Abstract

Sports neuropsychology has emerged as a specialty area within the field of clinical neuropsychology. The role of the sports neuropsychologist, rooted in baseline and post-concussion testing, has evolved to include other clinical domains, including the clinical assessment of potential draft picks. There is no published information on the neurocognitive characteristics of these draft picks. We sought to determine whether elite NFL draft picks differed from NFL roster athletes on neurocognitive (ImPACT) and biopsychosocial characteristics, and given that no published data exists for this population, adopted null hypotheses. Null hypotheses were rejected for two of the four ImPACT scores, as elite draft picks scored higher on measures of visual motor speed and reaction time than roster NFL athletes. Subtle but distinct neurocognitive differences are noted when comparing elite NFL draft picks with norms from a cumulative roster of a single NFL team.

Introduction

The use of baseline neuropsychological testing for concussion assessment and evaluation in the National Football League (NFL) began in 1993 when Dr. Mark Lovell, supported with a grant from NFL Charities, initiated a pilot program with a voluntary group of players from the Pittsburgh Steelers. Baseline neuropsychological testing for concussion gained momentum in the NFL and, in the mid-1990s, several teams voluntarily adopted this approach. By 1996 the NFL's Mild Traumatic Brain Injury (mTBI) Committee began a program attempting to collect baseline and post-concussion test data on participating athletes. These initial baseline testing efforts employed traditional paper and pencil neuropsychological tests adapted for sports concussion purposes. The NFL Neuropsychological Test Battery (Pellman, Lovell, Viano, Casson, and Tucker, 2004) included the Hopkins Verbal Learning Test, Brief Visuospatial Learning Test, Trail Making Test Parts A&B, Digit Span, Symbol Digit Modalities Test, Digits Forward/Backward, and Controlled Oral Word Association Test. Since 2000, several studies have been published in the literature providing normative neurocognitive data on NFL athletes using both paper and pencil and computerized neurocognitive testing.

The first published paper appeared in 2000, when Kutner and colleagues reported the results of a study of 53 active NFL players who were administered MicroCog and underwent Apolipoprotein e4 (ApoE e4) genetic testing, the latter with repeat confirmation. This study was focused on the relationship between age (which served as a surrogate index of concussion exposure), MicroCog test scores, and ApoE e4 status. Athletes were classified into groups based on high/low exposure (i.e., years of NFL play) and ApoE e4 status. Kutner and colleagues found that older athletes who were positive for ApoE e4 had lower cognitive test scores than any of the other groups. Summary data on the General Cognitive Functioning Summary Index were reported by group, but basic psychometric data were not reported for the entire sample, as this was not a primary purpose of the study.

The second paper to appear presented the results of the NFL mTBI's Committee's study of athletes' performances on the NFL Neuropsychological Test Battery (Pellman et al., 2004). A major purpose of this study was to assess the time to return to neurocognitive baseline in NFL athletes sustaining a concussion. Means and standard deviations on the paper and pencil measures were reported on these tests for 165–655 athletes (the number of athletes per test administered varied). The results of the follow-up sample of the NFL athletes participating voluntarily in this study (number of athletes ranged from 95 to 366 per test) were reported by Lovell and Solomon (2011), with few between-group differences noted among the groups of NFL athletes. However, the mean test score differences were found between roster NFL athletes and general population norms on immediate and delayed recall of the Hopkins Verbal Learning Test, Trails A&B, and COWAT (Lovell & Solomon, 2011).

The second study of NFL players' performance on computerized neurocognitive tests was published by Pellman, Lovell, Viano, and Casson (2006). Pellman and colleagues reported Immediate Post Concussion Assessment and Cognitive Testing (ImPACT; Maroon et al., 2000) and demographic data on 68 NFL athletes who completed baseline assessments and 48 NFL athletes who completed post-concussion assessments. A separate group of high school students was also evaluated for comparative purposes. The authors reported ImPACT baseline and post-concussion means and standard deviations for both groups on the ImPACT composite scores, and also provided summary demographic information including player age, years of education, prior number of concussions, on field indicators of concussive injury (e.g., loss of consciousness, amnesia, and disorientation) and position played. One major finding of this study was that the NFL athletes returned to neurocognitive baseline values more rapidly than high school athletes. A second area of interest in this study was the relationship between concussion history and neurocognitive scores, which was found to be non-significant. This study did not assess the relationships between neurocognitive scores and various medical and biopsychosocial variables such as history of psychiatric illness, chemical dependency, age, years of education, presence of attention deficit disorder (ADHD) or learning disability (LD), or specific reported symptoms.

The next study of normative neurocognitive ImPACT performances in NFL players and their relationships to various medical and biopsychosocial variables was published by Solomon and Haase (2008). This study detailed performances on ImPACT among a sample of “consecutively admitted” athletes to an NFL team's roster. This study also examined the effects of various biopsychosocial variables (e.g., age, years of education, prior concussions, history of ADHD or LD, and presence of headache) on ImPACT performance. The results of that study revealed a 6% rate of invalid tests, an absence of an effect for age, education, or number of prior concussions on ImPACT scores, and partial effects of ADHD/LD and headache on ImPACT scores. This study will be referred to as the “roster NFL” athletes in the remainder of this paper.

Over the past decade, the role of neuropsychology in the NFL has increased beyond baseline and postconcussion testing. This movement has been highlighted by the recent creation of the Sports Neuropsychology Society, which held an initial exploratory meeting in Chicago in October of 2011, and an organizational meeting in Denver in April of 2012. Clinical neuropsychologists are now involved in providing sports neuropsychological treatment services to NFL players, as well as providing consultation services to NFL management in areas such as personnel selection. This paper focuses in part on the application of neurocognitive tests to draft pick selection in the NFL.

