Immediate Post-Concussion Assessment and Cognitive Testing (ImPACT®) (contd.)

Normative Data for High School Students

Initial analyses were based a sample of 545 adolescents between the ages of 13 and 18, inclusive. A portion of these subjects reported some history of education-related problems, such as reading, math, or spelling difficulty; special education placement; or attention-deficit disorder. Athletes with any self-reported history of this nature were compared to those without a self-report history. The groups differed on the Verbal Memory Composite (p < .006; d = .32), Visual Memory Composite (p < .006; d = .31), Processing Speed Composite (p < .002; d = .37), and Reaction Time Composite (p < .045; d = .24). Therefore, subjects with a self-reported history of one or more of these problems were dropped from the normative sample.

The remaining subjects were 341 boys and 83 girls. The girls performed better on the Verbal Memory Composite (p < .01, d = .32), and there was a trend toward better performance on the Processing Speed Composite (p < .055, d = .24). Therefore, the normative data needed to be presented by gender.

The sample of 424 subjects was analyzed for age effects. The breakdown of subjects by age was as follows: 13 = 23, 14 = 122, 15 = 87, 16 = 87, 17 = 61, and 18 = 44. There was a significant main effect for age on the Processing Speed Composite (p < .00001) and the Reaction Time Composite (p < .03). Tukey planned comparisons revealed significantly higher Processing Speed scores for 16, 17, and 18 year olds compared to 13 and 14 year olds. There were no other differences. Tukey planned comparisons revealed no pairwise differences on the Reaction Time Composite. Therefore, the distinction between 13-14 year olds and 16-18 year olds was meaningful.

The sample was sorted into two groups, those between the ages of 13 and 15 and those between the ages of 16 and 18. The 15 year olds were included with the 13-14 year olds because they did not differ from younger or older subjects. The sample was then sorted by gender, and age-group comparisons were run. For the boys, the older subjects (aged 16-18) performed better on the Processing Speed Composite (p < .00001, d = .58), Reaction Time Composite (p < .0009, d = .37), and the Impulse Control Composite (p < .004, d = .32). There were no differences attributable to age among the girls, although the sample sizes, and thus power, were much smaller.

The normative tables are based on 183 boys between the ages of 13 and 15, 158 boys between the ages of 16 and 18, and 83 girls between the ages of 14 and 18. Normative data are based on the natural distributions of scores within these two samples.

The distributions of scores within these groups were examined and exact percentile ranks corresponding to the natural distribution of scores were assigned. Thus, these could be considered uniform percentile ranks. The distributions were not force-normalized, nor were raw scores converted to standard scores.

Sometimes it is useful to know if an athlete performs particularly poorly on a specific subtest. Cutoff scores for the 10th percentile and the 2nd percentiles for 11 scores derived from the 6 subtests are provided in Table 4. This table allows you to identify unusually and abnormally low subtest scores.

Normative Table 5: Boys Ages 13 - 15 (N = 183)