Athletic Training and Sports Health Care

Original Research 

Comparison of Neurocognitive Changes Over One Competitive Season in Adolescent Contact and Non-contact Athletes

Ashley S. Long, PhD, LAT, ATC; Janet P. Niemeier, PhD, ABPP (RP); Andrew McWilliams, MD, MPH; Charity G. Patterson, PhD, MSPH; Paul Perrin, PhD; Megan Templin, MPH, MS; David E. Price, MD

Abstract

Purpose:

To compare neurocognitive function in contact and non-contact high school athletes at three points during a competitive season.

Methods:

Fifty-two male high school athletes completed neurocognitive testing preseason, midseason, and postseason. Composite scores on neurocognitive measures of verbal memory, visual memory, visual motor speed, reaction time, and total symptom scores were measured.

Results:

Contact athletes demonstrated a significant decrease (P = .006) preseason to postseason in verbal memory and a significant improvement (P = .023) in visual motor speed composite score. Among the contact group, participants who began participation in organized football before the age of 10 years had significantly lower average scores, comparatively and over time, than those who began after 10 years of age.

Conclusions:

This study demonstrates that specific neurocognitive variables significantly vary over a competitive season, but only decline in contact athletes.

[Athletic Training & Sports Health Care. 201X;XX(X):XX–XX.]

Abstract

Purpose:

To compare neurocognitive function in contact and non-contact high school athletes at three points during a competitive season.

Methods:

Fifty-two male high school athletes completed neurocognitive testing preseason, midseason, and postseason. Composite scores on neurocognitive measures of verbal memory, visual memory, visual motor speed, reaction time, and total symptom scores were measured.

Results:

Contact athletes demonstrated a significant decrease (P = .006) preseason to postseason in verbal memory and a significant improvement (P = .023) in visual motor speed composite score. Among the contact group, participants who began participation in organized football before the age of 10 years had significantly lower average scores, comparatively and over time, than those who began after 10 years of age.

Conclusions:

This study demonstrates that specific neurocognitive variables significantly vary over a competitive season, but only decline in contact athletes.

[Athletic Training & Sports Health Care. 201X;XX(X):XX–XX.]

Although most head impacts do not result in concussions, there is concern, but little evidence, that repetitive low-magnitude head impacts with no visible acute signs or symptoms (subconcussive impacts) may lead to neurocognitive deficits.1 These types of subconcussive impacts are common among adolescent football players; however, the relationship between subconcussive head impacts and short- and long-term neurologic sequelae in this age group is not well studied or understood. Clinicians and sports stakeholders have limited evidence comparing adolescent athletes in contact and non-contact sports, and it remains unknown whether such insults to a developing brain may result in increased neurologic consequences later in life.2

Recent literature demonstrates that functional and/or structural changes occur in the brains of athletes who are exposed to cumulative subconcussive impacts even without a concussion diagnosis.3–10 In adult athletes, the cumulative effects of head impact forces have been reported to affect neuropsychological functioning and are possibly implicated in late-life cerebral pathology.11,12 Even in the absence of concussive symptoms, functional magnetic resonance imaging performed on high school athletes exposed to subconcussive impacts demonstrated changes in the brain.5,6,13 Additionally, several research studies have found memory deficits in participants exposed to repetitive impacts to the head and without clinically diagnosed concussion.3,4,14 Conversely, Gysland et al.15 discovered that the number of hits received from participation in collegiate football over one season resulted in no clinically significant change in neurocognitive measures. Additionally, Miller et al.16 found no changes in verbal memory or processing speed in college football players over one season. Other studies have even documented improvements in neurocognitive measures over one season of football for college-aged athletes, including verbal memory,9 reaction time,9,16 and visual memory.16

Although evidence of the magnitude of subconcussive effects is conflicting and limited, adolescent athletes do appear particularly at risk for head injuries and associated morbidity and mortality. Although there is a relatively low incidence of head injuries resulting in death or permanent neurological deficits, high school athletes are three times more likely to incur such an injury compared to college athletes.17 Given the potential vulnerability presented by a high school athlete's developing brain while being exposed to considerable quantity and magnitude of impact forces, further investigation into the effects of subconcussive impacts on the adolescent age group is warranted.

One potential modifiable variable that has been identified as related to the relationship between repetitive concussions and cognitive decline is age of first helmeted football participation. A retrospective study identified this association in adult former professional football players, but no studies have examined the relationship in adolescent players.18 These authors found that athletes who began contact play before age 12 years were at increased risk for future neurologic impairment compared to those who began after age 12 years. The authors hypothesized that the 2 years between ages 10 and 12 years are when critical neurodevelopment occurs. It should be noted that this study was retrospective in nature and conducted in a specific adult population; therefore, prospective studies are needed to better understand this initial association and to evaluate an adolescent population to see if the association is detectable earlier in life.

Using a prospective cohort design, this study aimed to investigate neurocognitive function in contact and non-contact high school athletes throughout a competitive season at one high school and to demonstrate the feasibility for conducting a larger study. An additional objective was to identify the influence of first age of participation in contact football on neurocognitive scores.

