The Derived Scores for ImPACT Version 4

In addition to the individual subtest scores, Composite Scores were developed by combining specific subtest scores that were selected by the test authors a priori. Composite scores were derived logically rather than through factor analysis and were designed to provide summary level information to facilitate interpretation of test taker performance by the healthcare provider. The Composite Scores provide a summary score that represents unweighted average raw scores for the component subtests. Scores are expressed as percentiles for each input device (computer mouse and trackpad) separately relative to individuals of their own age group and sex. Separate normative data was calculated for the computer mouse (n=38,095) and trackpad (n=33,720). Percentile ranks were created by using a standard computer software conversion program and any resulting irregularities were identified and smoothed manually. It should be noted that no trackpad normative data was collected for individuals aged 60-80. Thus, when testing individuals in this age group, the test should only be administered on a laptop or desktop with a computer mouse. There are four Composite Scores, also referred to as Core Composite Scores, (Verbal Memory Composite, Visual Memory Composite, Visual Motor Speed Composite, and Reaction Time), and two supplemental scores, Impulse Control Composite and Total Symptom Score.

New to ImPACT Version 4 is the option to obtain a Two-Factor Score in addition to the existing Composite Scores. Both the Composite Scores (Schatz, Pardini, Lovell, Collins, Podell, 2006; Schatz & Sandel, 2013) and the Two-Factor Scores (Schatz & Maerlender, 2013; Gerrard, Iverson, Atkins, et al 2017) have been validated in prior independent research.

Verbal Memory Composite#

Evaluates attentional processes, learning, and memory within the verbal domain This composite score represents the average performance on:

• Word Memory (Module 1) Total Percent Correct
• Symbol Match (Module 4) Total Correct Hidden/9*100
• Three Letters (Module 6) Percent Total Letters Correct

Example#

See tables below

Word Memory Total Percent Correct = 98%
Symbol Match (Total Correct Hidden)/9100 = 6/9100 = 66.7%
Three Letters Percent Total Letters Correct = 80%

Total Divided by 3 = 244.7/3 = 82

Verbal Memory Composite Score= 82

WORD MEMORY	Baseline	Post Injury
Hits (Immediate)	12	11
Correct Distractors (immed.)	12	7
Learning percent correct	100%	75%
Hits (delay)	11	8
Correct distractors (delay)	12	6
Delayed memory pct. correct	96%	58%
Total percent correct	98%	66.5%

SYMBOL MATCH	Baseline	Post Injury
Total correct (visible)	27	27
Avg. correct RT (visible)	1.52	1.85
Total correct (hidden)	6	4
Avg. correct RT (hidden)	1.94	2.79

THREE LETTERS	Baseline	Post Injury
Total sequence correct	3	1
Total letters correct	12	9
Pct. of total letters correct	80%	60%
Avg. time to first click	2.26	3.15
Avg. counted	12.8	9.6
Avg. counted correctly	12.8	9.6

Visual Memory Composite#

Evaluates visual attention and scanning, learning, and memory This score in its current form is comprised of the average of:

• Design Memory (module 2) Total Percent Correct
• X’s and O’s (module 3) (Total Correct Memory)/12*100

Example#

See tables below

Design Memory Total Percent Correct = 90%
X’s and O’s (Total Correct Memory)/12= 8/12*100 = 66.7%

Total Divided by 2 = 156.7/2 = 78

Visual Memory Composite Score= 78

DESIGN MEMORY	Baseline	Post Injury
Hits (Immediate)	12	10
Correct Distractors (immed.)	10	5
Learning percent correct	92%	63%
Hits (delay)	12	4
Correct distractors (delay)	9	7
Delayed memory pct. correct	88%	46%
Total percent correct	90%	54.5%

X’S AND O’S	Baseline	Post Injury
Total correct (memory)	8	5
Total correct (interference)	109	67
Avg. correct RT (interference)	0.55	1.17
Total correct (interference)	3	3
Avg. incorrect RT (interference)	0.43	1.04

Visual Motor Speed Composite#

Evaluates visual processing, learning and memory, and visual-motor response speed. This score is comprised of the average of following scores:

• Total Number Correct/4 during Interference of X’s and O’s (module 3).
• Average Counted Correctly*3 from Countdown Phase of Three Letters (module 6).

Example#

See tables below

Total Number Correct/4 during Interference of X’s and O’s= 109/4 = 27.25
Average Counted Correctly x 3 from Countdown Phase of Three Letters = 12.8x3 = 38.40

Total Divided by 2 = 65.65/2 = 32.83

Visual Motor Speed Composite Score= 32.83

X’S AND O’S	Baseline	Post Injury
Total correct (memory)	8	5
Total correct (interference)	109	67
Avg. correct RT (interference)	0.55	1.17
Total correct (interference)	3	3
Avg. incorrect RT (interference)	0.43	1.04

SYMBOL MATCH	Baseline	Post Injury
Total correct (visible)	27	27
Avg. correct RT (visible)	1.52	1.85
Total correct (hidden)	6	4
Avg. correct RT (hidden)	1.94	2.79

THREE LETTERS	Baseline	Post Injury
Total sequence correct	3	1
Total letters correct	12	9
Pct. of total letters correct	80%	60%
Avg. time to first click	2.26	3.15
Avg. counted	12.8	9.6
Avg. counted correctly	12.8	9.6

Reaction Time Composite#

Evaluates average response speed.
This score is comprised of the average of the following scores:

• Average Correct RT of Interference Stage of X’s and O’s (module 3).
• Symbol Match (module 4) Average Correct RT Visible/3.
• Color Match (module 5) Average Correct RT.

