Predicting Social Participation and Quality of Life in Individuals with TBI at Least One Year after Discharge from Rehabilitation

Abstract

This is an original follow-up study on predictors of social participation outcomes after traumatic brain injury (TBI) in a prospective longitudinal cohort followed within a universally accessible specialized rehabilitation pathway. The analytic sample comprised 98 individuals with complete clinical data that participated in inpatient and/or outpatient rehabilitation programs at three health centers and completed the outcome measures at a minimum of one year after discharge. The study sample included 26 participants with mild TBI, 23 with complex-mild TBI, 36 with moderate TBI, and 13 with severe TBI. Two outcome measures were administered, the Mayo-Portland Adaptability Inventory-4 (MPAI-4) and the Quality of Life after Brain Injury (QOLIBRI) questionnaire, to investigate social participation and health-related quality of life outcomes in patients with TBI at least one-year post-rehabilitation. Hierarchical regression models assessed the significance of various pre-, peri-, and post-TBI potential predictors in relation to the two outcome measures. Results showed that significant predictors of social participation (MPAI-4 Participation scale) at least one year after rehabilitation included the MPAI-4 Abilities subscale score at intake (p = 0.032) and the delay to rehabilitation intake (p = 0.012). The 8-predictor multivariate model accounted for 33.5% of the variance in social participation. For quality of life as measured by the QOLIBRI, significant predictors were the MPAI-4 Adjustment scale at intake (p = 0.031), employment status post-discharge (p = 0.024), and sports-related cause of injury (p = 0.000). The 10-predictor multivariate model accounted for 45.8% of the variance in quality of life. Our study highlights the importance of following up on individuals with TBI and studying predictors of social participation and quality of life in the years following rehabilitation since they can vary in time. This could help to target appropriate services to develop and offer once rehabilitation has ended, to support favorable long-term outcomes following a brain injury.

Keywords

predictors quality of life rehabilitation social participation traumatic brain injury

Introduction

The impact of traumatic brain injury (TBI) significantly challenges individuals’ ability to return to their social lives, in some cases, long after the TBI incident.^1–3 In general, most adults with severe TBI and many with moderate TBI experience significant difficulties in social and occupational reintegration. In the case of people with mild TBI, evolution is positive in most cases, but several studies show that between 15% and 20% are at risk of presenting persistent symptoms impacting several areas of their lives.^4–6

In the literature looking at TBI outcomes, especially social participation (e.g., return to work or other productive activities, community integration), there is a great deal of inconsistency about the significance of predictors as shown in the findings of several reviews conducted on predictors of social participation after TBI.^7–9 This is because such outcomes depend on the interplay between injury-related impairments and environmental factors, such as sociocultural context, social support, government policies, and accessibility to health care services.^10–12

This study is a follow-up to that conducted by Guerrette and McKerral (2023) within a universally accessible and organized trauma continuum of care. Their study investigated pre-injury, injury-related, and post-injury predictor variables of social participation following TBI as measured by the Mayo-Portland Adaptability Inventory—version 4 (MPAI-4) Participation scale at the end of rehabilitation.^13,14 The authors found that for individuals who underwent inpatient and outpatient rehabilitation, three variables significantly predicted their social participation outcome: years of education and scores for MPAI-4 Ability and Adjustment scales at rehabilitation intake. These variables accounted for 49% of the variance in social participation. For individuals who only received outpatient rehabilitation, five variables significantly predicted social participation outcome: premorbid hypertension and mental health diagnosis, total indirect rehabilitation hours received, and MPAI-4 Abilities and Adjustment scores at rehabilitation intake. These variables accounted for 47% of the variance in social participation.¹³

Their study captured social participation levels at the beginning and end of rehabilitation, and while it enabled measurement of improvements in social participation levels during rehabilitation and identification of its predictors at discharge, it did not permit evaluation of whether these improvements were sustained post-discharge from outpatient rehabilitation and if the predictors of social participation remained the same or evolved. In addition, the MPAI-4 measures functional impairment and extent of disruption in various domains of social participation; it does not address a person’s perceived well-being and quality of life, which are known to be often altered after a TBI.^15,16 Research has consistently shown that TBI can have a profound negative impact on both overall and health-related quality of life.^17–19 This underscores the need to also investigate the predictors of quality of life, particularly within rehabilitation settings and post-discharge. Studying these predictors can inform targeted interventions and improve outcomes for individuals recovering from TBI.

Therefore, our study objectives were to identify variables that could contribute to predicting social participation measured by the MPAI-4 Participation scale as well as quality of life measured by the Quality of Life after Brain Injury (QOLIBRI) scale at least one year following rehabilitation discharge. The QOLIBRI scale is validated and commonly used for assessing health-related quality of life in individuals with TBI.^20–22 This study is novel in that it reports on the one-year follow-up of individuals who received inpatient and/or outpatient rehabilitation in a public and universally accessible health setting and for whom social participation had been previously measured during rehabilitation.

Methods

Experimental design

This study was part of a prospective longitudinal cohort study, from which a subset of the initial cohort was followed up at one year post-rehabilitation. This study was registered at ClinicalTrials.gov (Identifier: NCT05022940) and is reported according to the STROBE statement for cohort studies.²³

Participants

Participants were recruited from a database of adults who spoke French or English and who had received TBI rehabilitation between April 2016 and June 2020 at two Integrated University Health and Social Services Centers and one Integrated Health and Social Services Center in the Greater area of Montreal, QC, Canada. All participants had been diagnosed with TBI and referred to rehabilitation by a health care professional. Ethical approval for the study was obtained from the Research Ethics Board of the Center for Interdisciplinary Research in Rehabilitation of Greater Montreal (project #CRIR-1219-0317).