Methods

Participants and Procedures

Inclusion criteria included being a male athlete participating in either football or cross-country at a single urban high school. The school hosts a student population of 59.9% white, 24.1% African American, 7.7% Hispanic, 6.4% Asian, and 2% other. Of the students at the school, 32.3% are classified as economically disadvantaged.19 The demographic distribution of our study population was similar to that of the overall student body. Participants were recruited by a member of the research team at the school's preseason parent/athlete meeting. All male cross-country and football players were invited to participate. Participants were excluded for a previous history of concussion within the past 6 months. Any participant diagnosed as having a concussion during the season was excluded from final data analysis. Participants were divided into two cohorts: contact and non-contact, consisting of football players and cross-country runners, respectively. All research protocols were approved by Atrium Health's Institutional Review Board prior to initiation of the research project.

Neurocognitive function of participants was assessed using the Immediate Post-Concussion Assessment and Cognitive Test (ImPACT) at three time points throughout one sport season: prior to formal fall sport participation (including “preseason” practice), midseason, and within 2 weeks after the conclusion of the fall 2014 season. The ImPACT is a commonly employed assessment tool used in sports concussion research9,13,16,20,21 and has been shown to be reliable over time.22,23 It assesses multiple dimensions of cognitive functioning, including attention span, working memory, sustained and selective attention time, response variability, non-verbal problem solving, and reaction time. Testing was administered in a group setting by a trained investigator (ASL) from the research team who controlled the environment for noise and distraction. Participants were tested in three groups, with approximately 17 testing at one time. Participants were asked to abstain from physical activity at least 90 minutes prior to testing in an effort to eliminate any influences of physical fatigue and ensure more uniform testing over time. All scores at each evaluation interval were compared to the participant's preseason scores, which were confirmed to be valid by the ImPACT software. At each testing session, participants were assessed by the same member of the research team for interval history of concussion and current signs and symptoms of concussion. Additionally, they were queried about their age of first participation in contact football, which was defined as contact being permitted and a requirement to wear a helmet. First age of participation in contact football was analyzed as a continuous predictor variable but, to visualize the effects through graphing, we dichotomized age using a median-split method (5 to 9 vs 10 to 14 years old).

Statistical Analysis

The primary analysis compared the neurocognitive and symptom scores between and within contact and non-contact athletes at three time points throughout the season. For normally distributed measures, means and standard deviations were reported and compared using a mixed model with a random effect for participants, which assumes compound symmetry for the correlation among time points. Baseline and demographic variables including age, height, weight, and race were compared between the two groups of participants. Chi-square tests were performed to compare the frequency of history-related items that are known to affect outcomes of neurocognitive testing in each group. The Student's t test was used for data measured on an interval scale and the Wilcoxon ranked-sum test was employed for non-normally distributed variables.

Pearson product moment correlations (r) reflect the strength of the linear association between neurocognitive test scores at preseason and postseason. Means and standard deviations of composite scores were calculated for all participants at preseason and postseason. We chose to calculate reliable change indices (RCIs) to assess whether changes between preseason and postseason were reliable and meaningful, and not influenced by practice effects, error, and demographic factors. A comparison of means may demonstrate a significant difference between contact and non-contact groups, but examination of the RCI allows observation of change that exceeds a threshold that can be attributed to the difference between contact and non-contact athletes and is a comparison to the normative values of our specific patient population.24 Although the ImPACT provides an indication when the test-taker has scored outside of the normal RCI, we calculated our own RCI based on the participants in our study, similar to previously published work25–27 and in an effort to account for the age differences between our participants (mean age = 15.5 years) versus the ImPACT's RCI testing group (mean age = 17.6 years). RCIs were calculated at the 80% level. When participants score outside of the RCI, we can conclude that there is a true clinical change as opposed to a statistical difference in mean scores between groups. Both are presented in this article.

Broglio and Puetz28 conducted a meta-analysis that showed effect sizes of computerized neurocognitive testing on individuals with concussion to be 0.81, but ranging from −0.11 to −14.92. We hypothesized that the effect size of a computerized test to detect differences in individuals without concussion would be less, therefore lowering the effect size and the power to detect differences. Our original power analysis required 40 individuals per group to achieve 95% power and an effect size of 0.81. Although the effect size is large, we anticipated smaller, more subtle differences between the groups because we were excluding individuals with concussion and only examining those who did not have concussion. With 26 individuals per group, we achieved 82% power to detect an effect size of 0.81, but we suspect that the effect size of subconcussive impacts is much lower than concussive impacts (closer to 0.5), therefore rendering us underpowered. Subsequently, with 26 individuals per group, assuming an effect size of 0.5 and an alpha level of 0.05, our power yielded 42%. Because we aimed to have a fully powered study, we planned the sample size around a smaller but moderate effect size of 0.5 with 85% power and an alpha level of 0.01 given the multiple outcomes that were being compared. With these parameters, we needed 107 individuals per group.

