-zero) in both groups throughout the study, with FT participants rated as having marginally lower levels of disability than NT participants at each time point.

Participation measures

SPRS-2: For the SPRS-2 total score, overall tests of the fixed effects of pre-post time slope (p < .0001), group (p = .003), and the group*pre-post time slope interaction (p = .046) were significant. Pre-intervention ratings were generally very high for both groups, albeit a mean of 4.37 points higher in FT (p = .003). While scores increased from pre- to post-intervention for both groups, the rate of improvement was greater for FT than NT (mean increase = 5.23 vs. 2.60). Scores were virtually unchanged from post-intervention to follow-up (p = .75).

For the SPRS-2 Occupational Activity for Work and Leisure subscale, overall tests of the fixed effects of pre-post time slope (p < .0001) and the group*pre-post time slope interaction (p = .0004) were significant. Pre-intervention scores did not differ between the groups. While both improved from pre- to post-intervention, the rate of improvement was greater for FT (mean increase = 3.92 vs. 1.56). Scores remained stable from post-intervention to follow-up (p = .52).

For the SPRS-2 Interpersonal Relationships subscale, only the overall test of the fixed effect of group was significant (p = .0001). Pre-intervention scores were a mean of 2.46 points higher in FT participants compared to NT (p = .0001) and this difference persisted at both subsequent time points.

For the SPRS-2 Independent Living Skills subscale, overall tests of the fixed effects of pre-post time slope (p < .0001) and pre-injury education (p = .005) were significant. Pre-intervention scores did not differ. Both groups showed a similar rate of improvement from pre- to post-intervention, and scores remained relatively stable from post-intervention to follow-up.

Discussion

The current study found evidence for the distinct nature of pathways targeting a return to pre-injury employment vs. seeking new employment following severe ABI. Importantly, the demand for new employment post-ABI was substantial, consistent with previous reports,^5,16–18 highlighting the importance of programs to support clients with these needs. For both pathways, the interventions were principally based in the workplace, lending further support to past findings that workplace intervention and more integration are associated with better RTW outcomes,³⁸ with outcomes from both pathways comparing favourably to previous studies.¹¹ However, participants in the new employment pathway presented greater challenges to achieving competitive employment due to their demographic and injury profile, their VR programs were by necessity of longer duration, and there was a lower success rate compared to participants returning to pre-injury employment. This increased understanding of the differences between the two pathways can be useful in identifying potential RTW candidates, helping to estimate the expected duration of VR interventions, and inform future service planning.

Comparing the clinical profiles of the two pathways, the first study hypothesis was only partially supported. In terms of demographic variables, participants seeking new employment did have significantly lower levels of education and were less likely to be in a married/defacto relationship, consistent with previous reports.^17,20,22,38 In the Australian context, having a tertiary qualification is associated with better job security.³⁹ This suggests that NT participants (who were more likely to have secondary school education) may have been faced with additional barriers when initially attempting to return to their pre-injury work role that ultimately proved unsuccessful. There was no difference between the pathways for other demographic variables such as sex and age, consistent with earlier studies.^17,22 Similarly, in relation to injury-related variables, no significant differences in injury severity between the two pathways emerged (NT: Med = 21.5 days vs. FT: Med = 29 days; PTA duration). Comparable injury severity was also observed in a later follow-up VIP effectiveness trial (VIP2.0) (NT: Med = 26 days vs. FT: Med = 14.5 days;⁴⁰). Where differences in injury severity according to RTW pathway have been reported previously, the analysis has compared mild and moderate-severe TBI,¹⁸ rather than focus on the broader spectrum of severe injuries as for the current study.

In contrast, time to treatment (i.e., time post-injury) was significantly longer for the new employment pathway (NT: Med = 26 months vs. FT: 4 months), suggesting a bimodal distribution across the two pathways. The time to treatment for the pre-injury employment pathway was similar to previous studies in which aggregated values of 2 months⁴¹ and 3.5 months⁴² were respectively reported. However, in a study of a state insurance-funded cohort employing pre-set time categories, later times for RTW were also documented (early RTW: < 6 months vs. late RTW: 7–34 months),⁴³ with the ‘late’ group in keeping with the profile of the new employment pathway in the current study. While it might be presumed that the majority of participants returning to pre-injury employment were situated within Spitz et al.'s⁴³ ‘early’ group, this was hard to determine, as employers received an insurer-funded wage subsidy for participants returning to their pre-injury employment, and this may have attenuated the normal period for which pre-injury jobs are held open.