Procedure and data collection

The French-Canadian version of the MPAI-4 had been previously systematically administered to all participants in the database by consensus of the clinical team upon intake to inpatient or outpatient rehabilitation according to a standardized clinical protocol previously implemented within the rehabilitation programs.²⁴ For the inpatients, this occurred within two weeks of admission, and for the outpatients, within four weeks. The MPAI-4 was also administered at the conclusion of outpatient rehabilitation, within one week of discharge.

Patients with TBI identified in the database were contacted by telephone at least one year after the end of their rehabilitation to ask them to complete the MPAI-4 and the QOLIBRI through a phone interview. A person identified at each rehabilitation site (e.g., research coordination staff, not related to the study research team) made initial contact with potential participants to obtain their consent to be contacted by our research team. The consenting participants were then contacted by the research team to solicit their participation in the study. Two members of the research team completed sociodemographic, MPAI-4, and QOLIBRI questionnaires with all the participants by phone using a detailed verbatim, relying on the participants’ self-reporting. All questionnaires were completed by the participants themselves with no proxies. There were 348 adults (≥16 y at time of TBI) who were contacted by the rehabilitation sites at least one year after their discharge, of which 106 (30.5%) were unreachable. Of the remaining 242 participants, 171 (70.7%) consented to the follow-up assessment (post-rehabilitation M = 19.1 months, SD = 7.2 months) and 71 (29.3%) declined.

Only data from participants who had a complete set of clinical information were included in the analyses (N = 98). A member of the research team had extracted additional sociodemographic and clinical data for each of these participants from the rehabilitation database during visits to the participating sites. Following a comprehensive review of the literature and available information, data were recorded to create both dichotomous and continuous independent variables for the analyses. Participants with complete clinical information (N = 98) and those without (N = 73) did not differ significantly on age, sex, TBI severity, delay to rehabilitation, or MPAI-4 subscale scores at intake or follow-up (all p > 0.05). We also compared the respondents and the nonrespondents (i.e., those who were unreachable or declined to participate) based on age, biological sex, severity, and education. In the respondent group (N = 171), there were 81 (47.4%) females, and the average age was 57.6. In the nonrespondent group (N = 177), there were 74 (41.8%) females, and the average age was 61.9. There were no significant differences between the two groups on these variables (all p > 0.05).

Independent variables

We considered sociodemographic and clinical variables to build our prediction models. These variables were grouped into three categories: pre-injury, injury-related, and post-injury.¹³ The MPAI-4 Participation and Total scores at discharge were not included in both prediction models and employment status post-discharge was not included in the social participation model due to multicollinearity issues. For the social participation model, variable selection was grounded in the findings of Guerrette and McKerral (2023), who predicted the same outcome in a similar population. Variables significant in their multivariate models were retained alongside our study variables passing a p < 0.05 bivariate Pearson correlation screen with outcome. For the quality of life model, no prior study has predicted QOLIBRI outcomes in a comparable TBI rehabilitation sample. Variable selection therefore relied on bivariate correlations (p < 0.05) between our study variables and outcome.

Outcome variables

The MPAI-4 Participation scale score, obtained at least one year after discharge from outpatient rehabilitation, was used as the measure of social participation. Participants who had only MPAI-4 data collected at intake at least one year before but not at end of rehabilitation (e.g., because they continued rehabilitation at a different site) were also included. The psychometric properties of the French-Canadian version of the MPAI-4 are described elsewhere.²⁴

The QOLIBRI was used as a measure of health-related quality of life. It assesses the cognitive, affective, functional, relational, emotional, and physical domains of quality of life.¹⁷ It has been adapted and validated in 10 languages.^18,19,25

Statistical analysis

We used linear hierarchical regression to evaluate our prediction models, ensuring adherence to statistical assumptions. Consistent with Guerrette and McKerral (2023), predictors were organized into three sequential blocks representing pre-injury, injury-related, and post-injury characteristics. A fourth block was added, comprising MPAI-4 Abilities and Adjustment subscale scores at rehabilitation intake. These variables were placed in a separate final block because they represent a standardized clinical assessment of TBI-related functional status at the start of rehabilitation, which is a conceptually distinct category from the sociodemographic, injury, and post-injury variables in the earlier blocks—and entering them last allowed us to evaluate how much additional predictive value functional status at intake provided beyond all other predictors.

The dataset contained no significant outliers and minimal missing data. Any variable with a small number of cases (n < 10) was excluded from the analysis because of their potential to limit the stability and the generalizability of the estimated effects. While most variables met assumptions of normality, a few exhibited skewness and kurtosis values slightly exceeding the recommended thresholds. However, following Field’s (2013) recommendation, this deviation was deemed acceptable given the use of bootstrapped confidence intervals.²⁶

Multicollinearity was assessed using variance inflation factors (VIF) and tolerance statistics, which were all within acceptable limits. The independence of errors was tested with the Durbin–Watson statistic, yielding satisfactory results. All statistical analyses were conducted with a significance level of 0.05 and 95% confidence intervals. Python was used for all analysis.

Results

Descriptive statistics

Table 1 presents the descriptive statistics for the study participants, detailing key sociodemographic and pre-injury variables, as well as other variables of interest.

Table 1.