Secondary mixed models with a random effect for participants examined whether continuous age at first play, time, and the age at first play-by-time interaction could predict memory scores across baseline, midseason, and postseason for the football group only. SAS Enterprise Guide software (version 6.1; SAS, Inc., Cary, NC) was used for all analyses. A two-tailed P value of less than .05 was considered statistically significant.

Results

Twenty-seven contact participants (mean age = 15.77 ± 1.1 years) and 26 non-contact participants (mean age = 15.23 ± 1.1 years) enrolled in the study. One contact participant was removed from the study after suffering a concussion during the season. Groups were similar on measures of age and height, but demonstrated significant differences in weight and race (Table 1). The Fisher's exact test confirmed our assumption that contact players did not differ from non-contact players in previous history of concussion (chi-square = 3.014, P = .084), history of migraines (chi-square = 1.083, P = .25), and learning disabilities (chi-square = 1.486, P =.16) (Table 1).

Demographic Characteristics and Relevant History for Contact and Non-contact Participants

Table 1:

Demographic Characteristics and Relevant History for Contact and Non-contact Participants

Neurocognitive

The results from ImPACT composite scores across time are displayed in Table 2 with means and standard deviations on neurocognitive measures for preseason, midseason, and postseason for contact and non-contact groups. Mixed model analyses demonstrated no significant differences in measures between contact and non-contact participants over time: verbal memory (F(1,52) = 2.84, P = .063), visual memory (F(1,52) = .01, P = .99), and visual motor speed (F(1,52) = 2.2, P = .11) or reaction time (F(1,52) = 0.058, P = .716). A generalized estimating equation showed no differences between contact and non-contact participants over time on total symptom score (Wald statistic = 2.19, P = .34). As a group, mixed model analysis demonstrated a significant decline over time on verbal memory (F(1,26) = 5.38, P = .006) and improvement in visual motor speed (F(1,26) = 3.94, P = .023) in contact participants. Visual memory and reaction time resulted in no significant change over time within each group (Table 2). Symptom scores were non-normally distributed and a generalized estimating equation showed neither main (Wald statistic = 2.19, P = .335) nor interaction (contact: P = .86; non-contact: P = .332) effects.

Neurocognitive Composite Scores by Group Across Testing Session

Table 2:

Neurocognitive Composite Scores by Group Across Testing Session

Mean scores and standard deviations for all participants during preseason and postseason were calculated. Pearson correlations between preseason and postseason assessments ranged from 0.51 to 0.56 (Table 3). RCIs were calculated for all composite and total symptom scores. The reliable change difference scores are presented at 80% confidence intervals (Table 3). Contact participants differed significantly (P = .025) from non-contact participants on visual motor speed, with 19% of contact participants improving beyond the 80% RCI versus 0% of non-contact participants (Table 4). Rates of improvement or decline outside of the 80% confidence interval were not significantly different between groups on any other measure.

Correlation and Reliable Change Difference Scores From Preseason to Postseason

Table 3:

Correlation and Reliable Change Difference Scores From Preseason to Postseason

Changes Outside of the 80% RCI From Preseason to Postseason

Table 4:

Changes Outside of the 80% RCI From Preseason to Postseason

Age of First Participation

Contact participants who began play between 5 and 9 years of age (n = 12) had significantly lower measures of verbal memory at all three time points (b = 1.42, t = 2.14, P = .042) when compared to contact participants who began playing contact football at age 10 years or older (n = 14). The group beginning play between 5 and 9 years of age also demonstrated a decline in verbal memory of 11.92 points from preseason to postseason, but the decline was not significant compared to the decline of 1.59 points in the group who began play at age 10 years or older (b = 0.49, t = 1.24, P =.221; Figure 1).

Verbal memory scores of participants beginning contact, helmeted football from 5 to 9 years old were plotted in comparison to those who began at ages 10 to 14 years old. Participants who began playing football at a younger age had lower verbal memory scores across three time points. There was a significant main effect (P = .042) of age on verbal memory over time.

Figure 1.

Verbal memory scores of participants beginning contact, helmeted football from 5 to 9 years old were plotted in comparison to those who began at ages 10 to 14 years old. Participants who began playing football at a younger age had lower verbal memory scores across three time points. There was a significant main effect (P = .042) of age on verbal memory over time.

Contact participants who began play at a younger age had significantly lower visual memory scores across the three time points (b = 2.01, t = 2.78, P = .010). Further, those who began playing football at a younger age showed a decline in visual memory over time, whereas visual memory remained constant for those who began playing football at an older age (b = 1.52, t = 3.24, P = .002; Figure 2).

Visual memory scores of participants beginning contact, helmeted football from 5 to 9 years old were plotted in comparison to those who began at ages 10 to 14 years. Participants who began playing football between the ages of 5 and 9 years demonstrated significantly lower visual memory scores at all three points in time (P = .002) and significantly lower than those beginning contact football at ages 10 to 14 years old (P = .010).