Each year significant resources are devoted to selecting draft picks for NFL teams. The process is in-depth and lengthy. The process may start at the high school level, intensifies at the collegiate level, and continues with bowl games and all-star games. Professional scouts from NFL teams routinely attend these events to assess potential draft picks. It is safe to say that tens of millions of dollars are spent annually observing, researching, evaluating, and selecting these athletes for participation in the NFL. After the completion of the collegiate football season, the National Invitational Camp holds an NFL Scouting Combine in Indianapolis in March of each year, and approximately 330 collegiate football players are invited to participate in this event.

The invitation-only Combine is attended by NFL coaches, player personnel specialists, and team medical staff. The athletes are evaluated on medical, radiographic, psychological, and behavioral domains. For many years, each athlete at the Combine has been administered (in group settings) the Wonderlic Personnel Test, a 12 min 50-item measure of general cognitive ability used historically for personnel assessment (Wonderlic & Hovland, 1937). The psychometric properties of the Wonderlic are available in the test manual (Wonderlic, 1992). Kuzmits and Adams (2008) found no predictive validity for this measure in terms of eventual NFL success, defined as draft pick order, games played, and salary over 3 years. Similarly, Lyons, Hoffman, and Michel (2009) found no relationship between Wonderlic scores and draft selection position, number of games started, and future NFL performance. Conversely, Gill and Brajer (2011) found a relationship between Wonderlic scores and draft pick position for quarterbacks, offensive linemen, and tight ends. Although scores obtained by each athlete at the Combine are reported to each NFL team, the data are considered confidential.

As of 2011, each athlete attending the Combine was also administered the ImPACT test in a group setting. ImPACT is a computer-based test designed specifically for the assessment of the neurocognitive functions and symptoms associated with sports concussion (Maroon et al., 2000). ImPACT yields composite score on the measures of verbal memory, visual memory, reaction time, visual motor (processing) speed, and total symptoms. ImPACT has been shown to be reliable at 1 month (Schatz & Ferris, 2012), 1 year (Elbin, Schatz, & Covassin, 2011), and at 2 years (Schatz, 2010). The validity of ImPACT in the assessment of sport-related concussion has been demonstrated in multiple studies (Iverson, Lovell, & Collins, 2003, 2005; Schatz, Pardini, Lovell, Collins, & Podell, 2006).

After completion of the Combine, NFL teams can invite selected athletes for an individual visit to team headquarters. During these individual visits, the potential draft picks are evaluated further by team personnel and consultants, including psychologists and neuropsychologists. To date, nothing has been published about the methodologies or the results of these psychological and neuropsychological evaluations of draft picks.

In an attempt to provide some initial empirical data about these draft picks, and in an effort to determine whether they differed on neurocognitive and biopsychosocial variables from roster NFL athletes (Solomon and Haase, 2008), we chose to study a convenience sample of 91 American collegiate football players who were being evaluated as part of a due diligence process for the NFL draft. The sample was composed of those players brought to a NFL team's headquarters for an individual visit after completion of the Combine. We assessed self-reported biopsychosocial data and ImPACT scores in these athletes, evaluated the relationships among these variables, and compared the results with those of the 158 players in the NFL roster study. We made no attempt to match athletes in the current study with the roster study, but were simply looking to determine whether any group differences existed.

Given that no published ImPACT data exists for the draft pick population, we adopted the null hypothesis of no significant group differences.

Method

In the present study, 24 potential draft picks were evaluated in Year 1, 30 athletes in Year 2, and 37 athletes in Year 3, resulting in a final sample of 91 athletes. The specific years covered are not reported in an effort to maintain confidentiality of the data. All 91 athletes in this study were invited to visit team headquarters by a single team after the Combine, but nearly all of the athletes were also evaluated by other NFL teams.

Athletes were chosen based on the NFL team's needs and overall due diligence process. After obtaining Institutional Review Board approval and written informed consent, each athlete underwent a semi-structured interview and a series of standardized neurocognitive tests. All athletes were seen individually by one of the authors. The assessment process was standardized and was composed of: (a) the initial part of the interview, including an assessment of data pertaining to medical, social, family, psychiatric, ADHD, LD, educational, chemical dependency, and legal history information, (b) the supervised administration of the desktop version of ImPACT (Lovell et al., 2002), (c) paper and pencil neuropsychological tests, and (d) concluded with the second part of the interview. The paper and pencil tests are neither listed nor are the results reported at the request of the NFL team, as these tests and scores are used for proprietary purposes. A priori ImPACT results with an Impulse Control Composite score of >30 were considered invalid (Lovell, 2007).

Results

Descriptive Statistics

Of the 91 initial participants, 2 were excluded from the study because of responses to ImPACT that were judged to be of questionable technical satisfaction (i.e., Impulse Control scores >30). Neither of these athletes acknowledged a history of concussion, neurologic illness/disease, ADHD, or LD. One excluded athlete was a first-round draft selection, whereas the other was a second-round selection. The draft selection status of the group is depicted in Table 1. One-third of the sample was first-round picks, and nearly 72% of the sample was eventually selected in the first three rounds of the draft. All athletes were ultimately signed by an NFL team.

Table 1.

Distribution of players by position

Position N 
DB 24 
DL 15 
LB 13 
OL 
QB 
RB 10 
TE 
WR 20 
Position N 
DB 24 
DL 15 
LB 13 
OL 
QB 
RB 10 
TE 
WR 20 

Of the 89 remaining participants, they were, on average, 23 years of age (SD = 1.23) and had 15.6 years of education (SD = 0.66). The sample's concussion history was as follows: 71% of the sample had experienced no concussions, 24% reported one concussion, 5% reported two concussions, and 1% reported having experienced three concussions. Among these 89 athletes, 7% reported having been diagnosed with ADHD, 10% reported an LD, and 3% underwent treatment from a physician for headache. None of the participants reported having sought psychiatric consultation, and no participant reported a history of chemical dependency. The 89 participants in this study represented virtually all the positions on an American football team. The largest group among the sample was the defensive backs (n = 24) and the smallest groups consisted of offensive linemen and quarterbacks (n = 2). The remaining position groups ranged from 3 to 20 individuals. The distribution of players by position is shown in Table 2. The relationship between player position and number of reported concussions incurred bordered on (but did not meet) the conventional level of statistical significance (p = .06).