Descriptive Statistics and Participant Characteristics (N = 98)

Variables	N	Frequency (%)	M (SD)
Pre-injury
Age			50.73 (19.6)
Gender
Male	53	54.1
Female	45	45.9
Ethnicity
White	73	74.5
Non-white	25	25.5
Employment status
Working/student	69	70.4
Retired	23	23.5
Not working	6	6.1
Health conditions
Diabetes	12	12.2
Hypertension	25	25.5
Cardiovascular disease	23	23.5
Pulmonary disease	5	5.1
Other diseases	15	15.3
Mental health diagnosis	36	36.7
Injury-related
Cause of accident
Motor vehicle	23	23.5
Sports	18	18.4
Other	57	58.1
Neck fracture	7	7.1
Skull fracture	26	26.5
Severity
Mild	26	26.5
Complex-mild	23	23.5
Moderate	36	36.7
Severe	13	13.3
Medical imaging
CT scan positive	67	68.4
MRI positive	11	11.2
Post injury
Orthopedic surgery	15	15.3
Delay before intake (days)			183.23 (237)
Employment status post-discharge
Working/student	35	35.8
Retired	34	34.6
Not working	29	29.6
MPAI scores at intake
Abilities scale			10.56 (6.73)
Adjustment scale			16.64 (6.59)
Participation scale			12.02 (7.27)
MPAI scores post-discharge^a
Abilities scale			13.5 (10.11)
Adjustment scale			14.88 (10.93)
Participation scale			8.18 (6.77)
QOLIBRI^a
Total score			64.21 (19.9)

At post-injury follow-up both the MPAI-4 and QOLIBRI were measured at least one year after discharge from rehabilitation.

N, number; M, mean; SD, standard deviation; CT, computed tomography; MRI, magnetic resonance imaging; MPAI-4, Mayo-Portland Adaptability Inventory-4; QOLIBRI, quality of life after brain injury.

Social participation

Table 2 shows the bivariate correlations between social participation outcome and significant predictors from Guerrette and McKerral (2023), as well as the significant bivariate correlations between remaining study variables and outcome. The initial variable pool for the social participation model included education, premorbid hypertension, premorbid mental health diagnosis, delay between TBI and start of rehabilitation, MPAI-4 Abilities at intake, and MPAI-4 Adjustment at intake. We also included the following significant variables from our bivariate screen (p < 0.05): age at TBI and cause of injury—sports versus all other causes. In post hoc analysis, we verified and confirmed that motor vehicle and other causes of injury did not contribute any independent variance to the model.

Table 2.

Variables Correlated with Social Participation at Least One Year After Discharge from Rehabilitation

Variables	R	P
Pre-injury
Age at TBI (years)	0.220	0.030*
Education (years)	−0.225	0.027*
Premorbid hypertension	0.164	0.107
Premorbid mental health diagnosis	0.057	0.574
Injury-related
Cause of Accident: sports vs others	−0.234	0.020*
Post-injury
Delay before intake (days)	0.161	0.113
MPAI-4 scores at intake
Abilities scale	0.437	0.000**
Adjustment scale	0.352	0.000**

*p < 0.05; **p < 0.001.

MPAI-4, Mayo-Portland Adaptability Inventory-4; TBI, traumatic brain injury.

Rehabilitation hours received, a significant predictor in Guerrette and McKerral (2023), could not be included because it was collected at rehabilitation discharge from clinical records and was not available in the present follow-up dataset.

Table 3 presents the results of the hierarchical regression model predicting social participation at least one year after rehabilitation discharge. The final social participation model comprised eight predictors in four hierarchical blocks. The model was statistically significant, accounting for 33.5% of the variance. The optimism-corrected R², estimated via bootstrap resampling (1,000 samples), indicated minimal overfitting.

Table 3.

Hierarchical Regression Predicting Social Participation (MPAI-4) at ≥1-Year Follow Up

Block/Variable	B	SE	β	95% CI	p
Block 1: Pre-injury, ΔR² = .098, ΔF(4, 89) = 2.41, p = .055^†
Age at TBI (years)	0.071	0.040	0.203	[−0.020, 0.160]	0.076^†
Education (years)	−0.299	0.174	−0.161	[−0.636, 0.012]	0.089^†
Pre-morbid hypertension	−0.014	1.719	−0.001	[−3.454, 3.168]	0.994
Pre-morbid mental health dx	0.516	1.344	0.037	[−1.938, 3.160]	0.702
Block 2: Injury-related, ΔR² = .017, ΔF(1, 88) = 1.72, p = .193
Cause: Sport vs all other causes	−2.195	1.658	−0.124	[−4.178, 0.128]	0.189
Block 3: Post-injury, ΔR² = .051, ΔF(1, 87) = 5.30, p = .024*
Delay to rehabilitation (days)	0.007	0.003	0.243	[0.003, 0.013]	0.012*
Block 4: MPAI-4 intake, ΔR² = .169, ΔF(2, 85) = 10.80, p < .001**
MPAI-4 Abilities (intake)	0.278	0.127	0.274	[0.043, 0.497]	0.032*
MPAI-4 Adjustment (intake)	0.195	0.125	0.189	[−0.003, 0.403]	0.123
R² = .335, Adj. R² = .273, F(8, 85) = 5.35, p < .001 \| Optimism-corrected R² = .380 \| DW = 2.09

95% CIs based on 1,000 bootstrap samples (percentile method). Age × Hypertension r = .547 (VIF age = 7.76; hypertension = 1.93). MPAI-4 Abilities × Adjustment r = .633 (VIF = 7.16 and 12.00); block-level ΔR² is primary inference for Block 4. The events-per-variable ratio (EPV) was 11.75.

^†p < 0.10; *p < 0.05; **p < 0.001.

B, unstandardized coefficient; SE, standard error; β, standardized coefficient.

In the model, two variables were significant independent predictors. Delay between TBI and start of rehabilitation was a significant positive predictor, indicating longer delays were associated with worse participation. MPAI-4 Abilities at intake was a significant positive predictor, indicating higher functional impairment at the start of rehabilitation predicted worse participation one year later. Age and education showed trend-level associations in the expected directions. The two MPAI-4 subscales were correlated; the block-level ΔR² provided the most reliable estimate of their combined contribution.

