Modeling community integration in workers with delayed recovery from mild traumatic brain injury

In this study of 94 workers with mTBI, we found that insomnia, head or neck pain, being married or in a relationship, TSI, and a diagnosis of possible/probable malingering were independently associated with limited CI.

Unfortunately, there are no normative data for CIQ scores in mTBI. However, the CIQ total and subscale scores in the present study were similar to the mean scores at 1 year post-injury in participants with more severe TBI, as reported by Sanders et al. (GCS: 8.43 ± 3.8) [40] and Seale et al. (GCS: 6.5 ± 3.7) [41]. As in previous studies [42, 43], the highest scores were observed in the social integration domain and the lowest scores were observed in the productive activities domain. However, our participants had slightly higher scores in the home integration domain. One possible explanation for this discrepancy is that our study evaluated workers with persistent symptoms, and the majority of participants were receiving disability benefits at their time of assessment. Thus, our data suggests that disability status is strongly associated with impaired productive activity, although it had a lesser effect on social and family integration in our study.

Our findings also highlight the associations between insomnia, pain, and depression, in agreement with earlier reports [42‐49]. However, insomnia has been viewed as a symptom that occurs in the context of another disorder (i.e., depression [42‐45], pain [46‐48], or brain injury [49]), rather than as an independent disorder within mTBI. In many depressed individuals, however, insomnia signals the onset of depression and may prevent recovery, even after adequate treatment [50, 51]. Similarly, the stress created by insomnia may aggravate or serve as a catalyst for the involvement of the hypothalamic orexinergic or hypocretinergic system in pain and headaches [52]. This may be relevant for our participants, as the majority continued to experience pain and depressive symptoms, despite being diagnosed with and treated for both at their assessment. Therefore, insomnia and its causes should be addressed separately from depression and pain in mTBI, in order to maximize treatment outcomes, especially regarding CI.

We also found that a diagnosis of possible/probable malingering was associated with a poorer CI. Although the presence of malingering or symptom exaggeration does not preclude the existence of symptoms or disorders, it does make the quantification of these issues impossible [53, 54]. Therefore, future research should be designed identify the determinants of a malingering diagnosis among injured workers. According to the DSM-IV, a diagnosis of malingering is appropriate when ≥2 of 4 criteria are met [36, 55]: (1) presentation of symptoms in a medico-legal context, (2) discordance between the individual’s stated disability and objective data, (3) uncooperative behavior during evaluation, and (4) presentation of antisocial personality disorder. However, our participants met the first two criteria, as they were referred by the WSIB for evaluation and mTBI is a clinical diagnosis that cannot be confirmed with objective data (i.e., none of our participants had positive MRI findings, and there is currently no sensitive and specific imaging technique to diagnose mTBI) [55, 56]. In addition, a worker with certain personality traits may find being questioned about their disability or injury to be unnerving, which may satisfy the third criterion independent of any malingering. Furthermore, recommendations have been published to implement neuropsychological testing for possible malingering in persons with TBI [57]. This raises the issue of language proficiency [49, 58], as these evaluations would be performed using tests that were developed in English and have not been validated in other languages. Therefore, English proficiency is an important variable to consider in future studies regarding performance validity.

We also found that a longer TSI was associated with a poorer CI after mTBI. Although this association was not evaluated longitudinally, our results may indicate that other relevant factors (e.g., psychological, medical, or other) can develop after the mTBI and strengthen over time, thereby impeding CI [59]. Similarly, psychological and psychosocial factors that present in the early post-injury period can influence outcomes across the entire recovery timeline [60]. Unfortunately, the cross-sectional nature of our study does not allow us to provide insight regarding this topic. Nonetheless, our results support the notion that patient assessments should begin as early as possible, which can establish baseline findings for CI and time-dependent outcomes.

Unexpectedly, being married was not related to a better CI, and our results indicate that marriage had a negative effect on CI. In contrast, earlier studies have reported that married people with a disability have higher levels of life satisfaction [61, 62], fewer handicaps [63], and longer life expectancy [64], compared to their unattached counterparts. However, our results should be interpreted with caution, as one-third of our participants were single, divorced, or widowed at the time of assessment, and we did not assess whether or not this status changed after their injury. Nevertheless, our findings suggest that workers who are married or in relationships may rely on their partners for assistance in performing their family and societal duties, which would hinder their independent CI. This observation highlights uncertainty regarding the role of marriage and/or partner support in the context of compromised CI, and longitudinal analysis of relationship status and patient outcomes (both personal and familial) may provide insight regarding their interdependence and independence.

The strengths of our study include a well-characterized sample of insured workers with a confirmed mTBI diagnosis. In addition, this is the first study to simultaneously address the prevalence of various sociodemographic, clinical, claim-related, and TBI-related variables, and their subsequent impact on CI. Furthermore, we utilized the TRIPOD checklist, which is a valuable reference for good reporting of multivariable prediction models. Therefore, the present study provides foundational data for a comprehensive longitudinal study that can evaluate the risk factors for prolonged recovery and reduced post-mTBI CI.

Our study has several limitations. First, our model for estimating CI was complex, which reflects the large number of covariates that were associated with CI and were not included in our final model. Therefore, a study with a large sample, standardized data collection and calculations will be needed to validate our model. Second, generalization of our findings may be limited, as a high prevalence of clinically relevant disorders was observed in our sample, and each of these disorders have been associated with poor post-injury outcomes. Furthermore, our study only aimed to evaluate workers who had a prolonged recovery after their injury, as they experience the greatest effects of mTBI and are the most difficult to rehabilitate. Third, the R² value of 0.366 for our final regression analysis indicates that that only 36. 6 % of the variance in the mean CI can be explained by the above named variables and insomnia explains most of the variance (i.e., 15.9 %). This finding may be due to the omission of information regarding various critical CI areas, such as psychological sense of community, satisfaction with community, and perception of safety [65‐67]. However, given that our results support the notion of CI as a time-dependent construct, and our data consisted of various time series, the R2 value for insomnia provides solid support of it as a covariate of CI, bearing in mind the fact that we are looking for meaningful associations in the context of delayed recovery from mTBI in the presence of many a priori defined relationships and a relatively small sample size.

Finally, this study highlighted the factors that were associated with CI in a population of workers with mTBI, although the longitudinal relationships between these factors and poor post-injury outcomes remain to be determined. Therefore, further research regarding this topic may facilitate the development of interventions that improve the CI of injured workers with mTBI.