A hybrid assessment of clinical mobility test items for evaluating individuals with mild traumatic brain injury

Abstract

Background and Purpose:

The Functional Gait Assessment (FGA) and High Level Mobility Assessment Tool (HiMAT) are clinical batteries used to assess people with mild traumatic brain injury (mTBI). However, neither assessment was specifically developed for people with mTBI; the FGA was developed to evaluate vestibular deficits, and the HiMAT was developed for individuals with more severe TBI. To maximize the sensitivity and reduce the time burden of these assessments, the purpose of this study was to determine the combination of FGA and HiMAT items that best discriminates persons with persistent symptoms from mTBI from healthy controls.

Methods:

Fifty-three symptomatic civilians with persistent symptoms from mTBI (21% male, age 31(9.5) years, 328 (267) days since concussion and 57 healthy adults (28% male, age 32(9.6) years) participated across three sites. The FGA and HiMAT were evaluated sequentially as part of a larger study. To determine the best combination of items, a lasso-based generalized linear model (glm) was fit to all data.

Results:

The area under the curve (AUC) for FGA and HiMAT total scores were 0.68 and 0.66, respectively. Lasso regression selected four items including FGA Gait with Horizontal Head Turns and with Pivot Turn, and HiMAT Fast Forward and Backward Walk, and yielded an AUC (95% CI) of 0.71 (0.61, 0.79) using standard scoring.

Discussion and Conclusions:

The results provide initial evidence supporting a reduced, hybrid assessment of mobility (HAM-4-mTBI) for monitoring individuals with mTBI. Future work should validate the HAM-4-mTBI and investigate its utility for tracking progression throughout rehabilitation.

INTRODUCTION

Balance and gait problems that persists for weeks-to-months are common problems after mild traumatic brain injury (mTBI)^1–4 that can be rehabilitated with physical therapy and targeted interventions.^5,6 While numerous clinical batteries have been used with mTBI populations to identify and monitor changes in functional mobility, few were designed specifically for evaluation in the mTBI population. For example, the Functional Gait Assessment (FGA) and High Level Mobility Assessment Tool (HiMAT) are both valid and reliable clinical batteries^7–9 that evaluate functional mobility, and the FGA is specifically recommended for use with mTBI.¹⁰ But, neither was explicitly designed for the heterogeneous mTBI population. The FGA was developed to evaluate gait and balance in individuals with vestibular dysfunction, as a modification of the Dynamic Gait Index, which was initially developed to assess the likelihood of falling in older adults by assessing various components of gait.¹¹ The HiMAT⁸ and revised HiMAT¹² were developed to test high-level mobility deficits and readiness for return to sport in individuals after a moderate to severe TBI and is the only outpatient assessment for TBI that is highly recommended by the Academy of Neurologic Physical Therapy TBI Evidence Database to Guide Effectiveness (EDGE) task force. The process for the development of the HiMAT used both critical evaluation of existing mobility scales and expert opinions from physical therapists using consensus meetings and rounds of questionnaires.^13,14 Both assessments were developed with a focus on functional requirements needed to successfully complete movements or activities rather than on the mechanistic or neural components underlying the specific deficits or disease processes. While both batteries are clinically useful in moderate-to-severe TBI, neither battery may be ideal for the moderately high-functioning population of individuals with mTBI and persistent symptoms. Given their development for assessment of individuals with other underlying pathologies, the FGA and HiMAT may also have ceiling and/or floor effects in people with mTBI.

One possible approach to patients with mTBI and persistent symptoms is to complete both FGA and HiMAT batteries, but time constraints on the patient and provider often make this solution impractical. Time constraints in clinical settings impact treatment sessions and adherence to clinical guidelines.¹⁵ Even the perception of a time constraint can effect clinical decision-making,¹⁶ and these constraints, perceived or real, can have a negative effect for both the clinician and the patient.¹⁷ Thus, there is a push to shorten clinical assessments to minimize the number of items and concerns about time.^12,18 Individuals with persistent symptoms from mTBI may report intolerance to prolonged evaluations due to fatigue and symptom exacerbation. Performing tests that adequately challenge the patient to determine the full extent of mobility deficits, while also considering suitable symptom load during evaluation, is important for patient management after mTBI.