Table 2.

Draft Pick status of the sample (n = 89)

Round Number chosen % of total sample 
31 34.83 
17 19.10 
16 17.98 
7.86 
4.49 
3.37 
3.37 
UFAa 8.99 
Round Number chosen % of total sample 
31 34.83 
17 19.10 
16 17.98 
7.86 
4.49 
3.37 
3.37 
UFAa 8.99 

aUFA = Undrafted free agent.

The means and standard deviations of the 4 ImPACT scales of verbal memory, visual memory, visual motor (processing) speed, and reaction time, as well as the total symptom score are presented in Table 3. The average scores on the ImPACT scales were similar to those of roster NFL players (Pellman et al., 2006; Solomon & Haase, 2008) on visual memory and reaction time.

Table 3.

Means and standard deviations of the neurocognitive and symptom variables (n = 89)

 Mean SD 
Verbal Memory 80.97 9.14 
Visual Memory 78.49 10.65 
Visual Motor Speed 37.5107 7.09 
Reaction Time 0.58 0.08 
Symptom Total 2.65 3.82 
 Mean SD 
Verbal Memory 80.97 9.14 
Visual Memory 78.49 10.65 
Visual Motor Speed 37.5107 7.09 
Reaction Time 0.58 0.08 
Symptom Total 2.65 3.82 

The draft pick group, on average, scored about ½ SD below the average of the NFL players on verbal memory and about ½ SD above the average of the NFL players on visual motor (processing) speed. No clear group differences in symptom endorsement were noted.

Relationships of Individual Biopsychosocial Characteristics to ImPACT Scales

We assessed the relationship between several demographic characteristics to the 4 ImPACT scores and the total symptom score in a manner similar to the roster NFL study. The relationship between the number of concussions to possible changes in memory, processing speed and reaction time was tested by four groups, one-way multivariate analysis of variance (ANOVA; Rencher, 2002) on four dependent variables. The analysis reveals no significant difference between concussion groups on the pooled optimal linear combination of the dependent variables (Pillai's Trace = 0.098, approximate F = 0.798, df = 12, 252, p = .739). In addition to the omnibus multivariate test, none of the univariate follow-up one-way ANOVA's showed significant differences between the four concussion groups. A similar four-group one-way ANOVA on the total symptom score showed the same lack of differences between concussion groups, F(3, 85) = 1.65, p = .920. We find no evidence among these athletes of any relationship of concussions to ImPACT scores. Means and SD are presented in Table 4.

Table 4.

Means and (SD) of neurocognitive variables by concussion group

Concussion group Verbal memory Visual memory Visual motor speed Reaction time Symptom total 
0 (N = 65) 80.55 (9.86) 77.85 (10.74) 37.05 (7.39) 0.58 (0.08) 2.86 (4.18) 
1 (N = 21) 82.00 (7.94) 79.67 (10.91) 37.87 (6.77) 0.58 (0.09) 2.62 (4.18) 
2–3 (N = 5) 83.40 (3.65) 82.40 (7.64) 36.62 (9.59) 0.59 (0.04) 2.00 (2.00) 
Concussion group Verbal memory Visual memory Visual motor speed Reaction time Symptom total 
0 (N = 65) 80.55 (9.86) 77.85 (10.74) 37.05 (7.39) 0.58 (0.08) 2.86 (4.18) 
1 (N = 21) 82.00 (7.94) 79.67 (10.91) 37.87 (6.77) 0.58 (0.09) 2.62 (4.18) 
2–3 (N = 5) 83.40 (3.65) 82.40 (7.64) 36.62 (9.59) 0.59 (0.04) 2.00 (2.00) 

Age and education were also assessed as to their relationship to the ImPACT scores and total symptoms. The Pearson correlations between age and verbal memory (r = −.004), visual memory (r = −.046), visual motor speed (r = .003), and reaction time (r = −.103) were all not significantly different from zero (all p's > .34). Similar findings emerged for the age-symptom total relationship (r = .043, p = .69). For the education variable, the pattern was similar to age with r's = .222, .333, and .333 for verbal memory, visual memory and reaction time, respectively. The exception to this pattern was a significant correlation between education and visual motor (processing) speed (r = .234, p < .027).

Among the remaining individual characteristics of headache treatment, ADHD, and LD, we found no evidence of a connection to the ImPACT scores. Two-group, one-way multivariate ANOVA tests reveal that treatment for headache (Pillai's Trace = 0.059, approximate F(4, 84) = 1.31, p > .274), presence of ADHD (Pillai's Trace = 0.068, approximate F(4, 84) = 1.53, p > .201), and presence of LD (Pillai's Trace = 0.057, approximate F(4, 84) = 1.28, p > .284) have no discernible relationship to the collection of 4 ImPACT scores. Similarly, the symptom total was not found to be related to any of these characteristics.

Tables 5–8 give the percentile scores for verbal memory, visual memory, visual motor (processing) speed, and reaction time for this sample of draft picks.

Table 5.

Percentile norms for verbal memory

Score Cumulative percent 
62 2.2 
64 5.6 
65 6.7 
66 9.0 
67 10.1 
68 11.2 
69 14.6 
70 18.0 
73 22.5 
74 23.6 
76 27.0 
77 28.1 
78 34.8 
79 44.9 
80 48.3 
81 51.7 
82 55.1 
83 58.4 
84 64.0 
85 68.5 
87 69.7 
88 77.5 
89 84.3 
90 85.4 
91 88.8 
92 92.1 
93 93.3 
94 94.4 
96 96.6 
99 98.9 
100 100.0 
Score Cumulative percent 
62 2.2 
64 5.6 
65 6.7 
66 9.0 
67 10.1 
68 11.2 
69 14.6 
70 18.0 
73 22.5 
74 23.6 
76 27.0 
77 28.1 
78 34.8 
79 44.9 
80 48.3 
81 51.7 
82 55.1 
83 58.4 
84 64.0 
85 68.5 
87 69.7 
88 77.5 
89 84.3 
90 85.4 
91 88.8 
92 92.1 
93 93.3 
94 94.4 
96 96.6 
99 98.9 
100 100.0 
Table 6.

