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

Table 19. Classifications, raw scores, and percentile ranks based on a sample of 588 regular education high school boys.

Table 20. Classifications, raw scores, and percentile ranks based on a sample of 119 regular education high school girls.

Table 21. Classifications, raw scores, and percentile ranks based on a sample of 803 regular education university men.

Table 22. Classifications, raw scores, and percentile ranks based on a sample of 236 regular education university women.

Table 23. Classifications, raw scores, and percentile ranks based on a sample of 156 high school boys with a history of "special education"³.

Table 24. Classifications, raw scores, and percentile ranks based on a sample of 31 high school girls with a history of "special education".

Table 25. Classifications, raw scores, and percentile ranks based on a sample of 196 university men with a history of "special education".

Table 26. Classifications, raw scores, and percentile ranks based on a sample of 60 university women with a history of "special education".

The reliable change methodology allows the clinician to reduce the adverse impact of measurement error on test interpretation. To represent clinically significant improvement, the change score must be statistically reliable. However, the converse is not true; a statistically reliable change does not necessarily guarantee a clinically meaningful change. For example, if an athlete demonstrated a major increase in symptoms measured 24 hours post injury, and then obtained a score that showed statistically reliable improvement a few days later, yet the symptom endorsement was still extremely high, this change might not be interpreted as clinically meaningful improvement. In other words, there was real change for the better, but the athlete was still far from recovered.

Using the earlier example of the concussed athletes, the test retest reliability was .80. The standard deviation for time 1 was 24.6 and the standard deviation for time 2 was 12.0. The SEM for time 1 was 11.0 and for time 2 is 5.4. Thus, the Sdiff = 12.2, and the .80 confidence interval = 15.7.

The problem with applying the reliable change methodology to concussed athletes is that their experience of postconcussion symptoms is rapidly changing over a short time period. Thus, the phenomenon under study is not reasonably stable. In the example of the 82 concussed athletes, only 2.4% got worse over time by 10 or more points, whereas 45% got better by 10 or more points. Ten points represents the 90% confidence interval surrounding the time 1 test score in concussed athletes (see Table 18, last column).

Thus, because concussions typically result in a radical change in symptom reporting from baseline, followed by rapid improvement, the reliable change methodology has serious limitations in its practical application.

Postconcussion Testing: Immediately following concussion, athletes often report a large number of symptoms on a postconcussion inventory. There typically is rapid resolution of these symptoms over the next several days, and sometimes weeks. Knowing normal and abnormal symptom score ranges for athletes is helpful for interpreting the clinical significance of the symptom reporting patterns, irrespective of the reliability of the measures.

PDF of Normative Data