The FGA and the HiMAT both have items that may be useful in the management of individuals with mTBI and persistent symptoms. However, these tools are also likely to have items not pertinent for assessment of this population. Hence, the purpose of this study was to determine the combination of FGA and HiMAT test items that best discriminate individuals with persistent, symptomatic mTBI from healthy controls. This targeted hybrid battery of select FGA and HiMAT items may 1) increase the sensitivity and specificity of the assessment 2) save time in the clinic, and 3) potentially minimize symptom load during evaluation sessions, which may improve tolerance and compliance. While the ultimate goal is to generate a quick, hybrid battery to track a patient’s progression over time after mTBI, this first step - determining whether a hybrid assessment can successfully differentiate individuals with mTBI and persistent symptoms (e.g., >3 weeks post-injury) from healthy individuals – serves as a critical starting point.

METHODS

Participants:

A convenience sample of 53 individuals with persistent symptoms following mTBI (11 male/42 female; mean age: 32.0 (SD 9.6) years; 328 (SD 267) days since injury, symptoms lasting >3 weeks) and 57 healthy control volunteers (16 male/41 female; mean age 31.1 (SD 9.5) years) were recruited and tested across three separate sites (Oregon Health & Science University, Portland, OR; Courage Kenny Research Center, Minneapolis, MN; and University of Utah, Salt Lake City, UT) between June 2019 and September 2020 as part of a larger study.¹⁹ Participants were recruited from the general population of the hospitals and local communities, and through concussion / TBI clinics or medical records of individuals with a mTBI diagnosis. Inclusion criteria were: (1) have a diagnosis of mTBI, (2) be between 18 and 50 years of age, and (3) be outside of the acute stage (>3 weeks post injury) but within 3 years of their most recent mTBI and still reporting symptoms. Control participants either had no history of mTBI or were more than 7 years removed from their most recent mTBI and had no reported residual symptoms. Potential participants were excluded if they: (1) had a history of any other injury, medical, or neurological illness that could potentially impair their balance (i.e., stroke, lower extremity injury, recent surgery), (2) had a current substance abuse disorder, (3) were pregnant, or (4) were unable to communicate in English. In order to reduce the chances of the participants overstating symptoms, participants were not provided with results or individual data from the study. Participants were provided a gift card in appreciation for their time and effort. The study was conducted in accordance with the Declaration of Helsinki (1964), and approved by the Institutional Review Boards at each of the sites. Informed consent was obtained prior to participation.

Procedure

All participants were tested on a single occasion in a clinical laboratory space or quiet hallway adjacent to the laboratory. Demographic information, exercise screening, symptom reports via the Neurobehavioral Symptom Inventory (NSI) and the Dizziness Handicap Inventory (DHI) and medical history (i.e., injury mechanism and any clinical diagnosis of comorbidities such as anxiety) were collected via self-report. A specific evaluation for peripheral vestibular disorders was not completed as part of the testing for this study. Participants completed the FGA and the revised HiMAT per published instructions.^8,9,12 The FGA is a 10-item gait assessment that uses both simple and complex tasks to assess gait and balance in individuals with vestibular disorders.⁹ Each item is scored on a 0–3 ordinal scale for a total of 30 points, with a higher score indicating better performance (i.e., 30/30 = best possible score). The revised HiMAT¹² is an 8-item test designed to assess high-level mobility performance in individuals with balance and mobility deficits secondary to TBI. Each item of the revised HiMAT is scored on a 0–4 ordinal scale primarily based on speed of completion time for a total of 32 points, with a higher score indicating better performance (i.e., 32/32 = best possible score).

Data Analysis:

To determine the best combination of individual items from the FGA and HiMAT for predicting group, a lasso-based generalized linear model (glm) was fit to all data. A binomial distribution was used for the response variable, mTBI / no mTBI, and each individual item’s score from the FGA and HiMAT were included as predictor variables. A 10-fold cross-validation was used to minimize the model deviance and tune the regularization parameter λ.

Individual items with a non-zero beta coefficient from the lasso glm were retained and used to create the Hybrid Assessment of Mobility for mTBI (HAM-4-mTBI). Two different HAM-4-mTBI scores were investigated: standard scoring and a coefficient-weighted scoring. The standard scoring used the sum of each item using routine scoring (0–3 points for FGA items, 0–4 points for HiMAT items); the coefficient-weighted scoring multiplied each item’s scoring by the coefficient from the lasso glm (i.e., equivalent to the predicted y-value from the glm) prior to summing each item.