Percentile Norms for visual memory

Score Percentile 
52 1.1 
55 3.4 
57 4.5 
59 5.6 
60 6.7 
63 7.9 
64 10.1 
65 13.5 
66 14.6 
67 18.0 
68 19.1 
69 21.3 
71 27.0 
72 29.2 
73 32.6 
74 33.7 
75 36.0 
76 39.3 
77 46.1 
78 48.3 
79 51.7 
81 57.3 
82 62.9 
83 65.2 
84 66.3 
85 69.7 
86 74.2 
88 76.4 
89 82.0 
90 85.4 
91 88.8 
92 91.0 
93 94.4 
94 97.8 
95 98.9 
96 100.0 
Score Percentile 
52 1.1 
55 3.4 
57 4.5 
59 5.6 
60 6.7 
63 7.9 
64 10.1 
65 13.5 
66 14.6 
67 18.0 
68 19.1 
69 21.3 
71 27.0 
72 29.2 
73 32.6 
74 33.7 
75 36.0 
76 39.3 
77 46.1 
78 48.3 
79 51.7 
81 57.3 
82 62.9 
83 65.2 
84 66.3 
85 69.7 
86 74.2 
88 76.4 
89 82.0 
90 85.4 
91 88.8 
92 91.0 
93 94.4 
94 97.8 
95 98.9 
96 100.0 
Table 7.

Percentile Norms for visual motor (processing) speed

Score Percentile 
16.5 1.1 
18.2 2.2 
23.88 3.4 
24.75 4.5 
26.53 5.6 
26.55 6.7 
28.15 7.9 
28.45 9.0 
28.5 10.1 
29.3 11.2 
29.75 12.4 
30.23 13.5 
31.05 14.6 
31.15 15.7 
31.25 16.9 
31.5 18.0 
31.85 19.1 
32.08 20.2 
32.63 21.3 
32.7 22.5 
32.98 23.6 
33.23 24.7 
33.25 25.8 
33.4 27.0 
33.55 28.1 
33.58 29.2 
33.7 31.5 
34.08 32.6 
34.8 33.7 
34.83 34.8 
34.88 36.0 
35 38.2 
35.3 39.3 
36.08 40.4 
36.23 41.6 
36.33 42.7 
36.43 43.8 
36.6 44.9 
36.7 47.2 
36.9 48.3 
37.13 49.4 
37.2 50.6 
37.53 51.7 
37.58 52.8 
37.7 53.9 
37.73 55.1 
38.38 56.2 
38.75 57.3 
39.1 59.6 
39.28 60.7 
39.4 61.8 
39.88 62.9 
40.2 64.0 
40.7 65.2 
40.75 66.3 
41.05 67.4 
41.15 68.5 
41.38 69.7 
41.4 70.8 
41.5 71.9 
41.53 73.0 
41.55 74.2 
41.8 75.3 
42.08 76.4 
42.25 77.5 
42.38 78.7 
42.45 79.8 
42.7 80.9 
42.98 82.0 
43.03 83.1 
43.43 84.3 
44.08 85.4 
44.73 86.5 
45.18 87.6 
46.2 88.8 
46.7 89.9 
47.08 91.0 
48.08 92.1 
48.23 93.3 
50.08 94.4 
50.73 95.5 
51.23 96.6 
51.28 97.8 
51.7 98.9 
51.83 100.0 
Score Percentile 
16.5 1.1 
18.2 2.2 
23.88 3.4 
24.75 4.5 
26.53 5.6 
26.55 6.7 
28.15 7.9 
28.45 9.0 
28.5 10.1 
29.3 11.2 
29.75 12.4 
30.23 13.5 
31.05 14.6 
31.15 15.7 
31.25 16.9 
31.5 18.0 
31.85 19.1 
32.08 20.2 
32.63 21.3 
32.7 22.5 
32.98 23.6 
33.23 24.7 
33.25 25.8 
33.4 27.0 
33.55 28.1 
33.58 29.2 
33.7 31.5 
34.08 32.6 
34.8 33.7 
34.83 34.8 
34.88 36.0 
35 38.2 
35.3 39.3 
36.08 40.4 
36.23 41.6 
36.33 42.7 
36.43 43.8 
36.6 44.9 
36.7 47.2 
36.9 48.3 
37.13 49.4 
37.2 50.6 
37.53 51.7 
37.58 52.8 
37.7 53.9 
37.73 55.1 
38.38 56.2 
38.75 57.3 
39.1 59.6 
39.28 60.7 
39.4 61.8 
39.88 62.9 
40.2 64.0 
40.7 65.2 
40.75 66.3 
41.05 67.4 
41.15 68.5 
41.38 69.7 
41.4 70.8 
41.5 71.9 
41.53 73.0 
41.55 74.2 
41.8 75.3 
42.08 76.4 
42.25 77.5 
42.38 78.7 
42.45 79.8 
42.7 80.9 
42.98 82.0 
43.03 83.1 
43.43 84.3 
44.08 85.4 
44.73 86.5 
45.18 87.6 
46.2 88.8 
46.7 89.9 
47.08 91.0 
48.08 92.1 
48.23 93.3 
50.08 94.4 
50.73 95.5 
51.23 96.6 
51.28 97.8 
51.7 98.9 
51.83 100.0 
Table 8.