The second study hypothesis regarding trajectory of RTW duration was supported. The new employment pathway took significantly longer than pre-injury employment, consistent with previous research.^17,22,40 It is likely that key RTW processes such as job-seeking and the relevant training necessary for a new position⁴⁴ require more intensive programs of longer duration. Participants seeking new employment were also more likely to drop out prior to completing their VR programs. As the new employment pathway intervention was, by definition, a trial in a new workplace, these participants did not have any of the incumbency advantages enjoyed by those aiming to return to their pre-injury employment (i.e., familiarity with the role/tasks, pre-existing relationship with the employer). There also greater potential for a mismatch between the client and the role/employer, leading the client to discontinue prematurely due to dissatisfaction and feeling they were a “poor fit” for the role.

The third hypothesis was supported in relation to the primary outcome (competitive employment), consistent with the previous report by Fabiano & Crewe.²¹ The rates of successful employment outcomes for both RTW pathways (FT: 89%; NT: 32% at file closure, 45% at follow-up) compare favourably with the range of VR outcomes reported within the 1 year and up to 5 year range by Gormley et al.¹¹ The delayed effect with the new employment pathway represents cases in which the work trial itself did not turn into an employment outcome, and further job seeking was required after the trial to find employment. The finding that participants in the new employment pathway were working significantly fewer hours than the pre-injury employment pathway at file closure and follow-up is in line with previous research^17,21,22 and suggests that the work they managed to obtain at the end of the trial tended to be less secure.

Some secondary outcomes also supported the third hypothesis. The higher levels of baseline dysexecutive problems among new employment pathway participants is consistent with previous retrospective research, in which the presence of challenging behaviours was more common among persons seeking new employment vs. those returning to pre-injury employment.¹⁷ The fact that dysexecutive syndrome was assessed by a validated measure in the current prospective trial extends the earlier retrospective finding that employed a single clinical indicator (presence vs. absence of challenging behaviours). Despite the initial differences between the two pathways, participants seeking new employment and those returning to pre-injury employment both achieved reductions in DEX scores by the end of the intervention, and these improvements were maintained at follow-up. Given that participants were concurrently receiving other rehabilitation interventions as part of their BIRP community programs, it cannot be concluded that such improvements were specific to the vocational intervention. The possibility that it reflects natural recovery from injury is supported by the fact that similar reductions in dysexecutive problems were observed across all three original arms of the VIP1.0 trial, including in the TAU arm who did not undertake a VR program.²³ For participation (as measured by the SPRS-2), the greater improvement over the intervention period in the pre-injury employment pathway was driven largely by the occupational activity (work/leisure) subscale, highlighting the important role that vocational attainment plays in supporting overall participation after ABI.⁴⁵

Clinical implications

Two important clinical implications arise from this study. First, it contributes to the evidence base supporting the value of work trials as an intervention for people seeking new employment.^46–48 Benefits of work trials include the real-world assessment of work capacity, building up work skills, and providing additional work experience to enhance a résumé, all without having to go through a formal (competitive) recruitment process.¹⁵ Second, up to half of those undertaking a VR program had a PTA duration longer than 30 days, flagging that individuals with extremely severe injuries can still be considered for possible RTW goals (see also Spitz et al.⁴³).

Methodological limitations

In considering the findings, the following limitations need to be kept in mind. In conducting the secondary data analysis, some statistical comparisons were likely to have been underpowered. Additionally, clinician-rated secondary data was not able to be collected at 3-month follow-up for participants in the H-VR arm due to staff resourcing limitations. This may have impacted the consistency of the effects of the VR programs on these outcomes from post-intervention to follow-up. Furthermore, there was some inconsistency in the VR activities available to new employment participants depending on the original VIP1.0 trial arm. For those in the VIP arm, the intervention was restricted to work trials, while some in the H-VR arm received additional activities such as job-seeking support to secure paid work. Finally, collection of data related to cultural, language, and ethnic diversity was limited to country of birth, which prevented in-depth analysis of any potential impacts of these factors on RTW outcomes within the Australian context.