Quality of life

Table 4 shows the significant bivariate correlations between study variables and quality of life outcome. The final quality of life model comprised 10 predictors in four hierarchical blocks: sex, premorbid mental health diagnosis, TBI severity (complex mild), skull fracture, cause of injury (sports), delay between TBI and start of rehabilitation, employment status post-discharge (working, retired), MPAI-4 Abilities at intake, and MPAI-4 Adjustment at intake.

Table 4.

Variables Correlated with Quality of Life at Least One Year After Rehabilitation

Variable	R	p value
Pre-injury
Gender	−0.26	0.011*
Mental health diagnosis	−0.22	0.033*
Injury-related
Cause (sport)	0.31	0.002**
Skull fracture	0.33	0.001**
Severity (complex mild)	0.26	0.013*
Post-injury
Delay before intake	−0.26	0.010**
MPAI-4 scores at intake
Abilities scale	−0.28	0.006**
Adjustment scale	0.41	<0.001***
Post-discharge employment
Employed	0.23	0.026*
Retired	−0.36	<0.001***

*p < 0.05; **p < 0.01; ***p < 0.001.

MPAI-4, Mayo-Portland Adaptability Inventory-4; QoL, quality of life; SP, social participation.

Table 5 presents the results of the hierarchical regression model predicting quality of life at least one year after rehabilitation discharge. The model was statistically significant, accounting for 45.8% of the variance in quality of life. The optimism-corrected R², estimated via bootstrap resampling (1,000 samples), indicated minimal overfitting. All four blocks contributed significantly to the model.

Table 5.

Hierarchical Regression Predicting Quality of Life (QOLIBRI) at ≥1-Year Follow-Up

Block/Variable	B	SE	β	95% CI (boot)	p
Block 1: Pre-injury, ΔR² = .0986, ΔF = 5.030), p = .0085*
Sex (female = 2)	−6.170	3.502	−0.157	[−12.741, 0.716]	0.082^†
Pre-morbid mental health dx	−4.747	3.500	−0.117	[−10.915, 1.305]	0.179
Block 2: Injury-related, ΔR² = .3316, ΔF = 10.344, p = .0000**
Severity: Complex mild (re f= mild)	5.495	4.313	0.118	[−5.565, 16.157]	0.206
Skull fracture	6.821	4.084	0.153	[−1.218, 14.900]	0.099^†
Cause: Sport vs all other causes	16.423	4.272	0.320	[8.537, 24.035]	0.000**
Block 3: Post-injury, ΔR² = .4008, ΔF = 3.311, p = .0238*
Delay to rehabilitation (days)	−0.010	0.008	−0.117	[−0.026, 0.002]	0.210
Employment (post-discharge): Working (ref = Not working)	10.002	4.353	0.244	[−3.067, 11.244]	0.024*
Employment (post-discharge): Retired (ref = Not working)	6.101	4.293	0.147	[−16.101, 4.471]	0.159
Block 4: MPAI-4 intake, ΔR² = .4584, ΔF = 4.461, p = .0144*
MPAI-4 Abilities (intake)	−0.057	0.355	-0.019	[−0.756, 0.679]	0.873
MPAI-4 Adjustment (intake)	−0.758	0.346	−0.253	[−1.395, −0.125]	0.031*
R² = 0.4584, Adj.R² = 0.3939, F(10,84) = 7.108, P = 0.0000 \| Optimism-corrected R² = 0.5130

95% CIs based on 1,000 bootstrap samples (percentile method). The events-per-variable ratio (EPV) was 9.5.

^†p < 0.10; *p < 0.05; **p < 0.001.

B, unstandardized coefficient; SE, standard error; β, standardized coefficient.

Three variables emerged as significant independent predictors of quality of life at follow-up. The cause of injury—sport versus all other causes—was the strongest predictor, indicating that participants who sustained a sport-related TBI reported substantially better quality of life one year after rehabilitation discharge compared with those whose injury resulted from other causes. Employment status post-discharge was another significant predictor when employment status was compared with the nonworking group as the reference category: participants who were employed following discharge reported significantly better quality of life than those who were not working, corresponding to approximately 10 points on the QOLIBRI scale. MPAI-4 Adjustment subscale score at rehabilitation intake was also a significant negative predictor, indicating that more adjustment difficulties at the start of rehabilitation were associated with lower quality of life at follow-up of one year after rehabilitation discharge.

Discussion

Our study aimed to identify significant predictors of social participation (Participation scale of the MPAI-4) and quality of life (QOLIBRI) for individuals with TBI at least one year after the end of their rehabilitation. Longer delays to rehabilitation access and higher functional impairment at intake predicted lower social participation. Sports-related injury and being employed post-discharge predicted better quality of life, while lower adjustment at intake predicted lower quality of life.