Figure 2.

Visual memory scores of participants beginning contact, helmeted football from 5 to 9 years old were plotted in comparison to those who began at ages 10 to 14 years. Participants who began playing football between the ages of 5 and 9 years demonstrated significantly lower visual memory scores at all three points in time (P = .002) and significantly lower than those beginning contact football at ages 10 to 14 years old (P = .010).

Discussion

No significant differences were observed between contact and non-contact participants on visual memory, verbal memory, or reaction time variables. However, we did observe significant decreases in verbal memory for the contact participants over one season of participation, and particularly in those who began participation before the age of 10 years. Visual memory also decreased in those beginning participation in contact football before the age of 10 years. Finally, there was a significant increase in visual motor speed in contact players over the season when compared to non-contact players.

Because this was a feasibility study with a small sample size, and the effect size of subconcussive impacts is presumably small, all of these findings must be interpreted as exploratory. It is likely that many more participants would be needed to examine across-group comparisons. The within-group declines and age at first play differences we observed support the need for additional research attention in this area. Additionally, this study demonstrates a study framework allowing future research to apply rigorous methodology that will begin to address the discrepancies and gaps in adolescent subconcussive evidence. However, the retrospective findings of interviews with National Football League retired players showing a long career in football is associated with cognitive, neurobehavioral, and psychological impairments29 and the contrasting finding30 of no evidence of deficits related to repetitive hits for high school and college football players show that the area of study is complex and findings inconsistent. Further study is warranted to explain these variations.

In addition, as with many domains of brain injury study, differences in the choices of baseline and outcome measures and lack of studies of young players make it difficult to decipher clinical implications. Perhaps most important is the ongoing post-career evidence of reduced capacity and risk of depression that players continue to face after retirement,31–33 which highlights the need for further research.

Verbal Memory

Poor verbal memory has been previously identified in high school football players without concussion exposed to cumulative impacts and has furthermore been linked to changes in white matter integrity.8 Verbal memory may be a particularly vulnerable function in adolescent athletes participating in contact sports. Neurophysiological changes observed with functional magnetic resonance imaging have been correlated with deficits in verbal working memory tasks in high school football athletes.4 It has been hypothesized that the increased vulnerability of sport concussion in adolescents could be due to the frontal portion of the brain, which is responsible for working memory and many executive functions, completing its final stages of maturation during this time.34–36 In contrast, previous literature measuring working memory in adults shows no difference after exposure to subconcussive hits.15,37 It is worth noting that our RCI analysis revealed 0% of contact participants improving from preseason to postseason, whereas 19% demonstrated a significant decline, outside of the RCI (F(1,52) = 0.011, P = .073) (Table 4). Although we did not observe a significant difference between groups for verbal memory, it warrants further investigation.

Visual Motor Speed

The contact cohort experienced a significant increase over time in visual motor speed composite scores compared to non-contact scores. Nineteen percent of contact participants scored above the RCI, indicating notable clinical improvement over the season that was significantly greater than in non-contact participants. This reveals an improvement in dimensions of visual processing, learning and memory, and visual motor speed. Although one previous study has shown significant improvement,38 most report no change in measure of visual motor speed or similar correlates preseason and postseason in contact athletes.9,16,39,40 Visual motor speed has been shown to be reliable in test–retest measures.41 One possibility for improvement may be a function of the varied demands between the two sports included in this study. For example, football may challenge visual motor speed more than cross-country, thus resulting in an improvement effect over time.

Visual Memory

Our study showed no significant change over time or significant difference between contact and non-contact groups on visual memory. However, when comparing those who began contact participation before the age of 10 years, there was a significant difference in visual memory for contact participants both overall and over time. Shenk et al.3 reported that visual working memory declined throughout a season of participation in high school football players without clinically diagnosed concussion, although their assessment method was different. High school football players exposed to increased collision events, particularly high impact to the top front of the helmet, develop functional changes in visual working memory comparable to athletes with diagnosed concussion.14 Studies conducted with college players have similarly shown no clinically significant change in visual memory over time.9,16 Although our RCI analysis reveals 19% of contact athletes falling below the RCI compared to 8% of non-contact athletes, the statistical comparison is not significant because of the similarity in improvements between groups. Our RCI was more conservative than the ImPACT's established RCI values. Still, almost 20% of our contact athletes fell below this conservative range, warranting further investigation of this variable.

Age of First Contact Exposure and Future Neurocognitive Function

Rather than choosing a specific age of vulnerability as a differentiation between groups a priori, we analyzed the data using age as a continuous variable, then examined the results of each neurocognitive component. We observed a significant reduction in performance as a group and over time on verbal and visual memory scores in athletes who began participation in organized contact football at age 9 years or younger. These participants had significantly lower verbal memory scores across the three time points. Also, the group had significantly lower scores for visual memory overall and a greater decrease over time when compared to the 10 to 14 year age group. A considerable amount of research in the area of traumatic brain injury suggests that age at the time of injury influences the functional outcomes, with a well-documented age-dependent vulnerability.42 Prime periods of distinct, rapid, neurodevelopment, such as increased myelination, synaptogenesis, and synaptic pruning, as well as varying white and gray matter volumes, may present specific functional detriment to children who suffer impacts.