The areas under the receiver-operator characteristic curves (AUCs) were calculated for each scoring method and 95% confidence intervals (CIs) were obtained using bootstrapping with 1,000 iterations. The AUCs and 95% CIs were also calculated for the FGA total score and the HiMAT total score. Finally, the optimal cutoff point was determined for each battery and sensitivity and specificity were determined at these points.²⁰ All analyses were conducted in MATLAB Version R2020a (The Mathworks, Natick, MA, USA).

RESULTS

Description of sample

This sample is part of a larger study – the subjects included here are identical to the subjects included in Parrington et al.²¹ Our healthy controls were matched demographically for age, and sex (Table 1). Twelve percent of our healthy control participants sustained a previous mTBI a median 13 (min 4, max 17) years prior to testing. Total scores on the FGA and HiMAT are provided in Table 1, with individual scores for each item provided in the online supplement. To assist in characterizing our sample, summary symptom report scores are also found in Table 1.

Table 1.

Demographic characteristics for each group reported as mean (SD) unless otherwise noted.

		mTBI	Control
Demographics	N (% male)	53 (21%)	57 (28%)
	Age (years)	32.0 (9.6)	31.1 (9.5)
	Height (m)	1.7 (0.3)	1.7 (0.1)
	Mass (kg)	72.1 (21.6)	72.4 (17.1)
	Days since concussion a,b	261 [21–989]	4564 [1441–6105]
Comorbidities c	Anxiety	9 (17%)	2 (4%)
	Depression	6 (11%)	2 (4%)
	Attention	2 (4%)	2 (4%)
	Learning disability	2 (4%)	1 (2%)
	Post-traumatic stress	1 (2%)	0 (0%)
Self-reported symptoms a	Total DHI (max 100) a	14 [0–58]	0 [0–10]
	N (%) with DHI > 0	45 (85%)	10 (18%)
	Total NSI (max 88)	27.68 (15.07)	5.07 (4.28)
	Somatosensory Subscore (max 28)	7.6 (4.5)	1.1 (1.5)
	Affective Subscore (max 24)	9.1 (5.4)	2.1 (1.9)
	Cognitive Subscore (max 16)	6.5 (4.1)	1.2 (1.4)
	Vestibular Subscore (max 12)	3.0 (2.4)	0.5 (0.7)
Mobility Function	FGA Total (max 30)	27.4 (2.1)	28.6 (1.5)
	N (%) with FGA = 30	7 (13%)	23 (40%)
	HiMAT Total (max 32)	23.8 (5.3)	26.5 (4.1)
	N (%) with HiMAT = 32	2 (4%)	6 (11%)

^a.Reported as median [min-max];

^b.7 of 57 healthy controls with remote history of concussion;

^cReported as n (percentage);

DHI, Dizziness Handicap Inventory; NSI, Neurobehavioral Symptom Inventory; FGA, Functional Gait Assessment; HiMAT, revised High Level Mobility Assessment Tool; mTBI, mild traumatic brain injury

Note: This sample is part of a larger study¹⁹ – the subjects included here are identical to the subjects included in Parrington et al.²¹

Selected test items

The lasso glm results are presented in Table 2. Four items were retained with non-zero coefficients: FGA Gait with Horizontal Head Turns, FGA Gait with Pivot Turn, HiMAT Fast Forward Walking, and HiMAT Fast Backward Walking. The AUCs (95% CI) for FGA total score and HiMAT total scores were 0.68 (0.57, 0.77) and 0.66 (0.52, 0.73), respectively. The HAM-4-mTBI achieved AUCs of 0.71 (0.61, 0.79) using standard scoring and 0.73 (0.62, 0.80) using coefficient-weighted scoring (Figure 1). The maximum possible score on the HAM-4-mTBI was 14 points, and a cutpoint of 12.5 points yielded optimal sensitivity and specificity using standard scoring; a score less than or equal to 12 points to denote abnormal function resulted in 68% sensitivity and 60% specificity. Conversely, sensitivity and specificity were 89% and 32%, respectively, when using a score of 11 or less to indicate abnormal function. Optimal sensitivity and specificity were 60% and 58%, respectively, for the FGA (cutpoint of 28.5), and 58% and 67%, respectively, for the HiMAT (cutpoint of 25.5). A total of 6 participants with mTBI (11%) and 18 healthy controls (32%) achieved the maximal 14 points on the HAM-4-mTBI, indicating a potential for small ceiling effects comparable to those seen in the standard FGA for people with persistent symptoms from mTBI.