Percentile Norms for reaction time

Score Percentile 
0.3 98.9 
0.45 97.8 
0.46 95.5 
0.48 89.9 
0.49 86.5 
0.5 85.4 
0.51 77.5 
0.53 73.0 
0.54 65.2 
0.55 59.6 
0.56 58.4 
0.57 52.8 
0.58 48.3 
0.59 39.3 
0.6 33.7 
0.61 29.2 
0.62 27.0 
0.63 22.5 
0.64 15.3 
0.65 14.6 
0.66 11.2 
0.67 7.9 
0.69 6.7 
0.7 5.6 
0.71 3.4 
0.73 2.2 
0.75 1.1 
0.83 0.0 
Score Percentile 
0.3 98.9 
0.45 97.8 
0.46 95.5 
0.48 89.9 
0.49 86.5 
0.5 85.4 
0.51 77.5 
0.53 73.0 
0.54 65.2 
0.55 59.6 
0.56 58.4 
0.57 52.8 
0.58 48.3 
0.59 39.3 
0.6 33.7 
0.61 29.2 
0.62 27.0 
0.63 22.5 
0.64 15.3 
0.65 14.6 
0.66 11.2 
0.67 7.9 
0.69 6.7 
0.7 5.6 
0.71 3.4 
0.73 2.2 
0.75 1.1 
0.83 0.0 

Discussion

We endeavored to study a group of elite NFL draft picks assessing demographic, concussion history, and neurocognitive (ImPACT) characteristics in an effort to present an initial psychometric portrayal of this group. We made comparisons between the draft pick group and previously published reports of roster NFL players (Pellman et al., 2006; Solomon & Haase, 2008). A total of 91 draft picks were enrolled initially in the study, and 2 were excluded due to neurocognitive data deemed technically unsatisfactory.

Athletes' level of effort on the neurocognitive tests was not formally assessed with commonly accepted clinical neuropsychological effort tests. As a proxy for a formal effort test, we employed the criterion suggested by Lovell (2007), which is to consider desktop ImPACT profiles with an Impulse Control score of >30 as being of doubtful validity. Only 2% of the draft picks produced ImPACT profiles considered technically unsatisfactory, which compares favorably with the prior report of 5.92% of invalid desktop ImPACT profiles noted in roster NFL athletes, 11.9% and 10.2% in high school and collegiate athletes, respectively ( Schatz, Moser, Solomon, Ott, & Karpf, 2012), and about an 11% rate in a high school football sample assessed with paper and pencil neurocognitive tests (Hunt, Ferrara, Miller, & Macciocchi, 2007).

Analysis of demographic data indicated that the draft pick sample was about 4 years younger than average compared with the roster NFL players, and had, on average, about 1 year less of education. The latter finding could indicate that the top-level collegiate athletes are leaving college football and entering the NFL draft sooner than in the past, possibly due to financial incentives as well as matters related to the timing of collective bargaining agreements. As in a previous study of roster NFL athletes, age was not correlated with ImPACT scores, reported baseline symptoms, or number of prior concussions, suggesting that junior and senior collegiate football players may be included with professional football players as a homogeneous group for neurocognitive and symptom assessment normative purposes.

The reported concussion history of the draft pick group was quite similar to that reported by the NFLroster athletes; the results are listed in Table 9.

Table 9.

Concussion history of draft picks and roster NFL athletes

Concussion history Draft picks (%) Roster NFL athletes (%) 
No concussions 71 62 
One concussion 24 22.6 
Two concussions 11.3 
Three or more concussions 4.4 
Concussion history Draft picks (%) Roster NFL athletes (%) 
No concussions 71 62 
One concussion 24 22.6 
Two concussions 11.3 
Three or more concussions 4.4 

The overall average concussion rates were 0.36 in the draft pick group and 0.6 in the NFL group. These rates are also consistent with results from the NCAA studies (Guskiewicz et al., 2003; McCrea et al., 2003) where concussed athletes reported an average of 0.58 (SD = 0.78) concussions over the past 7 years, with control collegiate athletes reporting an average of 0.39 (SD = 0.68) concussions during the same time frame.

The relationship between the number of concussions and player position among the draft picks did not reach the conventional level of statistical significance (p > .06). This is in contrast to the NCAA studies (Guskiewicz et al., 2003; McCrea et al., 2003) where it was shown that the highest risk positions for concussions was offensive lineman, with 20.9% of all concussions and a 0.96 concussion rate per 1000 athletic exposures (A-Es; 0.66–1.24 at the 95% confidence interval [CI]), linebackers, with 16.3% of all concussions and a 0.99 concussion rate per 1000 A-Es (0.65–1.33 at the 95% CI), and defensive backs, with 16.3% of all concussions and a 0.88 concussion rate per 1000 A-Es (0.67–1.18 at the 95% CI). Similarly, the results of the current draft pick sample differ with the NFL concussion studies (Pellman, Powell, Viano, et al., 2004), where quarterbacks, wide receivers, tight ends, running backs, and defensive backs had concussion incidence rates of 7.9%, 11.9%, 4.6%, 8.8%, and 18.2%, respectively. Risk per 100 game positions (defined as the number of concussions divided by the number of times the position was played during the observed period of 3,826 games, multiplied by 100, with 95% CIs in parentheses) was 1.62 for the quarterbacks (1.22–2.02), 1.23 for wide receivers (0.98–1.48), 0.94 for the tight ends (0.63–1.23), 0.90 for the running backs (0.69–1.11), and 0.90 for the defensive backs (0.78–1.08). We suspect that the small sample size of the draft picks (n = 89), when compared with the sample sizes of the NCAA (n = 1,631 athletes across 2,410 player-seasons) and NFL (n = 179,820 game positions played) studies, precluded the emergence of distinct concussion risk factors by position.

The draft pick groups' neurocognitive scores were, on average, about ½ SD below the roster NFL players' mean score on verbal memory, and about ½ SD above the roster NFL players' score on visual motor (processing) speed. As discussed below, the relative high percentage of athletes with acknowledged ADHD and/or LD may have been a factor in the group difference on verbal memory. The difference in processing speed may have been related to their elite draft pick status, as the draft pick group's average score was superior to that of the roster NFL players (Macciocchi, Barth, Littlefield, & Cantu, 2001).