The findings of this study suggest that further research on a larger sample population is needed to better understand both age at first play and the subsequent neurocognitive effects of participation in contact sports. Given the potential unique vulnerability of the developing brain, policy makers and clinicians need evidence to guide recommendations and define profiles of patients who are at risk for neurocognitive injury, as has been done in concussion.

This study has limitations. The small sample size in this feasibility study did not meet the desired size set in the original power analysis, and therefore limits exist to detect significant differences between groups. Although the ImPACT is a validated tool for concussion and is a clinically useful tool, its validity in detection of abnormalities in the subconcussive realm is unknown. Some of the group differences, particularly verbal memory, between contact and non-contact players were notable but did not reach statistical significance. The quantity and magnitude of hits received by the contact group were not assessed due to funding limitations. Incorporating these data in the future will add to a deeper understanding of the relationship between cumulative head impacts and neurocognitive changes. Also, ethnicities were not equally represented between groups and the effect of race on performance over time is unknown. Kontos et al.43 found no difference on baseline neuropsychological testing using the ImPACT software between African Americans and whites, but saw a difference between races in testing after concussion. A further limitation is that not all subconcussive impacts are accumulated in organized sports, so non-contact participants may have been exposed to impacts that are not accounted for. Also, it is possible that a more sensitive instrument is needed to capture neurocognitive deficits with repeated subconcussive hits. There may be other potential extraneous variables that influence the contact group and affect their performance on the ImPACT test that are not related to participation in a contact sport. Finally, we did not collect baseline intelligence data from participants, which could be an additional confounder that should be included in future research.

The findings from this study suggest that contact players experience a significant decline in verbal memory and a clinically detectable increase in visual motor speed. Age of first participation in contact football was associated with worse verbal memory and visual memory. Future research directions include conducting a larger study that incorporates measures of hit exposure, baseline intelligence testing, examination of a deficit profile, age of first contact play, length of time for which changes are detected, and the use of biomarker measures or advanced diagnostic imaging to examine the unique physiological response of the adolescent brain to subconcussive impacts. Because so little is known in the area of adolescent subconcussive research, novel approaches such as the contemporaneous control of a non-contact high school athlete group as deployed in this study are necessary.

Implications for Clinical Practice

Neurocognitive changes occur in contact athletes even over one competitive season. Athletic trainers and sports medicine physicians should be aware of the possibility of neurocognitive changes over one season even in asymptomatic contact athletes.