Table 2.

Beta coefficients from Lasso generalized linear model and counts of each score within mTBI and healthy control groups for the retained individual items

			Clinical Score
	Beta Coefficient from Lasso GLM		0	1	2	3	4
FGA: Gait with Horizontal Head Turns	−0.602	mTBI: Control:	0 0	1 1	18 4	34 52	N/A
FGA: Gait with Pivot Turn	−0.008	mTBI: Control:	0 0	1 0	4 1	48 56	N/A
HiMAT: Fast Walk Forward	−0.024	mTBI: Control:	0 0	0 0	10 4	29 28	14 25
HiMAT: Fast Walk Backward	−0.397	mTBI: Control:	0 0	0 0	10 3	26 21	17 33

FGA, Functional Gait Assessment; HiMAT, revised High Level Mobility Assessment Tool; mTBI, mild traumatic brain injury

Figure 1.

Receiver-operator characteristic (ROC) curves (solid lines) and 95% confidence intervals (CI; shaded bands) for the revised High Level Mobility Assessment Tool (HiMAT; blue), Functional Gait Assessment (FGA; green), and 4-Item Hybrid Assessment of Mobility for mTBI (HAM-4-mTBI; Red). The areas under the ROC curves (AUCs) are also included.

DISCUSSION

Our study identified a subset of four items that, when combined, differentiated individuals with persistent symptoms from mTBI from healthy controls as well as the full HiMAT or FGA batteries. This subset of items included: 1) Gait with Horizontal Head Turns, 2) Fast Backward Walking, 3) Fast Forward Walking, and 4) Gait with Pivot Turn. Taking into consideration the wide confidence intervals for the AUCs – the HAM-4-mTBI performed similarly to the individual HiMAT or FGA assessments – however, the pertinent clinical consideration is a reduction in clinical assessment time.

That the HAM-4-mTBI performed similarly to the FGA and HiMAT means the clinical assessment time can be meaningfully reduced to approximately 40% of either the FGA or HiMAT. With decreasing reimbursement from insurance companies and growing, unsustainable growth in health care expenditure in the United States,²² long evaluation and treatment times are becoming increasingly rare in physical therapy clinics. When asked about reasons for not using standardized outcome measures in their practice, physical therapists overwhelmingly reported they take too much time for the clinician and patient.²³ Utilizing outcome measures that require the patient to perform unnecessary and time-consuming tasks, ultimately increasing their overall symptom and cognitive load.²⁴ These preventable time constraints have then been shown to decrease the overall quality of patient care due to, for one, lack of adherence to clinical guidelines.¹⁵ Our results provide the first step towards developing a simplified outcome measure, the HAM-4-mTBI, that can be performed in a short period of time, minimize the symptom load of the patient, and provide the clinician with data that can potentially help inform diagnosis and management, and measure change over time. Subsequent steps will involve validation in a separate cohort, tracking the sensitivity of the HAM-4-mTBI to patient progression over rehabilitation, and examining whether this brief hybrid assessment generalizes to more populations with acute mTBI or to individuals with persistent symptoms after mTBI who are more symptomatic than our population.

While specific neurophysiologic mechanisms are not a part of this study, the present results agree with the notion that mTBI affects visual and vestibular function, speed, coordination and whole-body stability. Further, our results identified minimal, if any, difference between the HAM-4-mTBI scoring methods, meaning the traditional scoring methods that are well-known by physical therapists can be potentially used without issue.

1). Gait with Horizontal Head Turns

It is unsurprising that Gait with Horizontal Head Turns was retained in the HAM-4-mTBI given its persistent presence in a variety of clinical batteries including the DGI,²⁵ FGA,⁹ Balance Evaluation Systems Test (BEST),²⁶ and mini-BEST.¹⁸ Walking with horizontal head turns is a common task used by physical therapists to detect vestibular dysfunction and/or abnormal integration of vestibular inputs during walking.^27,28 In the FGA, walking with horizontal head turns is scored based on one’s ability to maintain balance and travel in a straight line while voluntary head turns alter visual and vestibular information. Given the predominating dysfunctional sensory integration in people with persistent symptoms after mTBI,^29,30 this task is well-suited for directly examining the ability to maintain balance while forward walking with self-induced vestibular challenges in people with mTBI.