Conclusion

Despite its challenges, VR interventions targeting new employment can lead to improved RTW outcomes for many individuals with severe ABI for whom a return to their pre-injury employer is not feasible. Future research could undertake a health economic analysis to examine the relative costs/benefits associated with the two RTW pathways, with particular attention given to the possible value achieved through the new employment pathway. Next, outcomes for the new employment pathway could be further tested with VR programs incorporating a broader range of job-seeking and other activities, in addition to work trials. Finally, in seeking to make decisions about the viability of clients to undertake a RTW program, the Work-ability Support Scale (WSS) has now been validated within the current severe TBI sample.⁴⁹ In that study, WSS scores demonstrated predictive validity, with a moderate degree of association with successful return to competitive employment. Further testing of the WSS could examine whether the strength of predictive validity is consistent across both pathways.

Supplemental Material

sj-docx-1-wor-10.1177_10519815241293893 - Supplemental material for Impact of return-to-work pathway on vocational outcomes following severe acquired brain injury: A secondary analysis of the VIP1.0 trial

Supplemental material, sj-docx-1-wor-10.1177_10519815241293893 for Impact of return-to-work pathway on vocational outcomes following severe acquired brain injury: A secondary analysis of the VIP1.0 trial by Thomas M Gates, Philippa McRae, Maysaa Daher and Grahame K Simpson in WORK

Footnotes

Acknowledgements

The authors wish to thank the NSW BIRP clinicians and clients across the seven sites who contributed data for this secondary analysis, and to the three appointed VR providers and BIRP teams involved in establishing the VIP service model. They would also like to thank Dr Joseph Descallar from the Ingham Institute of Applied Medical Research for his advice on multilevel modelling.

Supplemental material

Supplemental material for this article is available online.

Ethical approval

The VIP1.0 trial was approved by the Northern Sydney Local Health District Human Research Ethics Committee (RESP/15/161 & RESP/15/188).

Informed consent

In the VIP1.0 trial, all participants in the VIP arm provided written informed consent prior to enrolment. A second ethical approval was obtained to waive consent from H-VR participants before collecting personal information, as there was no direct patient contact beyond routine clinical care.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The VIP1.0 trial was jointly sponsored by the NSW icare Lifetime Care and Support Scheme and the State Insurance Regulatory Authority.

Declaration of conflicting interests

GS receives partial salary support from the NSW icare Lifetime Care and Support Scheme. The other authors have no relevant financial or non-financial interests to disclose.

Supplemental material

Supplemental material for this article is available online.

References

McAllister

. Neurobiological consequences of traumatic brain injury. Dialogues Clin Neurosci 2011; 13: 287–300.

McDonald

Black

Copland

, et al. Cognition in stroke rehabilitation and recovery research: consensus-based core recommendations from the second stroke recovery and rehabilitation roundtable. Int J Stroke 2019; 14: 774–782.

Donker-Cools

BHPM

Daams

Wind

, et al. Effective return-to-work interventions after acquired brain injury: a systematic review. Brain Inj 2016; 30: 113–131.

de Koning

Spikman

Coers

, et al. Pathways of care the first year after moderate and severe traumatic brain injury—discharge destinations and outpatient follow-up. Brain Inj 2015; 29: 423–429.

Grauwmeijer

Heijenbrok-Kal

Haitsma

, et al. A prospective study on employment outcome 3 years after moderate to severe traumatic brain injury. Arch Phys Med Rehabil 2012; 93: 993–999.

Wagner

Sweet

Butt

, et al. Measuring patient self-reported cognitive function: development of the functional assessment of cancer therapy-cognitive function instrument. J Support Oncol 2009; 7: W32–W39.

McRae

Hallab

Simpson

. Navigating employment pathways and supports following brain injury in Australia: client perspectives. Aust J Rehabil Couns 2016; 22: 76–92.

Matérne

Lundqvist

L-O

Strandberg

. Opportunities and barriers for successful return to work after acquired brain injury: a patient perspective. Work 2017; 56: 125–134.