Our findings about delay to access rehabilitation and functional impairment as predictors of social participation are consistent with previous research. Studies have shown that deficits in various elements of the MPAI-4 Abilities subscale affect social participation in individuals with TBI, including return to work, return to driving, and other variables.^27–33 While these studies did not use the MPAI-4 Abilities subscale directly, they addressed many of its components. For example, a systematic review³⁴ noted that cognitive and behavioral sequelae often hinder successful professional outcomes, even many years post-injury. Likewise, it has been found that individuals who reported greater cognitive challenges and reduced mobility one year after injury experienced more employment instability.³⁵

For the other significant predictor of social participation in our study, which was the delay before intake into rehabilitation, research suggests that such delays can negatively impact recovery a year or more after completing a rehabilitation program.^36–38 High Jr and colleagues³⁹ found that the greatest improvements on variables such as community integration were observed in patients who were admitted to post-acute rehabilitation within six months of their injury. These improvements continued even after discharge. Rehabilitation may exert its benefits not only by facilitating improvements in physical functions but also by preventing further declines in community functioning.⁴⁰

The second outcome studied was quality of life. The prediction model relied on bivariate screening applied to the variables of interest, as no prior study had examined predictors of quality of life in a comparable TBI rehabilitation cohort. Guerrette and McKerral’s (2023) findings pertaining to social participation could not be assumed to generalize to a different construct. The significance of the MPAI-4 Adjustment scale score at intake in our model may be explained by the cognitive and psychological impairments associated with TBI, which can continue to impact quality of life in the years after the injury.⁴¹ Research literature has extensively explored various sub-elements that are captured by the scale, including mood outcomes related to anxiety and depression, irritability and anger, pain and fatigue, altered self-awareness, and, crucially, social support.^42–46 The emphasis across studies is on the importance of accurate and timely assessment of these factors at the time of injury and the need for immediate, consistent intervention, beyond the rehabilitation phase, to ensure better and more sustainable psychosocial and behavioral adjustment outcomes.

Cause of injury was a significant predictor of quality of life, where TBI participants who had been injured through sports-related incidents had a better quality of life one year or more after discharge compared with those injured through other non-sports-related causes. Occupational reintegration appeared to be one of the associated mechanisms: 50% of sport-injured participants were employed post-rehabilitation discharge compared with 32.5% of those with other causes of injury. Previous research reached various conclusions in understanding this advantage^47–51 and suggested that some characteristics of sport-injured individuals may account for this advantage, such as showing better recovery of mental health and post-concussion symptoms than those injured in another context, which may be associated with quicker return to regular exercise and more resilience to some post-TBI effects.

Our results also showed that participants who were employed post-discharge reported significantly better quality of life than those who were not. This finding is consistent with a considerable body of literature documenting the central role of return to work in health-related quality of life following TBI.^52,53 More so, retired participants in our study did not differ significantly from nonworking participants, suggesting that the quality of life advantage of the employed group could reflect active occupational reintegration after TBI, rather than a general age-related effect or a pre-injury life course trajectory. These results underscore the importance of vocational and return-to-work objectives as integral components of TBI rehabilitation, since they clearly contribute to post-rehabilitation quality of life.

The social participation model in this study was developed as an extension of Guerrette and McKerral’ (2023) models, which predicted social participation in the same population using the same instrument. In both studies, functional abilities and delay before rehabilitation intake were significant predictors. This supports the conclusion that MPAI-4 functional status at the start of rehabilitation represents a durable prognostic marker whose predictive value persists well beyond the end of formal care. Likewise, no injury-related variable was a significant multivariate predictor of social participation, indicating that post-TBI impacts can vary across individuals with similar injury severity indicators. Compared with Guerrette and McKerral’s models that explained between 47% and 49% of variance in social participation, the present model explained 33.5%. A substantial part of this gap may be attributable to the rehabilitation hours variable, a significant predictor in their models, which could not be included in the present study as it was not available in the present follow-up dataset. It is also likely that different prognostic indicators of social participation that we did not capture are at play one year after rehabilitation versus at end of rehabilitation. It is noteworthy to underscore that the present study built on this previous work by adding a quality of life outcome at one year post-rehabilitation. The fact that the two outcomes studied have distinct predictor profiles reinforces that they measure different constructs and that both deserve attention in TBI rehabilitation and research.

Finally, it is worth noting that some participants in our study wished to emphasize after completing our questionnaires that they would have benefited from immediate counseling and guidance at the time of their TBI incident. They believed that early intervention could have helped them prevent rumination on their symptoms and their negative impacts. Future research should explore how access to early interventions and delay to rehabilitation intake affect such post-TBI difficulties. Many participants also stressed the wish to have benefited from regular follow-ups following rehabilitation. The availability and effects of such services should be appropriately studied since they could have a positive influence on TBI patients’ social participation and quality of life in the years after completing rehabilitation.

Study limitations

Several limitations must be considered in interpreting the findings of this study. First, the sample composition was relatively homogenous, possibly limiting the generalizability of findings to more diverse cultural backgrounds. Another issue lies in the inability to access some participants’ clinical information, which lowered sample size and may have reduced the robustness of the analysis. Also, there could be potential influence of unknown or unmeasured socio-environmental factors. For instance, participants who moved to a different city, experienced changes in access to professional resources, or faced the loss of family members might have experienced variations in their social participation and quality of life that are not captured in the study. Furthermore, a shift in the measurement modality represents a limitation of our study, as different rating methods were used at intake (clinician consensus rating approach) compared with the one-year follow-up (participant self-report via telephone interviews). These differing measurement modalities need to be taken into consideration when interpreting and generalizing the results. The reliance on self-reported measures to assess both social participation and quality of life can introduce bias. These responses may be influenced by individual perceptions, mood, or social desirability biases. Factors such as memory loss, impaired self-awareness, attention deficits, and fatigue could also affect participants’ responses on both the MPAI-4 and QOLIBRI instruments. These limitations should be addressed in future research to enhance the generalizability and validity of the findings.

Conclusions

Our study highlights the importance of studying specific predictors of social participation and quality of life outcomes of individuals with TBI in the years after completing rehabilitation. We conclude that employing the MPAI-4 and QOLIBRI measures are useful tools for evaluating patients’ self-perceived social participation and quality of life in the post-rehabilitation phase, respectively. This study demonstrated that participants with higher functional impairment and who showed a longer delay before accessing rehabilitation were at higher risk of experiencing sustained reductions in social participation. Participants with more adjustment difficulties had lower quality of life, and those whose TBI was sport-related and who were employed post-discharge reported higher quality of life. As these findings remain preliminary, future studies should focus on developing a more complete understanding of the variables that impact and predict social participation and quality of life in the years following rehabilitation. This should lead to the development of sustainable and cost-effective interventions that address the specific needs of patients with TBI and their support system after the end of formal rehabilitation services.