References

  1. Belanger HG, Vanderploeg RD, McAllister T. Subconcussive blows to the head: a formative review of short-term clinical outcomes. J Head Trauma Rehabil. 2016;31:159–166. doi:10.1097/HTR.0000000000000138 [CrossRef]
  2. Schneider GE. Is it really better to have your brain lesion early? A revision of the “Kennard principle.”Neuropsychologia. 1979;17:557–583. doi:10.1016/0028-3932(79)90033-2 [CrossRef]
  3. Shenk TE, Robinson ME, Svaldi DO, et al. fMRI of visual working memory in high school football players. Dev Neuropsychol. 2015;40:63–68. doi:10.1080/87565641.2015.1014088 [CrossRef]
  4. Talavage TM, Nauman EA, Breedlove EL, et al. Functionally-detected cognitive impairment in high school football players without clinically-diagnosed concussion. J Neurotrauma. 2014;31:327–338. doi:10.1089/neu.2010.1512 [CrossRef]
  5. Abbas K, Shenk TE, Poole VN, et al. Alteration of default mode network in high school football athletes due to repetitive subconcussive mild traumatic brain injury: a resting-state functional magnetic resonance imaging study. Brain Connect. 2015;5:91–101. doi:10.1089/brain.2014.0279 [CrossRef]
  6. Breedlove EL, Robinson M, Talavage TM, et al. Biomechanical correlates of symptomatic and asymptomatic neurophysiological impairment in high school football. J Biomech. 2012;45:1265–1272. doi:10.1016/j.jbiomech.2012.01.034 [CrossRef]
  7. Johnson B, Neuberger T, Gay M, Hallett M, Slobounov S. Effects of subconcussive head trauma on the default mode network of the brain. J Neurotrauma. 2014;31:1907–1913. doi:10.1089/neu.2014.3415 [CrossRef]
  8. Davenport EM, Whitlow CT, Urban JE, et al. Abnormal white matter integrity related to head impact exposure in a season of high school varsity football. J Neurotrauma. 2014;31:1617–1624. doi:10.1089/neu.2013.3233 [CrossRef]
  9. Volberding JL, Melvin D. Changes in ImPACT and graded symptom checklist scores during a division I football season. Athletic Training and Sports Health Care. 2014;6:155–160. doi:10.3928/19425864-20140710-02 [CrossRef]
  10. Puvenna V, Brennan C, Shaw G, et al. Significance of ubiquitin carboxy-terminal hydrolase L1 elevations in athletes after subconcussive head hits. PLoS One. 2014;9:e96296. doi:10.1371/journal.pone.0096296 [CrossRef]
  11. Moser RS, Schatz P, Jordan BD. Prolonged effects of concussion in high school athletes. Neurosurgery. 2005;57:300–306. doi:10.1227/01.NEU.0000166663.98616.E4 [CrossRef]
  12. Guskiewicz KM, Marshall SW, Bailes J, et al. Association between recurrent concussion and late-life cognitive impairment in retired professional football players. Neurosurgery. 2005;57:719–726. doi:10.1227/01.NEU.0000175725.75780.DD [CrossRef]
  13. Breedlove KM, Breedlove EL, Robinson M, et al. Detecting neurocognitive and neurophysiological changes as a result of subconcussive blows in high school football athletes. Athletic Training and Sports Health Care. 2014;6:119–127. doi:10.3928/19425864-20140507-02 [CrossRef]
  14. Talavage TM, Nauman EA, Breedlove EL. Functionally-detected cognitive impairment in high school football players without clinically-diagnosed concussion. J Neurotrauma. 2014;31:327–338. doi:10.1089/neu.2010.1512 [CrossRef]
  15. Gysland SM, Mihalik JP, Register-Mihalik JK, Trulock SC, Shields EW, Guskiewicz KM. The relationship between subconcussive impacts and concussion history on clinical measures of neurologic function in collegiate football players. Ann Biomed Eng. 2012;40:14–22. doi:10.1007/s10439-011-0421-3 [CrossRef]
  16. Miller JR, Adamson GJ, Pink MM, Sweet JC. Comparison of pre-season, midseason, and postseason neurocognitive scores in uninjured collegiate football players. Am J Sports Med. 2007;35:1284–1288. doi:10.1177/0363546507300261 [CrossRef]
  17. Boden BP, Tachetti RL, Cantu RC, Knowles SB, Mueller FO. Catastrophic head injuries in high school and college football players. Am J Sports Med. 2007;35:1075–1081. doi:10.1177/0363546507299239 [CrossRef]
  18. Stamm JM, Bourlas AP, Baugh CM, et al. Age of first exposure to football and later-life cognitive impairment in former NFL players. Neurology. 2015;84:1114–1120. doi:10.1212/WNL.0000000000001358 [CrossRef]
  19. Charlotte-Mecklenburg Schools. School Progress Report 2010–2011. Myers Park High School; 2011.
  20. Majerske CW, Mihalik JP, Ren D, et al. Concussion in sports: post-concussive activity levels, symptoms, and neurocognitive performance. J Athl Train. 2008;43:265–274. doi:10.4085/1062-6050-43.3.265 [CrossRef]
  21. Broglio SP, Macciocchi SN, Ferrara MS. Sensitivity of the concussion assessment battery. Neurosurgery. 2007;60:1050–1057. doi:10.1227/01.NEU.0000255479.90999.C0 [CrossRef]
