*4.2. Patient Characteristics and Direct Pathway*

The study population is similar to the western TBI population in terms of the proportion of males (69%), age (mean 57 years) and the dominant trauma mechanism being falls (56%) [27,28]. Severe somatic comorbidity was found in 29% (ASA-PS 3), which is higher than in the CENTER-TBI case-mix study, which also included patients with concussions (11%) [28]. This can likely be explained by a somewhat higher mean age in our study population and no exclusion criteria based on preinjury disease. This is also reflected by the 13% of patients living with assistance at home or at a nursing home at the time of injury. We found preinjury substance dependence in 17% of patients. However, patients were not systematically assessed for substance dependence; thus, there is reason to believe the actual number is somewhat higher. By comparison, a previous study from OUH found that 26% of TBI patients had significant preinjury substance dependence (mainly alcohol) [29].

Half of the patient population was admitted with complicated mild TBI, and one-third was discharged to home. In line with previous research, patients with moderate–severe TBI dominated in the direct pathway group [11,12]. For patients with severe TBI, 39% followed the direct pathway, which was slightly lower than that in previous European studies (40–48%) [8,12,14,30]. Similar to the study by de Koning et al. [30], 20% of moderate TBI followed the direct pathway. The results from the CENTER-TBI study demonstrated different and complex care pathways in the first six months after injury, particularly for patients with severe TBI [7]. Furthermore, rehabilitation needs were reported in 90% of patients with moderate–severe TBI in the first six months after injury [31]. Our data are limited to acute treatment at a Level 1 trauma hospital, and it is likely that several patients were referred to specialized rehabilitation at a later point. However, there are different organizations and expertise in rehabilitation between local hospitals and municipalities within the health region. Thus, it is reasonable to assume that the probability of unmet rehabilitation needs increases when the direct pathway is broken.

We found a positive and significant trend for utility of the direct pathway during the five years–a result of emphasized focus on TBI rehabilitation during the past decade [3,17]. However, the situation is fragile, and it is worth mentioning that neurorehabilitation beds were periodically converted to manage the impact of COVID-19 at OUH. However, this study did not include patients injured in the 2020–2021 pandemic years. Results on the access to rehabilitation in the pandemic years will be published in a subsequent paper.

#### *4.3. Factors Predicting the Direct Pathway*

The probability of following the direct pathway increased for patients with more severe TBI, younger age, and decreased functional level at discharge as measured by the GOS. These findings are in line with previous research [13,32]. The statistical models demonstrated the striking impact of age, which is discussed in a separate paragraph. Neurosurgical procedures were significant predictors in univariate regression but not in multivariate models, which is inconsistent with the study by Jacob et al. [11]. In fact, 42% of patients following the direct pathway were treated without surgical procedures. Severe disability (GOS) was the strongest clinical positive predictor in both models, suggesting that patients are clinically assessed and prioritized for the direct pathway. In this study, TBI severity was categorized by HISS, which is mainly based on the GCS score in the acute phase. The GCS score is essential because it partly guides acute treatment [22,33,34]. The GCS score, as a measure of TBI severity, is widely used in both clinical settings and research and correlates with outcome at the group level, but it should not be used as a single injury severity predictor of TBI outcome [34]. TBI is a complex condition with substantial individual variation in outcomes. TBI can be life-threatening in the acute phase, e.g., in cases with epidural hematoma, where the patient may have rapid and good recovery if treated with immediate neurosurgery and removal of hematoma. Likewise, patients with moderate or mild TBI in the acute phase may experience long-term disability. A recent proposal in assessing the severity of TBI suggests changing from severity labels to risk assessment over time [35]. Reduced functional levels at six months are reported for patients

with TBI admitted to the hospital [28]. The study identifying unmet rehabilitation needs emphasized the necessity of a more extensive and standardized assessment of functional impairments and corresponding rehabilitation needs [31]. Currently, there is no systematic assessment of rehabilitation needs at discharge from acute care units at OUH. Moreover, the decision for referral transfer to rehabilitation is not solely based on the clinical condition of patients. It can also be affected by pressure to free beds at the Level 1 trauma center, low capacity at early rehabilitation units, and professionals' knowledge about expected benefit from rehabilitation and long-term disabilities associated with TBI [12,36,37].

We found no support for the notion that patients with preinjury substance dependence are downgraded for the direct pathway, which was in contrast to findings by Jourdan et al. [12] but in line with other studies [11,13,14]. In our study, access to sociodemographic variables was limited. However, we do not believe that such variables have had an impact on the direct pathway in this study; this assumption is based on results regarding preinjury substance dependence. Nonetheless, variables such as marital status, education and employment are expected to be of importance at later stages of TBI when follow-up is more fragmented.

#### *4.4. The Impact of Age*

Age was an important explanatory variable for the direct pathway. The probability of management through the direct pathway decreased significantly with higher age, a situation not unique to our study population. In a systematic review [13], the only consistent negative predictor for discharge to specialized rehabilitation was increasing age. Preinjury comorbidity and functional impairments were highly associated with age in our study population [19]. We found these factors to have a significant negative impact on the probability of treatment in the direct pathway, similar to the literature on stroke [15,16]. Previous studies show that younger patients to a greater extent follow a direct pathway or are discharged to home, while older patients are more often discharged to general rehabilitation and rarely directly to home [13,14,38,39]. Clinical trials typically include patients aged 18–65 years and often have exclusion criteria based on comorbidities [28,40,41]. Older patients receive less aggressive therapy in the acute phase (medical and surgical) [42,43]; presumably, this may lead to directions for further treatment. Nonetheless, there is evidence that older patients with TBI may benefit from intensive inpatient rehabilitation [41]. It is clear from our results that young adults and patients of working age are prioritized for the direct pathway. Given the risk of life-long negative consequences of TBI, one can argue that it is appropriate to prioritize these patients. However, life expectancy is increasing, and in society, we observe many persons >64 years living an active life with social roles that include responsibilities with indirect socioeconomic impact (e.g., voluntary work, family obligations across generations). Moreover, it will presumably be socioeconomically beneficial if older patients regain preinjury functional levels and are able to live independently at home.