Transparency, Rigor, and Reproducibility Statement

This study was registered at ClinicalTrials.gov (Identifier: NCT05022940) and is reported in accordance with the STROBE statement for cohort studies. It is a longitudinal follow-up to Predictors of social participation outcome after traumatic brain injury differ according to rehabilitation pathways, conducted within an accessible and organized trauma continuum of care (Guerrette and McKerral, 2023). The sample size was determined by the number of participants meeting eligibility criteria (≥16 years at the time of TBI, with no premorbid neurodegenerative disease) and available for follow-up at least one year after discharge from rehabilitation. There were initially 348 adults (≥16 y at time of TBI) who were contacted by the rehabilitation sites at least one year after their discharge, of which 106 (30.5%) were unreachable. Of the remaining 242 participants, 171 (70.7%) consented to the follow-up assessment and 71 (29.3%) declined. Only data from participants who had a complete set of clinical information were included in the analyses (N = 98). Multiple linear regression analyses were conducted to evaluate the predictive models. The dataset contained no influential outliers and minimal missing data. Most variables met assumptions of normality; a small number exhibited skewness and kurtosis values slightly exceeding recommended thresholds. These deviations were considered acceptable given the use of bootstrapped confidence intervals, following Field (2013). Multicollinearity was assessed using VIF and tolerance statistics, and independence of errors was evaluated using the Durbin–Watson statistic. Dichotomous or categorical variables with fewer than 10 cases in any category were excluded to ensure robustness of the statistical models. All statistical analyses were conducted using a two-tailed significance level of 0.05 with 95% confidence intervals. Python was used for all analyses. The primary clinical outcome measures are established standards in the field. Validation of the MPAI-4 has been previously reported (Guerrette and McKerral, 2022), and validation of the QOLIBRI has been documented in prior studies (Giustini et al., 2014). The dataset generated and analyzed during the current study is not publicly available due to absence of prior authorization from participants to share their information with a third party, even if deidentified (Quebec Bill 25). Requests to share data can be addressed to the corresponding author; they will be considered on a case-by-case basis by the institutional review board and may require a data use agreement.

Authors’ Contributions

Z.A.: Data collection and management (lead), writing—original draft (equal), writing—review and editing (supporting), project administration (equal), visualization (lead), and validation (supporting). F.B.: Data collection (supporting), formal analysis (lead), writing—original draft (equal), writing—review and editing (supporting), methodology (equal), project administration (equal), and supervision (supporting). M.M.: Conceptualization (lead), funding acquisition (lead), methodology (equal), supervision (lead), validation (lead), and writing—review and editing (lead).

Footnotes

Acknowledgments

The authors wish to thank the clinicians and program managers from the TBI rehabilitation programs at the two Integrated University Health and Social Services Centres (CIUSSS) and the Integrated Health and Social Services Centre (CISSS) that participated in the study: the CIUSSS du Centre-Sud-de-l’Île-de-Montréal, the CIUSSS du Centre-Ouest-de-l’Île-de-Montréal, and the CISSS de Laval.

Author Disclosure Statement

The authors report no competing financial interests or other conflicts of interest.

Funding Information

This research was supported by a research grant from the Fonds de recherche du Québec—Santé (FRQS) to Michelle McKerral [grant number 254599].

Abbreviations Used

References

1. Pavlović

, Pekić

, Stojanović

, et al. Traumatic brain injury: Neuropathological, neurocognitive and neurobehavioral sequelae. Pituitary 2019;22(3):270–282.

2. Jourdan

, Bayen

, Pradat-Diehl

, et al. A comprehensive picture of 4-year outcome of severe brain injuries: Results from the PariS-TBI study. Ann Phys Rehabil Med 2016;59(2):100–106; doi: 10.1016/j.rehab.2015.10.009

3. Timmer

, Jacobs

, Schönherr

, et al. The spectrum of long-term behavioral disturbances and provided care after traumatic brain injury. Front Neurol 2020;11:246; doi: 10.3389/fneur.2020.00246

4. Theadom

, Barker-Collo

, Jones

, et al.; BIONIC4you Research Group. Work limitations 4 years after mild traumatic brain injury: A cohort study. Arch Phys Med Rehabil 2017;98(8):1560–1566; doi: 10.1016/j.apmr.2017.01.010

5. Cooksley

, Maguire

, Lannin

, et al. Persistent symptoms and activity changes three months after mild traumatic brain injury. Aust Occup Ther J 2018;65(3):168–175; doi: 10.1111/1440-1630.12457

6. Machamer

, Temkin

, Dikmen

, .t.al.; TRACK-TBI Investigators. Symptom frequency and persistence in the first year after traumatic brain injury: A TRACK-TBI study. J Neurotrauma 2022;39(5–6):358–370; doi: 10.1089/neu.2021.0348

7. Kersey

, Terhorst

, Wu

, et al. A scoping review of predictors of community integration following traumatic brain injury: A search for meaningful associations. J Head Trauma Rehabil 2019;34(4):E32–E41; doi: 10.1097/HTR.0000000000000442

8. Algethamy

. Baseline predictors of survival, neurological recovery, cognitive function, neuropsychiatric outcomes, and return to work in patients after a severe traumatic brain injury: An updated review. Mater Sociomed 2020;32(2):148–157; doi: 10.5455/msm.2020.32.148-157

9. Van Deynse

, 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(20):5750–5757; doi: 10.1080/09638288.2021.1954247

10.