  22. Nakayama Y, Covassin T, Schatz P, Nogle S, Kovan J. Examination of the test-retest reliability of a computerized neurocognitive test battery. Am J Sports Med. 2014;42:2000–2005. doi:10.1177/0363546514535901 [CrossRef]
  23. Schatz P. Long-term test-retest reliability of baseline cognitive assessments using ImPACT. Am J Sports Med. 2010;38:47–53. doi:10.1177/0363546509343805 [CrossRef]
  24. Barr WB. Neuropsychological testing for assessment of treatment effects: methodologic issues. CNS Spectr. 2002;7:300–302, 304–306. doi:10.1017/S1092852900017715 [CrossRef]
  25. Elbin RJ, Schatz P, Covassin T. One-year test-retest reliability of the online version of ImPACT in high school athletes. Am J Sports Med. 2011;39:2319–2324. doi:10.1177/0363546511417173 [CrossRef]
  26. Roebuck-Spencer TM, Vincent AS, Schlegel RE, Gilliland K. Evidence for added value of baseline testing in computer-based cognitive assessment. J Athl Train. 2013;48:499–505. doi:10.4085/1062-6050-48.3.11 [CrossRef]
  27. Van Kampen DA, Lovell MR, Pardini JE, Collins MW, Fu FH. The “value added” of neurocognitive testing after sports-related concussion. Am J Sports Med. 2006;34:1630–1635. doi:10.1177/0363546506288677 [CrossRef]
  28. Broglio SP, Puetz TW. The effect of sport concussion on neurocognitive function, self-report symptoms and postural control: a meta-analysis. Sports Med. 2008;38:53–67. doi:10.2165/00007256-200838010-00005 [CrossRef]
  29. Casson IR, Viano DC, Haacke EM, Kou Z, LeStrange DG. Is there chronic brain damage in retired NFL players? neuroradiology, neuropsychology, and neurology examinations of 45 retired players. Sports Health. 2014;6:384–395. doi:10.1177/1941738114540270 [CrossRef]
  30. Montenigro PH, Alosco ML, Martin BM, et al. Cumulative head impact exposure predicts later-life depression, apathy, executive dysfunction, and cognitive impairment in former high school and college football players. J Neurotrauma. 2017;34:328–340. doi:10.1089/neu.2016.4413 [CrossRef]
  31. Hart J Jr, Kraut MA, Womack KB, et al. Neuroimaging of cognitive dysfunction and depression in aging retired NFL players: a cross-sectional study. JAMA Neurol. 2013;70:326–335. doi:10.1001/2013.jamaneurol.340 [CrossRef]
  32. Pryor J, Larson A, DeBeliso M. The prevalence of depression and concussions in a sample of active North American semi-professional and professional football players. J Lifestyle Med. 2016;6:7–15. doi:10.15280/jlm.2016.6.1.7 [CrossRef]
  33. Didehbani N, Munro Cullum C, Mansinghani S, Conover H, Hart J Jr, . Depressive symptoms and concussions in aging retired NFL players. Arch Clin Neuropsychol. 2013;28:418–424. doi:10.1093/arclin/act028 [CrossRef]
  34. Luna B, Garver KE, Urban TA, Lazar NA, Sweeney JA. Maturation of cognitive processes from late childhood to adulthood. Child Dev. 2004;75:1357–1372. doi:10.1111/j.1467-8624.2004.00745.x [CrossRef]
  35. Luna B, Padmanabhan A, O'Hearn K. What has fMRI told us about the development of cognitive control through adolescence?Brain Cogn. 2010;72:101–113. doi:10.1016/j.bandc.2009.08.005 [CrossRef]
  36. O'Hare ED, Lu LH, Houston SM, Bookheimer SY, Sowell ER. Neurodevelopmental changes in verbal working memory load-dependency: an fMRI investigation. Neuroimage. 2008;42:1678–1685. doi:10.1016/j.neuroimage.2008.05.057 [CrossRef]
  37. McAllister TW, Flashman LA, Maerlender A, et al. Cognitive effects of one season of head impacts in a cohort of collegiate contact sport athletes. Neurology. 2012;78:1777–1784. doi:10.1212/WNL.0b013e3182582fe7 [CrossRef]
  38. Schatz P, Putz BO. Cross-validation of measures used for computer-based assessment of concussion. Appl Neuropsychol. 2006;13:151–159. doi:10.1207/s15324826an1303_2 [CrossRef]
  39. Munce TA, Dorman JC, Odney TO, Thompson PA, Valentine VD, Bergeron MF. Effects of youth football on selected clinical measures of neurologic function: a pilot study. J Child Neurol. 2014;29:1601–1607. doi:10.1177/0883073813509887 [CrossRef]
  40. Matser EJ, Kessels AG, Lezak MD, Jordan BD, Troost J. Neuropsychological impairment in amateur soccer players. JAMA. 1999;282:971–973. doi:10.1001/jama.282.10.971 [CrossRef]
  41. Resch J, Driscoll A, McCaffrey N, et al. ImPact test-retest reliability: reliably unreliable?J Athl Train. 2013;48:506–511. doi:10.4085/1062-6050-48.3.09 [CrossRef]
  42. Semple BD, Blomgren K, Gimlin K, Ferriero DM, Noble-Haeusslein LJ. Brain development in rodents and humans: identifying benchmarks of maturation and vulnerability to injury across species. Prog Neurobiol. 2013;106–107:1–16.
  43. Kontos AP, Elbin RJ 3rd, Covassin T, Larson E. Exploring differences in computerized neurocognitive concussion testing between African American and White athletes. Arch Clin Neuropsychol. 2010;25:734–744. doi:10.1093/arclin/acq068 [CrossRef]