There was also no significant relationship between concussion history and neurocognitive tests scores, a finding similar to the study of the roster athletes. These results also are consistent with several prior studies (Broglio, Ferrara, Piland, & Anderson, 2008; Collie, McCrory, & Makdissi, 2006; Iverson, Brooks, Lovell, & Collins, 2006) but inconsistent with others (Killam, Cautin, & Santucci, 2005; Moser & Schatz, 2002). To date, most of the studies indicating cumulative neurocognitive effects of sports-related concussion appear to apply to younger athletes, as well as to those at the older end of the athletic age spectrum. There are several autopsy reports of former NFL players that have been presented as evidence of cumulative neuropathological (and presumptive neurocognitive effects) of sports-related concussion (Omalu et al., 2005, 2006), but these studies have been criticized on neuropathological and methodological grounds (Casson, Pellman, & Viano, 2006a, 2006b). Other studies have indicated a greater risk for late-life cognitive impairment (Guskiewicz et al., 2005) and depression (Guskiewicz et al., 2007; Kerr, Marshall, Harding, & Guskiewicz, 2012) in retired professional football players, but these studies also have been criticized (Reider, 2012; Solomon, Ott, & Lovell, 2011). The hypothesis of the potentially cumulative effects of concussion on neurocognitive functioning is a hotly debated topic, and will only be answered with any degree of certainty with prospective, well-controlled, longitudinal studies.

It is reported that in the general population, 6% seek treatment for a serious psychiatric illness (NIMH, 2012), while 26.6% seek treatment for substance abuse (Kessler, McGonagle, Zhao, Nelson, Hughes, & Eshelman, 1994). None of the draft picks acknowledged any history of treatment for psychiatric illness or chemical dependency. These results are quite similar to those reported previously for roster NFL players (Solomon & Haase, 2008), where a frequency of <1% for either psychiatric or substance abuse disorders was reported. It appears that diagnosed (or reported) psychiatric and/or chemical dependency disorders may be rather low in draft pick and NFL football athletes. It is also quite possible that the actual history of psychiatric/chemical dependency treatment was minimized by the draft pick athletes, given the context in which they were evaluated.

In the general population, it is reported that about 13–16% seek treatment for headache from a physician (Duckro, Tait, & Margolis, 1989). In the draft pick sample, only 3% acknowledged seeking headache treatment, whereas in the roster NFL players, the frequency was 15%. Among the NFL players, 10.2% of the players with a benign concussion history sought treatment for headache. The comparable figure for those with one or two concussions was 16.7%, and for those with three or more concussions, the rate was 42.9%. These data would suggest that seeking treatment from a physician for headache increases linearly during an athlete's tenure in the NFL. It is unclear, however, whether this is a cohort effect, an age (maturational) effect, an effect of concussion, or possibly an interaction among these (or other) factors.

Education was correlated only with the processing speed score in the draft pick sample (r = .23, p < .03). These results are very similar to the findings in the NFL sample (Solomon & Haase, 2008), where education was not significantly correlated with any of the ImPACT scores or concussion history. The restricted range on the education variable among the athletes may have contributed to the lack of association.

The prevalence of ADHD in the general population is reported to be about 4% (Wilens, Faraone, & Biederman, 2004), while it has been estimated that 2%–10% of the general population is diagnosed with an LD (American Psychiatric Association, 1994). We found reported rates of 7% and 10% for ADHD and LD, respectively, in the draft pick group. This compares with a rate of 9% for ADHD and/or LD in the roster NFL sample reported previously (the two conditions were considered unitarily in the study of roster players). Among the NCAA athletes (Guskiewicz et al., 2003; McCrea et al., 2003), a 2% rate was reported for both ADHD and LD among the concussed and control athletes. The draft picks and the roster NFL athletes have a prevalence of ADHD and LD that appears to be considerably higher than the base rates for these disorders among the general population and in prior reports of NCAA football players.

We found no correlation between the presence of ADHD (n = 6) or LD (n = 9) and any of the ImPACT scores among the draft picks. In a previous study of roster NFL athletes, ADHD and LD were grouped together. In this combined group (ADHD and/or LD), significant correlations of −.16 and −.24 were noted on verbal and visual memory scores, respectively. In one study (Collins et al., 1999), the presence of ADHD and/or LD was noted to have an adverse effect on post-concussion paper and pencil neurocognitive scores among collegiate athletes. Iverson, Collins, and Lovell (2004) found that adolescents with ADHD performed more poorly on ImPACT visual memory and visual motor (processing) speed indexes than adolescents matched for age, gender, education, and concussion history. In other studies of sports-related concussion (Broglio, Ferrara, Macciocchi, Baumgartner, & Elliott, 2007), athletes with ADHD and/or LD have been excluded from participation in the study due to the presumed effects of these disorders on baseline and post-concussion neurocognitive results. The inconsistent findings suggest that, for time being, ADHD and LD may need to be considered separately as moderator variables in assessing the effects of sports-related cognitive functioning, both at baseline and post-injury in all age ranges. This is consistent with the Concussion in Sport Group's (McCrory et al., 2009) delineation of ADHD/LD as “moderating” variables in sport-related concussion.

Limitations of the study

There are several limitations to our study. First, all data related to prior medical, psychiatric, chemical dependency, concussion, and LD/ADHD history information was based on athlete self-report, and no independent confirmation was available. Secondly, this was an observational, retrospective study of a convenience sample, and athletes were not selected randomly for participation. Thirdly, to our knowledge, this is an initial empirical study of this elite group of collegiate football players, and any conclusions should be viewed as preliminary and in need of cross-validation. Fourthly, this study reports the results of a single neurocognitive computerized test (ImPACT), and it is entirely possible that the use of paper and pencil neurocognitive tests or other computerized sports concussion tests might yield different findings. Finally, despite our attempts to review the literature as comprehensively as possible, it is entirely possible that we have inadvertently omitted studies which have bearing on the topics at hand.

Conflict of Interest

None declared.

