Predicting Length of Care Following Concussion

Abstract

Concussions have become a prominent injury treated across medical care settings. Due to a complicated underlying pathophysiological process, concussions are subtle and produce a wide array of signs, including cognitive, somatic, and sensorimotor symptoms. The diverse symptomatology makes clinical assessment and management of concussions challenging for clinicians. Additionally, this complexity makes determining recovery trajectory after injury difficult for practitioners. Identifying individuals at risk for long durations of treatment early in their care would allow for early restorative interventions, reduced cost of care, and improved patient quality of life. This need lead to a series of studies attempting to add evidence to support the clinical assessment process of individuals following both suspected and diagnosed concussions. Our first study evaluated the rates and reasons for under-reporting of concussions and symptoms in female high school athletes, a neglected population of study in concussion research. We found female athletes often failed to report concussion symptoms to coaches and trainers. Additionally, only 2/3rds of athletes reported receiving concussion education that may explain the lack of reporting. Different clinical assessment techniques were also explored. Balance and reaction time assessment are commonly completed to help identify individuals with suspected concussion, but these assessments are subjective and often require equipment not available outside of clinical settings. We evaluated the validity and reliability of a mobile device application to assess balance and reaction time as part of a concussion screening. We found SwayTM to be comparable to force platform balance assessments and computerized reaction time measurement in healthy subjects, indicating these assessments have clinical utility. Additionally, the assessment only required a mobile device, allowing for assessment in clinics and sports settings without computers or force platform equipment. The last assessment method we evaluated involved neuroimaging techniques. Diffusion tensor imaging (DTI) has been used in the assessment of individuals with moderate to severe brain injury to objectively track recovery of function after trauma. We explored the similar use of DTI to objectively track recovery of function after concussion. We assessed 10 female participants experiencing slowed recovery of function after a concussion. We found no significant associations between DTI metrics for white matter pathway integrity and changes in symptoms over time, indicating that symptom improvement may occur in a non-congruent fashion with underlying neurological functioning. We did find improvements in pathway integrity were associated with improved reaction time and visual processing, adding to the clinical importance of assessing these skills. A lack of understanding as to which objective clinical features leads to increased risk of prolonged care adds to the complexity of determining a person’s recovery trajectory following concussion. This gap in evidence lead to an exploratory study evaluating differences in clinical presentation patterns between groups of individuals with typical treatment durations (≤ 28 days) and extended lengths of care ( 28 days). The exploratory factor analysis yielded different factor structures for each group. Notable differences included age, sex, history of psychiatric disorders, and neurocognitive abilities at the evaluation. These findings suggest clinical differences between individuals with longer and shorter lengths of care are observable at evaluation. After detecting differences between these groups, we sought to clarify the impact of each variable on length of care. A survival analysis was completed, evaluating the effects of clinical, demographic, and social factors on estimating length of care after concussion. We found males had shorter durations of care, while individuals with a history of concussion and better access to housing and transportation had longer durations of care. This study added evidence to support sex and concussion history as influential of longer durations of care. Additionally, while significant, housing and transportation access is likely not a protective feature for concussion recovery, but likely reflects the ability to attend follow-up treatments for concussion care. A final study was conducted to assess the predictive ability for each variable identified in the previous studies, with the primary outcome focused on determining which features best predict whether someone will have a typical duration of care (≤ 28 days) or an extended duration of care ( 28 days). A multivariable, logistic regression model was used to evaluate the significance of each variable. History of previous concussion and slowed reaction times at the evaluation had significant effects. Additionally, individuals from lower socioeconomic areas had lower odds of an extended length of care, although these findings should reflect ability to pursue care, rather than a reduction of recovery time. In summary, these studies add to the foundational knowledge of concussion assessment for diagnostic and recovery purposes. Additionally, these studies contribute to our understanding of factors that contribute to prolonged lengths of care for individuals after concussion. Clinicians should continue to complete thorough evaluation of individuals with suspected concussion, documenting concussion histories, sex, age, reaction times, and external social factors that may prevent or impede necessary follow-up care.