Bloom

Thomas

Ahrensberg

, et al. A systematic review and meta-analysis of return to work after mild traumatic brain injury. Brain Inj 2018; 32: 1623–1636.

10.

Sigurdardottir

Andelic

Wehling

, et al. Return to work after severe traumatic brain injury: a national study with a one-year follow-up of neurocognitive and behavioural outcomes. Neuropsychol Rehabil 2020; 30: 281–297.

11.

Gormley

Devanaboyina

Andelic

, et al. Long-term employment outcomes following moderate to severe traumatic brain injury: a systematic review and meta-analysis. Brain Inj 2019; 33: 1567–1580.

12.

Watanabe

. Vocational rehabilitation for clients with cognitive and behavioral disorders associated with traumatic brain injury. Work 2013; 45: 273–277.

13.

Kendall

Muenchberger

Gee

. Vocational rehabilitation following traumatic brain injury: a quantitative synthesis of outcome studies. J Vocat Rehabil 2006; 25: 149–160.

14.

Van Deynse

Ilunga Kazadi

Kimpe

, et al. Predictors of return to work after moderate-to-severe traumatic brain injury: a systematic review of current literature and recommendations for future research. Disabil Rehabil 2022; 44: 5750–5757.

15.

Tate

Simpson

McRae

, et al. Traumatic brain injury. In: Escorpizo

Brage

Homa

(eds) Handbook of vocational rehabilitation and disability evaluation: application and implementation of the ICF. New York (NY): Springer, 2015, pp. 263–294.

16.

Fleming

Tooth

Hassell

, et al. Prediction of community integration and vocational outcome 2–5 years after traumatic brain injury rehabilitation in Australia. Brain Inj 1999; 13: 417–431.

17.

Simpson

McRae

Hallab

, et al. Participation in competitive employment after severe traumatic brain injury: new employment versus return to previous (pre-injury) employment. Neuropsychol Rehabil 2020; 30: 995–1012.

18.

Thor

Mazlan

Waran

. Employment status after traumatic brain injury and the effect of concomitant injuries on return to work. Brain Inj 2021; 35: 949–956.

19.

van Velzen

van Bennekom

Edelaar

, et al. How many people return to work after acquired brain injury?: a systematic review. Brain Inj 2009; 23: 473–488.

20.

Cuthbert

Pretz

Bushnik

, et al. Ten-Year employment patterns of working age individuals after moderate to severe traumatic brain injury: a national institute on disability and rehabilitation research traumatic brain injury model systems study. Arch Phys Med Rehabil 2015; 96: 2128–2136.

21.

Fabiano

Crewe

. Variables associated with employment following severe traumatic brain injury. Rehabil Psychol 1995; 40: 223–231.

22.

Wang

Fort

Beaudoin-Gobert

, et al. Indicators of long-term return to work after severe traumatic brain injury: a cohort study. Ann Phys Rehabil Med 2019; 62: 28–34.

23.

Simpson

McRae

Gates

, et al. Is the vocational intervention program effective in enhancing return to work after severe acquired brain injury? A controlled trial. Ann Phys Rehabil Med 2023; 66: 101787.

24.

Simpson

Daher

Hodgkinson

, et al. Comparing the injury profile, service use, outcomes, and comorbidities of people with severe TBI across urban, regional, and remote populations in New South Wales: a multicentre study. J Head Trauma Rehabil 2016; 31: E26–E38.

25.

Coetzer

Du Toit

. HoNOS–ABI; a clinically useful outcome measure? Psych Bull 2001; 25: 421–422.

26.

Honan

McDonald

Tate

, et al. Outcome instruments in moderate-to-severe adult traumatic brain injury: recommendations for use in psychosocial research. Neuropsychol Rehabil 2019; 29: 896–916.

27.

McCauley

Levin

Vanier

, et al. The neurobehavioural rating scale-revised: sensitivity and validity in closed head injury assessment. J Neurol Neurosurg Psych 2001; 71: 643–651.

28.

Burgess

Alderman

Wilson

, et al. The Dysexecutive Questionnaire. In: Wilson

Alderman

Burgess

(eds) Behavioural assessment of the dysexecutive syndrome. Bury St. Edmunds, U.K.: Thames Valley Test Company, 1996.