10. Silva

, Hoffman

, O’Neil-Pirozzi

, et al. Environmental barriers are associated with rehabilitation needs 10 to 15 years after traumatic brain injury: A Veterans Affairs TBI model systems study. J Head Trauma Rehabil 2025;40(1):44–53; doi: 10.1097/HTR.0000000000000864

11.

11. Hart

, Rabinowitz

. Changes in social participation between 1 and 2 years following moderate-severe traumatic brain injury. Front Rehabil Sci 2022;3:945699; doi: 10.3389/fresc.2022.945699

12.

12. Mollayeva

, Mollayeva

, Pacheco

, et al. Systematic review of sex and gender effects in traumatic brain injury: Equity in clinical and functional outcomes. Front Neurol 2021;12:678971; doi: 10.3389/fneur.2021.678971

13.

13. Guerrette

M-C

, McKerral

. Predictors of social participation outcome after traumatic brain injury differ according to rehabilitation pathways. J Neurotrauma 2023;40(5–6):523–535; doi: 10.1089/neu.2022.0232

14.

Foundation to Advance Brain Rehabilitation (FABR): Rochester, MN; 2023.

15.

15. Cicerone

, Mott

, Azulay

, et al. Community integration and satisfaction with functioning after intensive cognitive rehabilitation for traumatic brain injury. Arch Phys Med Rehabil 2004;85(6):943–950; doi: 10.1016/j.apmr.2003.07.019

16.

16. Johnston

, Goverover

, Dijkers

. Community activities and individuals’ satisfaction with them: Quality of life in the first year after traumatic brain injury. Arch Phys Med Rehabil 2005;86(4):735–745; doi: 10.1016/j.apmr.2004.10.031

17.

17. Anderson

JFI

. Heterogeneity of health-related quality of life after mild traumatic brain injury with systemic injury: A cluster analytic approach. Disabil Rehabil 2025;47(2):347–356; doi: 10.1080/09638288.2024.2345278

18.

18. Mamman

, Grewal

, Garrone

, et al. Biopsychosocial factors of quality of life in individuals with moderate to severe traumatic brain injury: A scoping review. Qual Life Res 2024;33(4):877–901; doi: 10.1007/s11136-023-03511-0

19.

19. Rauen

, Späni

, Tartaglia

, et al. Quality of life after traumatic brain injury: A cross-sectional analysis uncovers age- and sex-related differences over the adult life span. Geroscience 2021;43(1):263–278; doi: 10.1007/s11357-020-00273-2

20.

Truelle

, Koskinen

, Hawthorne

et al.; Qolibri Task Force et al.; Qolibri Task Force. Quality of life after traumatic brain injury: The clinical use of the QOLIBRI, a novel disease-specific instrument. Brain Inj 2010;24(11):1272–1291; doi: 10.3109/02699052.2010.506865

21.

21. Gorbunova

, Zeldovich

, Voormolen

, et al. Reference values of the QOLIBRI from general population samples in the United Kingdom and The Netherlands. J Clin Med 2020;9(7):2100; doi: 10.3390/jcm9072100

22.

22. Iderdar

, Covic

, Al Wachami

, et al. Cross-cultural adaptation of the Quality of Life after Brain Injury (QOLIBRI) questionnaire in patients after traumatic brain injury in Morocco. Disabil Rehabil 2025;47(15):3958–3967; doi: 10.1080/09638288.2024.2436983

23.

23. von Elm

, Altman

, Egger

, et al.; STROBE Initiative. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: Guidelines for reporting observational studies. J Clin Epidemiol 2008;61(4):344–349; doi: 10.1016/j.jclinepi.2007.11.008

24.

24. Guerrette

, McKerral

. Validation of the Mayo-Portland Adaptability Inventory-4 (MPAI-4) and reference norms in a French-Canadian population with traumatic brain injury receiving rehabilitation. Disabil Rehabil 2022;44(18):5250–5256; doi: 10.1080/09638288.2021.1924882

25.

25. Giustini

, Longo

, Azicnuda

, et al. Health-related quality of life after traumatic brain injury: Italian validation of the QOLIBRI. Funct Neurol 2014;29(3):167–176; doi: 10.11138/fneur/2014.29.3.167

26.

26. Field

. Discovering Statistics Using IBM SPSS Statistics, Paperback ed. Sage Publications: Thousand Oaks, CA; 2013.

27.

27. Colantonio

, Salehi

, Kristman

, et al. Return to work after work-related traumatic brain injury. NeuroRehabilitation 2016;39(3):389–399; doi: 10.3233/NRE-161370

28.

28. Odgaard

, Pedersen

, Poulsen

, et al. Return to work predictors after traumatic brain injury in a welfare state. Acta Neurol Scand 2018;137(1):44–50; doi: 10.1111/ane.12806

29.

29. Erler

, Whiteneck

, Juengst

, et al. Predicting the trajectory of participation after traumatic brain injury: A longitudinal analysis. J Head Trauma Rehabil 2018;33(4):257–265; doi: 10.1097/HTR.0000000000000383

30.

30. Xiong

, Martin

, Sravanapudi

, et al. Factors associated with return to work in men and women with work-related traumatic brain injury. Disabil Health J 2016;9(3):439–448; doi: 10.1016/j.dhjo.2015.12.002

31.

31. Cullen

, Krakowski

, Taggart

. Functional independence measure at rehabilitation admission as a predictor of return to driving after traumatic brain injury. Brain Inj 2014;28(2):189–195; doi: 10.3109/02699052.2013.862738

32.