References

American Psychiatric Association
Diagnostic and statistical manual of mental disorders
 , 
1994
4th ed.
Washington, DC
American Psychiatric Press
Broglio
S.
Ferrara
M.
Macciocchi
S.
Baumgartner
T.
Elliott
R.
Test–retest reliability of computerized concussion assessment programs.
Journal of Athletic Training
 , 
2007
, vol. 
42
 (pg. 
509
-
514
)
Broglio
S.
Ferrara
M.
Piland
S.
Anderson
R.
Concussion history is not a predictor of computerized neurocognitive performance
British Journal of Sports Medicine
 , 
2008
, vol. 
40
 (pg. 
802
-
805
)
Casson
I.
Pellman
E.
Viano
D.
Chronic traumatic encephalopathy in a National Football league player
Neurosurgery
 , 
2006a
, vol. 
58
 pg. 
E1003
  
(Letter)
Casson
I.
Pellman
E.
Viano
D.
Chronic traumatic encephalopathy in a National Football league player: Part II
Neurosurgery
 , 
2006b
, vol. 
59
 pg. 
E1152
  
(Letter)
Collie
A.
McCrory
P.
Makdissi
M.
Does concussion history affect current cognitive status?
British Journal of Sports Medicine
 , 
2006
, vol. 
40
 (pg. 
550
-
551
)
Collins
M.
Grindel
S.
Lovell
M.
Dede
D.
Moser
R.
Phalin
B.
, et al.  . 
Relationship between concussion and neuropsychological test performance in college football players
Journal of the American Medical Association
 , 
1999
, vol. 
282
 (pg. 
964
-
970
)
Duckro
P.
Tait
R.
Margolis
R.
Prevalence of very severe headache in a large US metropolitan area
Cephalagia
 , 
1989
, vol. 
15
 (pg. 
199
-
205
)
Elbin
R. J.
Schatz
P.
Covassin
T.
One-year test-retest reliability of the online version of ImPACT in high school athletes
American Journal of Sports Medicine
 , 
2011
, vol. 
39
 (pg. 
2319
-
2324
)
Gill
A.
Brajer
V.
Wonderlic, race, and the NFL draft
Journal of Sports Economics
 , 
2011
Guskiewicz
K.
Marshall
S.
Bailes
J.
McCrea
M.
Cantu
R.
Randolph
C.
, et al.  . 
Association between recurrent concussion and late-life cognitive impairment in retired professional football players
Neurosurgery
 , 
2005
, vol. 
57
 (pg. 
719
-
726
)
Guskiewicz
K.
Marshall
S.
Bailes
J.
McCrea
M.
Harding
H.
Mathews
A.
, et al.  . 
Recurrent concussion and risk of depression in retired professional football players
Medicine & Science in Sports and Exercise
 , 
2007
, vol. 
39
 (pg. 
903
-
909
)
Guskiewicz
K.
McCrea
M.
Marshall
S.
Cantu
R.
Randolph
C.
Barr
W.
, et al.  . 
Cumulative effects associated with recurrent concussion in collegiate football players: The NCAA concussion study
Journal of the American Medical Association
 , 
2003
, vol. 
290
 (pg. 
2549
-
2555
)
Hunt
T.
Ferrara
M.
Miller
L.
Macciocchi
S.
The effect of effort on baseline neuropsychological test scores in high school football athletes
Archives of Clinical Neuropsychology
 , 
2007
, vol. 
22
 (pg. 
615
-
621
)
Iverson
G.
Brooks
B.
Lovell
M.
Collins
M.
No cumulative effect for one or two previous concussions
British Journal of Sports Medicine
 , 
2006
, vol. 
40
 (pg. 
72
-
75
)
Iverson
G.
Collins
M.
Lovell
M.
Preseason neurocognitive testing in athletes with ADHD
2004
Paper presented at the National Association of School Psychologists
Dallas, TA, USA
Iverson
G. L.
Lovell
M. R.
Collins
M. W.
Interpreting change on ImPACT following sport concussion
The Clinical Neuropsychologist
 , 
2003
, vol. 
17
 (pg. 
460
-
467
)
Iverson
G. L.
Lovell
M. R.
Collins
M. W.
Validity of ImPACT for measuring processing speed following sports-related concussion
Journal of Clinical and Experimental Neuropsychology
 , 
2005
, vol. 
27
 (pg. 
683
-
689
)
Kerr
Z.Y.
Marshall
S.W.
Harding
H.P.
Jr.
Guskiewicz
K. M.
Nine-year risk of depression diagnosis increases with increasing self-reported concussions in retired professional football players
American Journal of Sports Medicine
 , 
2012
, vol. 
40
 (pg. 
2206
-
2212
)
Kessler
R.
McGonagle
K.
Zhao
S.
Nelson
C.
Hughes
M.
Eshleman
S.
Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States: Results from the national Comorbidity Survey
Archives of General Psychiatry
 , 
1994
, vol. 
51
 (pg. 
8
-
19
)
Killam
C.
Cautin
R.
Santucci
A.
Assessing the enduring residual neuropsychological effects of head trauma in college athletes who participate in contact sports
Archives of Clinical Neuropsychology
 , 
2005
, vol. 
20
 (pg. 
599
-
611
)
Kutner
K. C.
Erlanger
D. M.
Tsai
J.
Jordan
B.
Relkin
N. R.
Lower cognitive performance of older football players possessing Apolipoprotein E e4
Neurosurgery
 , 
2000
, vol. 
47
 (pg. 
651
-
657
)
Kuzmits
F. E.
Adams
A. J.
The NFL combine: Does it predict performance in the National Football League?
Journal of Strength and Conditioning Research
 , 
2008
, vol. 
22
 (pg. 
1721
-
1727
)
Lovell
M.
ImPACT 2007 (6.0) clinical interpretation manual. ImPACT applications
 , 
2007
Pittsburgh, PA: ImPACT Applications
Lovell
M.
Collins
M.
Maroon
J.
Powell
J.
Barber-Foss
K.
Powell
J. W.
ImPACT concussion management software (Version 5.0): ImPACT Applications
 , 
2002
Pittsburgh, PA: ImPACT Applications
Lovell
M. R.
Solomon
G. S.
Psychometric data for the NFL neuropsychological test battery
Applied Neuropsychology
 , 
2011
, vol. 
18
 (pg. 
197
-
209
)
Lyons
B. D.
Hoffman
B. J.
Michel
J. W.
Not much more than g? An examination of the impact of intelligence on NFL performance
Human Performance
 , 
2009
, vol. 
22
 (pg. 
225
-
245
)
Macciocchi
S.
Barth
J.
Littlefield
L.
Cantu
R.
Multiple concussions and neuropsychological functioning in collegiate football players
Journal of Athletic Training
 , 
2001
, vol. 
36
 (pg. 
303
-
306
)
Maroon
J. C.
Lovell
M. R.
Norwig
J.
Podell
K.
Powell
J. W.
Hartl
R.
Cerebral concussion in athletes: evaluation and neuropsychological testing
Neurosurgery
 , 
2000
, vol. 
47
 (pg. 
659
-
669
)
McCrea
M.
Guskiewicz
K.
Marshall
S.
Barr
W.
Randolph
C.
Cantu
R.
, et al.  . 
Acute effects and recovery time following concussion in collegiate football players: The NCAA concussion study
Journal of the American Medical Association
 , 
2003
, vol. 
290
 (pg. 
2556
-
2563
)
McCrory
P.
Meeuwisse
W.
Johnston
K.
Dvorak
J.
Aubry
M.
Molloy
M.
, et al.  . 
Consensus statement on concussion in sport: The third international conference on concussion in sport held in Zurich, 2008
British Journal of Sports Medicine
 , 
2009
, vol. 
43
 (pg. 
76
-
84
)
Moser
R.
Schatz
P.
Enduring effects of concussion in youth athletes
Archives of Clinical Neuropsychology
 , 
2002
, vol. 
17
 (pg. 
91
-
100
)
National Institute of Mental Health [NIMH]
Statistics
2012
 