2). Fast Backward Walking

Backward walking is a component of both the FGA and HiMAT, signifying its clinical importance for assessing individuals with fall risk or vestibular and balance disorders. Nonetheless, it is not commonly considered in the assessment of mTBI. The sole available study that tested backward walking in individuals with mTBI (i.e., concussion) examined backward tandem gait with eyes open and eyes closed in 14 to 18 year old adolescents with sport related concussion.³¹ Using scoring based on balance control and instability, the authors found that backwards walking with eyes closed was the most sensitive component for identifying the adolescents with sport related concussion (81%). Here, the backwards walking from the HiMAT was retained but not the backwards walking from the FGA, and thus scoring was based simply on speed rather than on balance control and instability. The “as fast as possible” component of the HiMAT may provide the additional challenge that aided in differentiating case and control participants.

Brief backward walking requires increased cognitive demands and postural control³² as compared to forward walking. In certain diagnoses such as breast cancer,³³ Parkinson’s Disease,³⁴ and dementia or aging,³⁵ decreased backward walking speed and shorter steps or other spatiotemporal gait parameters may be related to decreased muscular strength and poorer postural stability. We do not typically associate mTBI with loss of muscle strength, but slower speed and shorter steps are often considered to be indicative of a more cautious gait strategy that is commonly observed after sport-related concussion or mTBI.^36–39 The retention of backwards walking in the HAM-4-mTBI suggests that walking in reverse coupled with different sensory information (visual flow away from the body) may be sensitive to detecting impaired neuromuscular coordination, particularly for new tasks. Previous work suggested backwards walking can unmask walking deficits that are not detected during typical forward walking,⁴⁰ which can be important in individuals with persistent symptoms following an mTBI where deficits are often more subtle. While few studies have examined backwards walking in people with mTBI, the use of backward walking speed to identify mobility deficits is gaining popularity.³²,⁴⁰,⁴¹ Future work should consider more detailed investigations of backwards walking in people with mTBI.

3). Fast Forward Walking

In addition to Fast Backward Walking from the HiMAT, Fast Forward Walking from the HiMAT was also retained in the HAM-4-mTBI. The deleterious effects of mTBI on self-selected gait speed are well described,³⁸ but deficits are most readily apparent during dual-task gait or complex gait. Fast walking, in contrast to walking at a self-selected pace, may similarly represent added complexity that stresses intersegmental coordination, cognitive processing efficiency, muscle power, and balance.⁴² The importance of fast walking is also supported by previous work in individuals with acute and chronic, mild to severe TBI.⁴³ Further, fast walking was a foundation for constructing the HiMAT; higher self-selected and fast walking speeds are strongly associated with increased functional ability following brain injury.⁴⁴ The ability, or willingness, to increase speeds is associated with muscle power and balance in individuals who are months to years removed from a moderate-to-severe TBI.⁴⁵ Slower fast walking speeds may also be indicative of symptomology; given that the acceleration of the head oscillates at larger amplitudes with faster walking speeds.⁴⁶ Therefore, it is possible that slower, regulated ‘fast as possible’ walking speeds do not represent a neuromechanical deficit in function, but a behavioral decision to limit speed based on head acceleration and possible symptom exacerbation. Similar to results with turning speed,⁴⁷ individuals with mTBI may choose to regulate their speed to avoid the larger amplitudes of head acceleration despite being prompted to walk fast. We suspect the multiple factors that play a role in fast walking contribute to the retention of this item; heterogeneous presentations following mTBI may all result in slower maximal walking speeds despite varying underlying deficiencies.

