Development and initial validation of a digital divided-attention neurocognitive test for use in concussion assessment.

Objective: Computerized neurocognitive tests are commonly used after a concussion injury. The use of reliable and valid tests that utilize a divided-attention task may improve assessment. Therefore, the purpose of this study is to test a digital divided-attention neurocognitive test for test-retest reliability, practise effects, and initial validity. Methods and procedures: One hundred ninety-two subjects (159 healthy, 33 concussed) were assessed utilizing the neurocognitive test. Group comparisons were made between subjects with concussions and matched controls to determine the initial sensitivity of the test. Results: Intraclass correlation coefficients remained high (greater than 0.50) across all time points tested, and practise effects were largest in first retest session but we correlated (single task: R2 = 0.89, divided-attention: R2 = 0.85). Subjects who had experienced concussions performed significantly worse than matched controls on both the maths computation task and shape matching task during the divided-attention test. Conclusion: The mathematical computation component of the divided-attention test yielded high reliability. Practise effects were seen between the first and second testing sessions with smaller, insignificant improvements seen thereafter. Sensitivity to injury was comparable to other digital neurocognitive tests suggesting ongoing testing is warranted.

Acute Sport-Related Concussion Screening for Collegiate Athletes Using an Instrumented Balance Assessment.

CONTEXT: Without a true criterion standard assessment, the sport-related concussion (SRC) diagnosis remains subjective. Inertial balance sensors have been proposed to improve acute SRC assessment, but few researchers have studied their clinical utility. OBJECTIVE: To determine if group differences exist when using objective measures of balance in a sample of collegiate athletes with recent SRCs and participants serving as the control group and to calculate sensitivity and specificity to determine the diagnostic utility of the inertial balance sensor for acute SRC injuries. DESIGN: Cross-sectional cohort study. SETTING: Multicenter clinical trial. PATIENTS OR OTHER PARTICIPANTS: We enrolled 48 participants with SRC (age = 20.62 ± 1.52 years, height = 179.76 ± 10.00 cm, mass = 83.92 ± 23.22 kg) and 45 control participants (age = 20.85 ± 1.42 years, height = 177.02 ± 9.59 cm, mass = 74.61 ± 14.92 kg) at 7 clinical sites in the United States. All were varsity or club collegiate athletes, and all participants with SRC were tested within 72 hours of SRC. MAIN OUTCOME MEASURE(S): Balance performance was assessed using an inertial balance sensor. Two measures (root mean square sway and 95% ellipse sway area) were analyzed to represent a range of general balance measures. Balance assessments were conducted in double-legged, single-legged, and tandem stances. RESULTS: A main effect for group was associated with the root mean square sway measure ( F1,91 = 11.75, P = .001), with the SRC group demonstrating balance deficits compared with the control group. We observed group differences in the 95% ellipse sway area measure for the double-legged ( F1,91 = 11.59, P = .001), single-legged ( F1,91 = 6.91, P = .01), and tandem ( F1,91 = 7.54, P = .007) stances. Sensitivity was greatest using a cutoff value of 0.5 standard deviations (54% [specificity = 71%]), whereas specificity was greatest using a cutoff value of 2 standard deviations (98% [sensitivity = 33%]). CONCLUSIONS: Inertial balance sensors may be useful tools for objectively measuring balance during acute SRC evaluation. However, low sensitivity suggests that they may be best used in conjunction with other assessments to form a comprehensive screening that may improve sensitivity.