Omalu
B.
DeKosky
S.
Hamilton
R.
Minster
R.
Kamboh
M.
Shakir
A.
, et al.  . 
Chronic traumatic encephalopathy in a National Football Player: Part II
Neurosurgery
 , 
2006
, vol. 
59
 (pg. 
1086
-
1093
)
Omalu
B.
DeKosky
S.
Minster
R.
Kamboh
M.
Hamilton
R.
Wecht
C.
Chronic traumatic encephalopathy in a National Football League player
Neurosurgery
 , 
2005
, vol. 
57
 (pg. 
128
-
134
)
Pellman
E.
Lovell
M.
Viano
D.
Casson
I.
Concussion in professional football: Recovery of NFL and high school athletes assessed by computerized neuropsychological testing—Part 12
Neurosurgery
 , 
2006
, vol. 
58
 (pg. 
263
-
274
)
Pellman
E.
Lovell
M.
Viano
D.
Casson
I.
Tucker
A.
Concussion in professional football: Neuropsychological testing—Part 6
Neurosurgery
 , 
2004
, vol. 
55
 (pg. 
1290
-
1305
)
Pellman
E. J.
Powell
J. W.
Viano
D. C.
Casson
I. R.
Tucker
A. M.
Feuer
H.
, et al.  . 
Concussion in professional football: Epidemiological features of games injuries and review of the literature—Part 3
Neurosurgery
 , 
2004
, vol. 
54
 (pg. 
81
-
96
)
Reider
B.
Melancholy thoughts
American Journal of Sports Medicine
 , 
2012
, vol. 
40
 (pg. 
2197
-
2199
)
Rencher
A.
Methods of multivariate analysis
 , 
2002
New York
Wiley
Schatz
P.
Long-term test-retest reliability of baseline cognitive assessments using ImPACT
American Journal of Sports Medicine
 , 
2010
, vol. 
38
 (pg. 
47
-
53
)
Schatz
P.
Ferris
C.
One month test–retest reliability of ImPACT (2012)
Archives of Clinical Neuropsychology
 , 
2012
, vol. 
27
 (pg. 
622
-
623
)
Schatz
P.
Moser
R. S.
Solomon
G. S.
Ott
S. D.
Karpf
R.
Prevalence of invalid computerized baseline neurocognitive test results in high school and collegiate athletes
Journal of Athletic Training
 , 
2012
, vol. 
47
 (pg. 
289
-
296
)
Schatz
P.
Pardini
J.E.
Lovell
M.R.
Collins
M.W.
Podell
K.
Sensitivity and specificity of the ImPACT Test Battery for concussion in athletes
Archives of Clinical Neuropsychology
 , 
2006
, vol. 
21
 (pg. 
91
-
99
)
Solomon
G. S.
Haase
R. F.
Biopsychosocial characteristics and neurocognitive test performance in National Football League players: An initial assessment
Archives of Clinical Neuropsychology
 , 
2008
, vol. 
23
 (pg. 
563
-
577
)
Solomon
G. S.
Ott
S. D.
Lovell
M. R.
Long-term neurocognitive dysfunction in sports: What is the evidence?
Clinics in Sports Medicine
 , 
2011
, vol. 
30
 (pg. 
165
-
177
)
Wilens
T.
Faraone
S.
Biederman
J.
Attention-deficit/hyperactivity disorder in adults
Journal of the American Medical Association
 , 
2004
, vol. 
292
 (pg. 
619
-
623
)
Wonderlic
E. F.
Hovland
C. I.
The Personnel Test: A restandardized abridgment of the Otis S-A Test for business and industrial use
Journal of Applied Psychology
 , 
1937
, vol. 
23
 (pg. 
685
-
702
)
Wonderlic Personnel Test, Inc.
Wonderlic Personnel Test and Scholastic Level Exam
 , 
1992
Libertyville, IL
Author