3). Gait with Pivot Turn

The fourth item included in the HAM-4-mTBI was the Gait with Pivot Turn from the FGA. This item involves a rapid whole-body turn and, in contrast to the horizontal head turn, requires integrated motion of the head and body to turn and face a new direction of travel. Whole-body turning involves sophisticated predictive and feedback control of proprioceptive, vestibular, and visual information to rotate the body, maintain balance, and stabilize visual information.⁴⁸ Often times, the center-of-mass of the body can extend outside of the base-of-support,^49,50 making turning a complex dynamic action that challenges balance control. Individuals with sport-related concussion exhibit abnormal balance control during turning that lasts beyond symptom resolution,⁵¹ suggesting this complex coordination for balance is impaired by the injury. Coming to a complete stop at the end of the pivot turn increases the balance and stability requirement over normal turning when additional steps can help correct for any unwanted motion of the center of mass. In addition to balance-related issues, the speed of turning is qualitatively assessed using the FGA scoring criteria and may be affected by mTBI. The speed of whole body turning distinguishes individuals with chronic vestibular loss⁵² and is slower during 180 degree turns at a self-selected walking speed in service members with persistent mTBI symptoms as compared to healthy controls.⁵³ Further, the rapid reorientation of the head and body can elicit somatic symptoms in populations with vestibular disorders.⁵⁴ Slower turning speeds in individuals with persistent complaints of imbalance were related to self-report of somatosensory symptoms,⁴⁷ suggesting that slower turning speeds may be a strategy to avoid provoking symptoms.

Limitations and a need for replication and validation

Several limitations should be considered when interpreting the results of this study. First and foremost, these results are based on a relatively small, single sample of adults with and without mTBI. While 110 individuals are included in this sample, this sample size prohibited us from splitting our sample into separate datasets for cross-validation of AUCs. Therefore, our results present a promising approach that should be further researched. Robust replication of the results is needed before the HAM-4-mTBI is recommended for clinical implementation. Secondly, we recommend the standard scoring method for the HAM-4-mTBI based on the simplicity and alignment with standard scoring of the FGA and HiMAT. However, this scoring does introduce a discrepancy between the FGA items and the HiMAT items; FGA items (Gait with Horizontal Head Turns, Gait with Pivot Turn) are scored from 0–3 possible points, while HiMAT items (Fast Backward Walking, Fast Forward Walking) are scored from 0–4 possible points. Despite the apparent difference, the HiMAT is only scored a 0 if the participant is unable to perform the task. Therefore, all items will retain a 4-point range (0–3 or 1–4) in all but the most severe cases. While this may present a limitation, future research should investigate whether all items should be equally scaled. Additionally, this standard scoring may still elicit ceiling effects. Other assessments of function, including the Community Balance & Mobility Scale (CB&M)⁵⁵ and the short CB&M (s-CB&M),⁵⁶ may not be vulnerable to the same ceiling effects⁵⁷ but similar concerns about assessment time remain given the number of individual items. While many items between the CB&M, HiMAT, and FGA are similar, future work may consider replicating the development of the HAM-4-mTBI using new items from the CB&M or s-CB&M such as lateral foot scooting. We elected to use the FGA and HiMAT as starting points given their ubiquitous use in physical therapy clinics for people with mTBI. However, the choice of many physical therapists may not reflect the most ideal assessments for a given pathology. For instance, familiarity in scoring, time considerations (as mentioned above), and recommendations for Core Outcome measures across populations may influence a provider’s decisions on which assessment battery to use. The recent Clinical Practice Guidelines for Physical Therapy Evaluation and Treatment after Concussion / mTBI recommends the “selection of timing of motor performance assessments should be based on clinical judgement about which evaluation strategies are most appropriate” due to “insufficient evidence to inform selection of motor function assessments.”⁵⁸ A more comprehensive evaluation that starts from scratch, rather than using existing assessments common to clinical practice, may find different results. Finally, it is possible that a single battery is not ideal for all participants with persistent symptoms from mTBI. Heterogeneous symptom profiles, activity levels, and time since injury may make specific items more relevant for some subsets of the population. Future work may consider if the HAM-4-mTBI has different utility based on specific patient characteristics.

CONCLUSIONS:

These findings in adults with and without mTBI indicate that a four-item hybrid combination of FGA and HiMAT items achieves equivalent classification between mTBI and healthy control participants in less than half of the time required by current batteries. The four items retained in the HAM-4-mTBI, in order of weighting, were: Gait with Horizontal Head turns, Fast Backward Walking, Fast Forward Walking, and Gait with Pivot Turns. These items target components of movement, stability, and balance that are likely affected in individuals with persistent symptoms following mTBI. These results should be viewed as preliminary, given the sample size and lack of independent replication. Still, the results are promising; and with further research, this battery may aid in assessment and management of individuals with persistent symptoms following mTBI.