Demographic Characteristics and Relevant History for Contact and Non-contact Participants

CharacteristicContact (n = 26)Non-Contact (n = 26)P
DemographicsMean ± SDMean ± SDP (Wilcoxon)

Age (August 1)15.77 ± 1.1415.23 ± 1.14.0947
Height (inches)69.73 ± 2.4168.42 ± 3.59.1156
Weight (pounds)169.73 ± 33.79131.38 ± 22.44< .0001

EthnicityFrequency (Col %)FrequencyTotal

African American7 (26.92%)0 (0%)7
Asian/Asian American0 (0%)1 (3.85%)1
Hispanic/Latino1 (3.85%)0 (0%)1
Mixed race2 (7.69%)0 (0%)2
Non-Hispanic/White15(57.69%)25 (96.15%)40
Not reported1 (3.86%)0 (3.86%)1

Medical historyFrequency (Col %)FrequencyChi-square, P

Previous concussion5 (19.2%)1 (3.8%)3.014, .083
Learning disability2 (7.7%)5 (19.2%)1.486, .223
Previous diagnosis of migraine1 (3.8%)3 (11.5%)1.083, .298

Neurocognitive Composite Scores by Group Across Testing Session

ScorePreseason (n = 52)Midseason (n = 52)Postseason (n = 52)P




Mean ± SDES (95% CI)Mean ± SDES (95% CI)Mean ± SDES (95% CI)Sport by TimeTime Effect
Verbal memory composite
  Contact81.31 ± 11.340.0063 (−0.53, 0.54)80.85 ± 10.14−0.023 (−0.56, 0.52)74.15 ± 15.15−0.51 (−1.06, 0.04).0632.0060a
  Non-contact81.23 ± 13.830.0063 (−0.53, 0.54)81.15 ± 15.71−0.023 (−0.56, 0.52)81.38 ± 13.00−0.51 (−1.06, 0.04).0632.9954
Visual memory composite
  Contact73.35 ± 11.89−0.12 (−0.66, 0.42)72.38 ± 14.68−0.12 (−0.65, 0.42)69.92 ± 17.57−0.073 (−0.61, 0.47).9946.4437
  Non-contact74.92 ± 13.51−0.12 (−0.66, 0.42)74.04 ± 13.98−0.12 (−0.65, 0.42)71.19 ± 17.47−0.073 (−0.61, 0.47).9946.3720
Visual motor speed composite
  Contact34.55 ± 7.21−0.075 (−0.61, 0.46)36.30 ± 8.14−0.060 (−0.60, 0.48)37.96 ± 8.910.32 (−0.22, 0.86).1160.0225a
  Non-contact35.08 ± 6.94−0.075 (−0.61, 0.46)36.75 ± 6.92−0.060 (−0.60, 0.48)35.33 ± 7.330.32 (−0.22, 0.86).1160.3402
Reaction time composite
  Contact0.60 ± 0.09−0.12 (−0.66, 0.42)0.61 ± 0.070 (−0.54, 0.54)0.61 ± 0.070.14 (−0.40, 0.68).6817.7594
  Non-contact0.61 ± 0.08−0.12 (−0.66, 0.42)0.61 ± 0.080 (−0.54, 0.54)0.60 ± 0.070.14 (−0.40, 0.68).6817.5965

Correlation and Reliable Change Difference Scores From Preseason to Postseason

Composite MeasurePreseason (n = 52) Mean ± SDPostseason (n = 52) Mean ± SDDifference Mean ± SDrReliable Change Difference Score (80%)
Verbal memory81.3 ± 12.577.8 ± 14.4−3.5 ± 12.80.55916
Visual memory74.1 ±12.670.6 ± 17.4−3.6 ± 15.10.53219
Visual motor speed34.8 ± 7.036.6 ± 8.21.8 ± 7.50.52510
Reaction time0.6 ± 0.10.6 ± 0.10.0 ± 0.10.5490.09

Changes Outside of the 80% RCI From Preseason to Postseason

MeasureNegative Change > 80% RCIChange < 80% RCIPositive Change > 80% RCIP
Verbal memory
  Football5 (19%)21 (76%)0 (0%).073
  Cross-country2 (8%)20 (77%)4 (15%)
Visual memory
  Football5(19%)21(81%)0 (0%).419
  Cross-country2 (8%)23 (88%)1 (4%)
Visual motor speed
  Football0 (0%)21 (81%)5(19%).025a
  Cross-country2 (8%)24 (92%)0 (0%)
Reaction time
  Football1 (4%)22 (85%)3 (12%).683
  Cross-country3 (12%)20 (77%)3 (12%)
Total symptom score
  Football2 (8%)23 (88%)1 (4%).502
  Cross-country1 (4%)21 (81%)4 (15%)
Authors

From the Department of Family Medicine (ASL), the Center for Outcomes Research and Evaluation (AM, MT), and the Primary Care Sports Medicine Fellowship (DEP), Carolinas HealthCare System, Charlotte, North Carolina; the John F. Butzer Center for Research and Innovation, Mary Free Bed Hospital, Grand Rapids, Michigan (JPN); the Department of Physical Therapy and the Physical Therapy Data Center, University of Pittsburgh, Pennsylvania (CGP); the Department of Psychology, Virginia Commonwealth University, Richmond, Virginia (PP); and the Department of Family Medicine, University of North Carolina School of Medicine, Chapel Hill, North Carolina (DEP).

Support for this research was provided by Carolinas Trauma Network Research Center of Excellence.

The authors have no financial or proprietary interest in the materials presented herein.

The authors thank Mrs. Stephanie Sousa, LAT, ATC, for her assistance in coordinating data collection sessions.

Correspondence: Ashley S. Long, PhD, LAT, ATC, Department of Family Medicine, Carolinas HealthCare System, CMC – Mercy, MMP, Suite 400B, 2001 Vail Avenue, Charlotte, NC 28207. E-mail: Ashley.S.Long@carolinashealthcare.org

Received: August 10, 2017
Accepted: January 10, 2018
Posted Online: June 25, 2018

10.3928/19425864-20180430-01

Sign up to receive

Journal E-contents