Work-related and non-work-related mild traumatic brain injury: Associations with mental health and substance use challenges in a Canadian population-level survey

1 Introduction

An estimated 4.6 million individuals sustain a traumatic brain injury (TBI) in the United States and Canada each year [1]. The somatic, cognitive, and behavioural symptoms associated with even mild TBI (mTBI) can be substantial, persisting beyond 6 months and being classified as post-concussion syndrome (PCS) in between 11% and 38% of patients, depending on the classification system used [2–6]. Symptoms of PCS have adverse consequences for overall health, employment, and social and family life [3]. It is believed that 70 to 90% of all brain injuries are considered mild [7]. The incidence of hospital-treated mTBI is estimated to range from 100 to 300 per 100,000 globally, but because many mTBIs are not treated at hospital [8, 9] and because mild cases are often missed or misdiagnosed by primary care physicians, the true incidence is suspected to exceed 600 per 100,000 [7, 11]. Population-level surveys capture cases of TBI that go unreported in healthcare databases [7, 11]; in Ontario, Canada, estimated incidence of past-year TBI in the student and adult populations ranged between 3% and 6% [12, 13] with the prevalence of lifetime TBI ranging between 16% and 20% [12, 13].

TBIs can occur in many situational contexts, but those that occur in the workplace represent an important subgroup of injury, and have been suggested to have “poorer vocational, psychological, neurological, functional, psychosocial and general health outcomes” [14, p.2]. Although there has been an overall decline in claims for work-related injury in Canada, the proportion of these injuries involving TBI has increased [15], indicating a need to better understand the correlates and sequelae of work-related (wr) TBI. A recent systematic review identified just 18 studies in a 6-year period that focused specifically on wrTBI [15]. As with TBI in the general population, not all wrTBI are reported to employers and medical care is not always sought; thus, prevalence of wrTBI is poorly understood. Although likely to be an underestimate of the true incidence of wr-mTBI, current evidence suggests that 1 in 4 mTBIs occurs in the workplace [16]. The majority of individuals who experience mTBI return to work within 3 to 6 months, with most estimates suggesting that only about 20% are unable to return to work within one year of their injury [17–19]. Return to work is associated with increased quality of life post-injury [19, 20].

Mental health (MH) and substance use (SU) challenges have been identified as correlates, if not sequelae of TBI [21, 22]. Recent TBI has been associated with increased risk of depressive and anxiety disorders [23–27] and post-traumatic stress disorder [27, 28]. Lifetime experience of TBI is associated with past-year substance use problems (including problem gambling), MH challenges, and aggression [13, 29–31], which are in turn associated with negative outcomes such as injury, increased suicidality, and chronic health problems such as heart and liver disease (e.g., [32–35]). Multiple studies have identified an association between MH challenges and worse TBI symptom severity, disability, and quality of life post-injury [36–39], and there is evidence to suggest that in at least some patients, MH challenges are a precursor to functional impairments [40]. Comorbidity of TBI with hazardous drinking behaviour among adults is associated with higher likelihood of roadway aggression [21] compared to those experiencing either TBI or hazardous drinking alone. Among adolescents, however, this same comorbidity is associated with higher likelihood of psychological distress, suicide contemplation and attempts, self-rated fair/poor MH, use of medication for depression or anxiety, and a host of conduct behavioural issues (e.g., damaging property, stealing more than $50, beating up or hurting someone) [41].

Research examining the association between MH and SU challenges and brain injury among individuals with a wr-mTBI is limited, particularly at a population-level. A comparison of wrTBI and non-wrTBI based on a review of the Ontario Trauma Registry revealed that individuals with a wrTBI were less likely to have a history of psychiatric disorder pre-injury [42]. Yet, based on an outpatient clinic sample studied by Terry et al. [43], patients with wr-mTBI were more likely to experience post-traumatic stress symptoms than patients who had experienced a similar injury outside of the workplace.

The current study determined the prevalence of self-reported lifetime wr-mTBI and non-wr-mTBI from a representative sample of adults and assessed the association between these injuries and experience of MH and SU challenges. Consistent with existing data [21, 22], we hypothesized that individuals with wr-mTBI and non-wr-mTBI would be at greater risk of MH and SU challenges than individuals without TBI history.