32. Eastvold

, Walker

, Curtiss

, et al. The differential contributions of posttraumatic amnesia duration and time since injury in prediction of functional outcomes following moderate-to-severe traumatic brain injury. J Head Trauma Rehabil 2013;28(1):48–58; doi: 10.1097/HTR.0b013e31823c9317

33.

33. Malone

, Erler

, Giacino

, et al. Participation following inpatient rehabilitation for traumatic disorders of consciousness: A TBI model systems study. Front Neurol 2019;10:1314–1310; doi: 10.3389/fneur.2019.01314

34.

34. Manoli

, Delecroix

, Daveluy

, et al. Impact of cognitive and behavioural functioning on vocational outcome following traumatic brain injury: A systematic review. Disabil Rehabil 2021;43(18):2531–2540; doi: 10.1080/09638288.2019.1706105

35.

35. Ponsford

, Spitz

. Stability of employment over the first 3 years following traumatic brain injury. J Head Trauma Rehabil 2015;30(3):E1–E11; doi: 10.1097/HTR.0000000000000033

36.

36. Andelic

, Bautz-Holter

, Ronning

, et al. Does an early onset and continuous chain of rehabilitation improve the long-term functional outcome of patients with severe traumatic brain injury? J Neurotrauma 2012;29(1):66–74; doi: 10.1089/neu.2011.1811

37.

37. Hayashi

, Kamo

, Momosaki

. Effectiveness of early rehabilitation interventions in patients with traumatic brain injury using a large database. Pm R 2025;17(2):170–177; doi: 10.1002/pmrj.13243

38.

38. Buh

, Hutchinson

PJA

, Anwar

, et al. Buh

, Hutchinson

PJA

, Anwar

, et al. Early neurorehabilitation in traumatic brain injury: The need for an African perspective. BMC Med 2023;21(1):290; doi: 10.1186/s12916-023-03009-z

39.

39. High

Jr , Roebuck-Spencer

, Sander

, et al. Early versus later admission to postacute rehabilitation: Impact on functional outcome after traumatic brain injury. Arch Phys Med Rehabil 2006;87(3):334–342; doi: 10.1016/j.apmr.2005.11.028

40.

40. Cicerone

. Participation as an outcome of traumatic brain injury rehabilitation. J Head Trauma Rehabil 2004;19(6):494–501; doi: 10.1097/00001199-200411000-00006

41.

41. Azouvi

, Jourdan

, Ghout

, et al. Predictors and indicators of disability and quality of life 4 years after a severe traumatic brain injury: A structural equation modelling analysis from the PariS-TBI study. Ann Phys Rehabil Med 2015;58:e145; doi: 10.1016/j.rehab.2015.07.344

42.

42. Miles

, Hammond

, Neumann

, et al. Evolution of irritability, anger, and aggression after traumatic brain injury: Identifying and predicting subgroups. J Neurotrauma 2021;38(13):1827–1833; doi: 10.1089/neu.2020.7451

43.

43. Agtarap

, Campbell-Sills

, Jain

, et al.; TRACK-TBI Investigators. Satisfaction with life after mild traumatic brain injury: A TRACK-TBI study. J Neurotrauma 2021;38(5):546–554; doi: 10.1089/neu.2020.7055

44.

44. Fahey

, Bain

, Critchfield

. I couldn’t be better: Assessing self-awareness with the mayo-Portland adaptability inventory-4 following TBI. J Head Trauma Rehabil 2021;36(6):E373–E380; doi: 10.1097/HTR.0000000000000670

45.

45. Muliira

, Lazarus

, Jacob

, et al. The needs of families caring for patients with traumatic brain injury: A scoping review. Disabil Rehabil 2024;46(20):4586–4594; doi: 10.1080/09638288.2023.2278178

46.

46. Takada

, Sashika

, Wakabayashi

, et al. Social participation and quality of life of patients with traumatic brain injury living in the community: A mixed methods study. Brain Inj 2016;30(13–14):1590–1598; doi: 10.1080/02699052.2016.1199901

47.

47. Lennon

, Rigney

, Creese

, et al. Sports-related concussion not associated with long-term cognitive or behavioural deficits: The PROTECT-TBI study. J Neurol Neurosurg Psychiatry 2025;96(4):397–405; doi: 10.1136/jnnp-2024-334039

48.

48. Roby

, Mozel

, Arbogast

, et al. Postinjury outcomes after non-sport-related concussion: A CARE consortium study. J Athl Train 2024;59(3):289–296; doi: 10.4085/1062-6050-0181.23

49.

49. Nichols

, Schindler

, Kapoor

, et al. Motor vehicle collision associated TBI: Predictors of injury severity. Orthop Rev (Pavia) 2025;17:137678; doi: 10.52965/001c.137678

50.

50. Brady

, Hume

, Mahon

, et al. What is the evidence on natural recovery over the year following sports-related and non-sports-related mild traumatic brain injury: A scoping review. Front Neurol 2021;12:756700; doi: 10.3389/fneur.2021.756700

51.

51. Ntikas

, Stewart

, Ietswaart

, et al.; CENTER-TBI participants and investigators. Contrasting characteristics and outcomes of sports-related and non–sports-related traumatic brain injury. JAMA Netw Open 2024;7(1):e2353318; doi: 10.1001/jamanetworkopen.2023.53318

52.

52. Andelic

, Howe

, Hellstrøm

, et al. Disability and quality of life 20 years after traumatic brain injury. Brain Behav 2018;8(7):e01018; doi: 10.1002/brb3.1018

53.

53. Klang

, Molero

, Bergström

, et al. Five-year follow-up of work disability after traumatic brain injury: A nationwide Swedish matched cohort study of 98,000 individuals. Neurology 2026;106(5):e214674; doi: 10.1212/WNL.0000000000214674