29.

Burgess

Alderman

Evans

, et al. The ecological validity of tests of executive function. J Int Neuropsychol Soc 1998; 4: 547–558.

30.

Bennett

Ong

Ponsford

. Measuring executive dysfunction in an acute rehabilitation setting: using the dysexecutive questionnaire (DEX). J Int Neuropsychol Soc 2005; 11: 376–385.

31.

Bennett

Ong

Ponsford

. Assessment of executive dysfunction following traumatic brain injury: comparison of the BADS with other clinical neuropsychological measures. J Int Neuropsychol Soc 2005; 11: 606–613.

32.

Rappaport

Hall

Hopkins

, et al. Disability rating scale for severe head trauma: coma to community. Arch Phys Med Rehabil 1982; 63: 118–123.

33.

Williams

Smith

. Clinical utility and psychometric properties of the disability rating scale with individuals with traumatic brain injury. Rehabil Psychol 2017; 62: 407–408.

34.

Tate

Simpson

Soo

, et al. Participation after acquired brain injury: clinical and psychometric considerations of the Sydney psychosocial reintegration scale (SPRS). J Rehabil Med 2011; 43: 609–618.

35.

Tate

Pfaff

Veerabangsa

, et al. Measuring psychosocial recovery after brain injury: change versus competency. Arch Phys Med Rehabil 2004; 85: 538–545.

36.

Fleminger

Leigh

Eames

, et al.

HoNOS-ABI: a reliable outcome of neuropsychiatric sequelae to brain injury?

Psychiatr Bull 2005; 29: 53–55.

37.

Morel

Bokossa

Neerchal

. Small sample correction for the variance of GEE estimators. Biom J 2003; 45: 395–409.

38.

Petty

McLennan

Kendall

, et al. Scoping review of return-to-work interventions for persons with traumatic brain injury. Disabil Rehabil 2024; 46: 3243–3255.

39.

Australian Bureau of Statistics. Qualifications and work [Internet]. Canberra: ABS, 2022–23 [cited 2024 July 1]. https://www.abs.gov.au/statistics/people/education/qualifications-and-work/latest-release.

40.

McRae

Kobel

Lukersmith

, et al. What does it take to get somebody back to work after severe acquired brain injury? Service actions within the vocational intervention program (VIP 2.0). Int J Environ Res Public Health 2022; 19: 9548.

41.

Trexler

Parrott

Malec

. Replication of a prospective randomized controlled trial of resource facilitation to improve return to work and school after brain injury. Arch Phys Med Rehabil 2016; 97: 204–210.

42.

Watkin

Phillips

Radford

. What is a ‘return to work’ following traumatic brain injury? Analysis of work outcomes 12 months post TBI. Brain Inj 2020; 34: 68–77.

43.

Spitz

Mahmooei

Ross

, et al. Characterizing early and late return to work after traumatic brain injury. J Neurotrauma 2019; 36: 2533–2540.

44.

Murray

Watter

McLennan

, et al. Identifying models, processes, and components of vocational rehabilitation following acquired brain injury: a systematic scoping review. Disabil Rehabil 2022; 44: 7641–7654.

45.

Wagner

Hammond

Sasser

, et al. Return to productive activity after traumatic brain injury: relationship with measures of disability, handicap, and community integration. Arch Phys Med Rehabil 2002; 83: 107–114.

46.

Murphy

Chamberlain

Weir

, et al. Effectiveness of vocational rehabilitation following acquired brain injury: preliminary evaluation of a UK specialist rehabilitation programme. Brain Inj 2006; 20: 1119–1129.

47.

Tyerman

. Vocational rehabilitation after severe traumatic brain injury: iI-specialist interventions and outcomes. J Occup Psychol Employ Disabil 2000; 2: 13–20.

48.

O’Brien

. Achieving a successful and sustainable return to the workforce after ABI: a client-centred approach. Brain Inj 2007; 21: 465–478.

49.

Gates

Daher

McRae

, et al. Validation of the work-ability support scale in individuals seeking to return to work after severe acquired brain injury. Disabil Rehabil 2023; 46: 3164–3175.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB