Immediate Post-Concussion Assessment and Cognitive Testing (ImPACT®)
Normative Data: for Foxpro (Original Platform) Version of ImPACT 2003 (For Reference Only)
| Grant L. Iverson, Ph D. | |
| University of British Columbia & Riverview Hospital | |
| Mark R. Lovell, Ph.D. | Michael W. Collins, Ph.D. |
| University of Pittsburgh Medical Center | University of Pittsburgh Medical Center |
Address correspondence to: Mark Lovell, Ph.D., UPMC Sports Concussion Program; Department of Orthopaedic Surgery;
Center for Sports Medicine; 3200 South Water Street, Pittsburgh, PA 15203.
Purpose
The purpose of this handout is to provide normative data for ImPACT Version 2.0 (Immediate Post-Concussion Assessment and Cognitive Testing). ImPACT is a computerized neuropsychological test battery developed specifically for the evaluation of sports concussion.
Description of Test
ImPACT is a computer administered neuropsychological test battery that consists of 6 individual test modules that measure aspects of cognitive functioning including attention, memory, reaction time, and processing speed (see Table 1). Some of these test modules have two distinct subtests that measure different cognitive functions (e.g., working memory and processing speed).
Table 1. ImPACT Neuropsychological Test Modules.
| Test Module | Ability Areas |
| Word Memory | Immediate and delayed memory for words |
| Design Memory | Immediate and delayed memory for designs |
| X's and O's | Attention, concentration, working memory, reaction time |
| Symbol Match | Visual processing speed, learning and memory |
| Color Match | Focused attention, response inhibition, reaction time |
| Three Letters | Attention, concentration, working memory, visual-motor speed |
| Results from above tests are computed into composite scores. | |
| Computation of Composite Scores | |
Verbal Memory Composite Score Average of these scores:
Visual Memory Composite Score Average of these scores:
|
Reaction Time Composite Score Average of these scores:
Processing Speed Composite Score Average of the following scores:
Impulse Control Composite Score (experimental; not normed yet) Sum of the following scores:
Postconcussion Scale: Total Score |
Conceptualizing Normative Scores
The profession of clinical neuropsychology has a long history of over-pathologizing test scores. The most obvious and pervasive example is the use of the term "impaired." It is extremely common for researchers to state that a specific group of patients has impaired cognitive abilities because, as a group, they had statistically lower scores than a group of control subjects. This often occurs when the effect sizes for these differences are small or modest. Moreover, it is frequently the case that the mean scores for the patient group on various neuropsychological tests, although lower than the control group, still fall in the average or low average classification range; thus, they represent a presumed lowering, decline, diminishment, or decrement in performance, but not an impairment.
Although it can be argued that the term impairment simply refers to a negative change in function, for most people the term carries much more serious connotations. This is a particularly important issue when working with people who have sustained mild injuries or disease processes that could have affected their brains. Neuropsychologists must guard against iatrogenesis (i.e., health care providers making the problem worse). It is quite possible that by over-pathologizing test scores, the health care provider can inadvertently make the patient worse. Focusing, dwelling, and worrying about symptoms and "brain damage" can magnify them and protract the recovery period. Having stated this, it is important to accurately detect change that has occurred, and to determine whether this is a statistically and clinically meaningful change.
A basic conceptualization of initial level of performance is provided below. Standardized tests yield scores that fall within certain classification ranges. The following classification ranges and their corresponding percentile rank ranges are commonly used, although not universally accepted: Mildly Impaired < 2nd percentile; Borderline 3rd - 9th percentile; Low Average 10th - 24th percentile; Average 25th - 75th percentile; High Average 76th - 90th percentile; Superior 91st - 98th; Very Superior > 99th percentile. Thus, if an individual obtained a score at the 42nd percentile, this would mean that his performance would be greater than or equal to 42% of his same-aged peers in the general population, and that his score would fall in the Average classification range.
Different normative scores and their corresponding descriptors (i.e., their classification ranges) are illustrated in Table 2. It is important to note that there is not precise agreement in our profession as to where exactly the cutoffs should fall between certain classification ranges (e.g., some may call a percentile rank of 9 low average instead of borderline, because it corresponds to an IQ of 80). There is also disagreement as to the three "impaired" classification ranges. The system below is similar to the more traditional IQ classifications corresponding to mild, moderate, and severe mental retardation.
Table 2. Normative scores and classification ranges in neuropsychology
| Descriptor / Classification Range |
Scaled Scores M=10, SD=3 |
IQs/Index Scores M=100, SD=15 |
T-Score M=50, SD=10 |
Percentile Rank |
| Severely Impaired | <1 | <55 | <20 | <.13 |
| Moderately Impaired | 1 | 55-59 | 20-23 | .13 - .35 |
| Mildly Impaired | 2 - 4 | 60 - 69 | 24 - 29 | .38 - 1.9 |
| Borderline | 5 - 6 | 70 - 79 | 30 - 36 | 2 - 9 |
| Low Average | 7 | 80 - 89 | 37 - 43 | 10 - 24 |
| Average | 8 - 12 | 90 - 109 | 44 - 56 | 25 - 75 |
| High Average | 13 | 110 - 119 | 57 - 63 | 76 - 90 |
| Superior | 14 - 15 | 120 - 129 | 64 - 69 | 91 - 97 |
| Very Superior | 16 - 19 | 130+ | 70+ | 98+ |
(M = Mean (average), SD = Standard deviation)