There are increasing interests in the population and in the research about mTBI prevention, diagnosis, and prognosis to improve treatment and clinical decision making (5, 19, 20, 65). Multiple systematic reviews synthesized information about this population and prognostic factors associated with various outcomes. We aimed to present a synthesis of systematic reviews concerning any variable associated with persistent post-concussion symptoms after 3 months. We provided a comprehensive overview of the state of the evidence available concerning the 35 prognostic factors that have been systematically reviewed. Premorbid characteristics, such as the age, the sex, prior concussions and mental/physical problems need to have special attention having stronger evidence that demonstrated a relation with poor outcome 3 months post-injury. Having somatic complaints (e.g. headaches, neck or back pain) after the injury and negative perceptions or expectations related to the recovery were also associated with persistent symptoms following mTBI. However, we could not conclude to a clear prognostic relation regarding physiological characteristics after mTBI such as the presence of specific biomarkers.
Other authors had interests of reviewing, understanding and identifying prognostic or risk factors associated with persistent problems related to a specific condition, such as whiplash injury (66–68), neck pain (69, 70), orthopedic trauma (71), and all TBI (72).
Walton et al. (66) found 9 significant predictors for persistent problems following whiplash injury, such as no postsecondary education, female gender, history of previous neck pain, baseline neck pain intensity greater than 55/100, presence of neck pain at baseline, presence of headache at baseline, catastrophizing, whiplash-associated disorder grade 2 or 3, and no seat belt in use at time of collision. Scholten-Peeters et al. (67) showed that of over 100 different prognostic factors examined, only one (high initial pain intensity) demonstrated a strong evidence association with persisting symptoms after whiplash-associated disorders. Finally, a meta-review of Sarrami et al. (68) presented five associated factors with the prognosis for people with whiplash injury, which were post-injury pain, disability and anxiety, catastrophizing, compensation and legal factors and early use of healthcare. Walton et al. (69) suggested that the prognosis of neck pain of various causes is generally poor and there are relatively few factors that allow high or moderate confidence in their use as predictors of outcome. Clay et al. (71) reported strong evidence supporting the association of female sex, older age, high pain intensity, preinjury anxiety or depression, and fewer years of education with persistent pain outcomes following acute orthopedic trauma. Willemse-van et al. (72) reported in their review that older age, pre-injury unemployment, pre-injury substance abuse, and more severe disability at rehabilitation discharge were strong predictors for long-term disability after TBI. Hence, these conditions seem to have similar prognosis factors for persistent symptoms. Older age, female gender and presence of somatic complaints after the event, such as the intensity of pain can be considered as general prognostic factors to examine from the outset.
Most prognostic factors highlighted in these reviews and in our overview are premorbid characteristics or early on symptoms, which are variables related to the injury that cannot be modified. It is however possible that prioritizing individuals with a higher number of prognosis factors would lead to better outcome in care. We also discuss that interventions could target some perpetuating factors, by reducing life stressors, improving quality of sleep and focusing on realist expectations and helping to manage symptomatic complaints. Although there is no clear evidence on the effectiveness of interventions for reducing persistent post-concussion symptoms (73–76), educational and behavioural interventions seems promising like a multidimensional psychoeducative and counseling intervention (SAAM) based on a biopsychosocial model, addresses misconception and perception of mTBI recovery and postconcussion symptoms through up-to-date psychoeducation, and provides reassurance and counseling about recovery and motivation for change (77). Thus, post-acute recommended interventions tend to match with these objective, such as education regarding normalizing symptoms, expected outcomes and positive recovery, technique to manage stress and gradual return to activities and life roles (78). While commonly used treatment remains educational and reassurance early after the injury, these interventions showed mixed results concerning effectiveness (76, 79). The current evidence tend to suggest psychological and rehabilitative strategies (76), such as cognitive behavioral therapy, cognitive rehabilitation (65), psychotherapeutic interventions, social work interventions, self-management strategies (80) and specialized interdisciplinary rehabilitation (81) for reducing chronic post-concussive symptoms. In addition, future studies should explore whether a patient’s outcome after mTBI can be improved by removing or reducing a prognostic factor. For example, while we know that patients experiencing intense headaches or pain after mTBI are more likely to have poor long-term outcomes, more research are needed to examine the effects of intervention reducing this prognostic factor on persistent post-concussion symptoms outcomes.
It is possible that clinical decision making and organizational interventions can be developed by taking into consideration these prognostic factors. For example, adopting more rigorous criteria to identify patients who would benefit from further treatment, such as identifying those who are at high risk of developing persistent symptoms by having ≥3 symptoms according to the RPQ during the early phase post-injury (74). Rytter et al. (81) suggested focus on effective identification of patients who are at risk of maintaining persistent post-concussive symptoms in order to initiate a treatment plan in a timely fashion. Early and effective identification of risk factors of persistent symptoms may indicate earlier intervention and prevent of such chronicity after mTBI (64).
Evidence have highlighted the importance of early identification of this at-risk population and how to prevent persistent symptoms before onset. However, we need to acknowledge that some individuals may fall through the cracks of the system and they can experience a prolonged recovery (more than 3 months) (6, 7). Once these patients show signs of chronic symptomatology, prognostic factors are therefore more than important to consider, as time is running out. Authors are unanimous that patients experiencing persistent consequences following mTBI must receive healthcare services in a timely manner (12, 81). Referral to a specialized multidisciplinary mTBI clinic can be appropriate for patients with persistent symptoms that do not respond to treatment in a primary care setting (82). Access to such specialized rehabilitation services can be complex and patients often face long waiting time (83–85). Results of this OvR could help multiple stakeholders, such as clinicians and healthcare managers, to understand the prognosis of their patients and to focus their time and resources on patients needing the most. It could also inform decision-makers and policymakers about the challenge of early identification of prognostic factors in order to prevent onset of persistent symptoms.
It is known that overviews of reviews often lack methodological rigor because there are no pre-established reporting guidelines (86). However, we have based our methods on Cochrane Handbook (29) and several previous works (30–35, 87) that can appropriately guide us through a rigorous process. We must acknowledge that our review has some limitations. The first is that, by the nature of an overview, we limited our data analysis to what was presented in the systematic reviews included. Thus, results presented in our OvSR had already been synthesized by previous authors, so prognostic factors may have been omitted as we did not analyze the primary articles. Our intent was to have a broad picture of all prognostic factors reviewed in the literature and we are aware that our results should be interpret with caution. Even if we showed a very small CCA value, we did not conduct further analysis regarding the primary studies in each review, so two reviews may have analyzed data from the same studies. While most of the included systematic reviews targeted only adult population with mTBI, some have examined a broader population, which made it more difficult to extract the relevant data in some systematic reviews. We chose to include all adult population with mTBI (veterans, military personnel and adults with sport-related injury) to be in adequation with our objective, but it may have led to complicate the data synthesis of prognostic factors into a single comprehensive model. We also need to mention that few systematic reviews presented results of prognostic factors that did not demonstrate association with outcomes, which shows several publication biases.