2 Methods

Data from the 2019 Centre for Addiction and Mental Health (CAMH) Monitor cross-sectional telephone (landline and cellphone) survey of Ontario adults aged≥18 years (N = 1792; 28% response rate, comparable to national surveys) were analyzed. The 2019 survey data were an accumulation of rolling samples, collected quarterly with independent samples of approximately 750 completions each (see [44] for more details on the CAMH Monitor). Participants were asked if they had experienced a head or neck injury that resulted in headache, neck pain, dizziness, blurred or double vision, vomiting, feeling confused or ‘dazed’, or resulted in problems remembering, being knocked out or losing consciousness. If so, participants were asked a similarly worded question focused on identifying mild injury, which this time specified loss of consciousness as lasting ‘up to 30 minutes’. Those who reported experiencing a mild injury were asked if it occurred while working for income. Based on a complex sample design with six strata (region) and final post-stratification adjustment to restore the age by gender distribution reported by the most recently available census figures, Taylor series linearization available in SAS 9.4 was used to test four binary logistic regression models, treating missing data listwise. These models adjusted for sex, age, marital status, education, and income. A score of≥8 on the Alcohol Use Disorders Identification Test (AUDIT; [45]) indicated hazardous use of alcohol. A score of≥4 on the cannabis subscale of the Alcohol, Smoking and Substance Involvement Screening Test (ASSIST; [46]) indicated moderate to high risk of problematic cannabis use. A score of≥8 on the Kessler Psychological Distress Scale (K6; [47]) indicated moderate to serious distress. Finally, an item from the Centers for Disease Control’s Health-Related Quality of Life Measures [48] asked participants to rate their MH (binary scoring: excellent/very good/good versus fair/poor). The STROBE cross-sectional reporting guidelines were used to prepare this report [49].

3 Results

Lifetime prevalence of wr-mTBI and non-wr-mTBI was 7.3% and 24.9%, respectively. wr-mTBI and non-wr-mTBI were more prevalent among males compared to females (9.3% vs. 5.5% and 30.1% vs. 20.2%, respectively). wr-mTBI was more prevalent among 35–54 year-olds (9.2%), whereas non-wr-mTBI was more prevalent among 18–34 year-olds (31.0%). Non-wr-mTBI was more prevalent among those respondents who reported hazardous drinking (42.8% vs. 22.5%) and higher risk of problematic cannabis use (37.3% vs. 22.8%). Both wr-mTBI and non-wr-mTBI were more prevalent among those reporting fair or poor MH compared to those reporting good to excellent MH (10.2% vs. 6.9% and 33.1% vs. 23.8%, respectively). Likewise, both wr-mTBI and non-wr-mTBI were more prevalent among those reporting moderate to serious risk of psychological distress compared to those reporting low risk (12.2% vs. 6.2% and 31.1% vs. 23.5%, respectively). See results of chi-square analyses in Table 1.

Adjusting for sociodemographic variables, non-wr-mTBI was significantly associated with all MH and SU variables (see regression analyses in Table 2), demonstrating increased odds of hazardous alcohol (AOR = 2.12, 95% CI = 1.41, 3.19) and cannabis (AOR = 1.61, 95% CI = 1.05, 2.45) use, psychological distress (AOR =  1.68, 95% CI = 1.14, 2.49), and fair/poor MH (AOR = 1.70, 95% CI = 1.11, 2.59). Lifetime wr-mTBI was significantly associated with MH variables only, demonstrating increased odds of reporting psychological distress (AOR = 3.40, 95% CI = 1.93, 5.97) and fair/poor MH (AOR = 2.16, 95% CI = 1.12, 4.19). Notably, for each outcome where wr-mTBI was significantly associated, its adjusted odds ratio was nearly double the odds ratio of non-wr-mTBI.

Although the prevalence of wr-mTBI and non-wr-mTBI was similar, MH and SU correlates for these two types of injury diverged. Whereas SU outcomes were significantly associated with non-wr-mTBI only, MH outcomes were significantly associated with both wr-mTBI and non-wr-mTBI. Moreover, based on size of the odds ratios, MH outcomes were more strongly associated with wr-mTBI than non-wr-mTBI. This finding is consistent with the findings of Terry et al. [43] who reported increased risk of post-traumatic stress symptoms among outpatients with experience of wr-mTBI relative to non-wr-mTBI; authors speculated that this may have been the result of differences in the mechanism of injury (e.g., fewer sports injuries and more traumatic injury circumstances among cases of wrTBI).

Recognizing the correlational nature of the data, the opposite direction of effect could be argued, placing individuals with pre-existing MH issues at greater risk of both wr-mTBI and non-wr-mTBI. By extension, individuals with pre-existing SU issues may be less likely to be employed in positions where a wr-mTBI is likely to occur, such as construction or transportation [15]. However, a review of the Ontario Trauma Registry indicated that individuals with wrTBI were less likely than those with non-wrTBI to have a history of MH issues prior to their injury [42].

Eligibility for financial compensation, which in many jurisdictions (including Ontario) applies primarily to workplace injuries, has been associated with slower recovery and return to work, possibly due to potential financial gain or increased stress resulting from challenges in pursuing compensation. However, even in jurisdictions where non-workplace injuries are eligible for compensation, wrTBI is associated with poorer outcomes than non-wrTBI [14]. Employers may be less willing to accommodate a worker’s injury, and fellow employees may be less empathetic, if the injury occurred in the workplace. Likewise, the worker may experience more MH challenges and be less likely to return to work if their injury occurred in the workplace [14]. Further research exploring causes for adverse MH outcomes following wrTBI is needed.