Example#

See tables below

Average Correct RT of Interference Stage of X’s and O’s = 0.55
Symbol Match Average Correct RT Visible/3 = 1.52/3 = 0.51
Color Match Average Correct RT = 0.76

Total Divided by 3 = 1.82/3= 0.61

Reaction Time Composite Score= 0.61

X’S AND O’S	Baseline	Post Injury
Total correct (memory)	8	5
Total correct (interference)	109	67
Avg. correct RT (interference)	0.55	1.17
Total correct (interference)	3	3
Avg. incorrect RT (interference)	0.43	1.04

SYMBOL MATCH	Baseline	Post Injury
Total correct (visible)	27	27
Avg. correct RT (visible)	1.52	1.85
Total correct (hidden)	6	4
Avg. correct RT (hidden)	1.94	2.79

COLOR MATCH	Baseline	Post Injury
Total correct	9	9
Avg. correct RT	0.76	1.58
Total commissions	0	1
Avg. commissions RT	0	1.31

Impulse Control Composite#

While not one of the core composites for ImPACT scoring, this composite provides a measure of errors on testing and is useful in determining test validity. This score indicates the sum of errors committed during different phases of the test, and while clinical decisions should not be based on this composite, its inclusion may help in the interpretation of other composites. Scores above 30 should be viewed as invalid.
This score is obtained by adding:

• Total Incorrect on the Interference Phase of X’s and O’s (module 3).
• Color Match Total Commissions (module 5).

Example#

See tables below

Total Incorrect on the Interference Phase of X’s and O’s = 3
Color Match Total Commissions = 0
Total= 3

Impulse Control Composite Score= 3

X’S AND O’S	Baseline	Post Injury
Total correct (memory)	8	5
Total correct (interference)	109	67
Avg. correct RT (interference)	0.55	1.17
Total correct (interference)	3	3
Avg. incorrect RT (interference)	0.43	1.04

COLOR MATCH	Baseline	Post Injury
Total correct	9	9
Avg. correct RT	0.76	1.58
Total commissions	0	1
Avg. commissions RT	0	1.31

Total Symptom Score#

This score presents summary information regarding the individual’s self-reported symptom data. A higher score reflects a higher symptom total.

Two-Factor Score#

A new scoring approach, called a Two-Factor Score comprised of a Memory component and a Speed component was independently developed and validated for ImPACT (Schatz and Maerlander, 2013) and, will be provided as an additional option to the existing Composite Scores approach provided in the predicate device. The Two-Factor Scores are calculated as Z-scores (subtracting the user's score from the group mean and dividing by the standard deviation: (X - Mean)/SD) and are presented as percentile ranks. The Two Factor Score is calculated using the raw score mean and standard deviation (based on age and gender) as described below.

Speed Composite#

LBL-01_v14_ImPACT-Version-4_Administration_Manual_source

The Speed Composite is comprised of the Motor Speed and Reaction Time and is calculated by converting the Motor Speed composite score into a Z-score by subtracting the age- and gender-appropriate mean from the test-taker's raw score and dividing by the standard deviation. Similarly, the Reaction Time composite score is also converted into a Z-score, and the Speed Composite is calculated by subtracting the Motor Speed and Reaction Time Z-scores and dividing by 2.

Memory Composite#

LBL-01_v14_ImPACT-Version-4_Administration_Manual_source

The Memory Composite is comprised of the Visual Memory and Verbal Memory and is calculated by converting the Verbal Memory composite score into a Z-score by subtracting the age- and gender-appropriate mean from the test-taker's raw score and dividing by the standard deviation. Similarly, the Visual Memory composite score is also converted into a Z-score, and the Memory Composite is calculated by summing the Verbal Memory and Visual Memory Z-scores, and dividing by 2.

The Two-Factor Score (speed and memory) provides an additional summary score approach for interpretation of the test data. Research has shown that in addition to the existing four Composite Scores that have historically been used to describe the ImPACT test results, a Two-Factor Score can also provide a valid approach to interpreting the data. In addition, this research has shown that the two-factor structure has improved test-retest reliability with no loss of sensitivity/specificity and may improve understanding and interpretability of ImPACT test results. This increased reliability is likely a result of more items included in each factor, thus possibly reducing error.

Over the years there has been some debate as to the most parsimonious explanation of the constructs underlying ImPACT. As described in the device labeling, the authors of the test used an a priori, logical method of assigning subtests to 4 Composite Scores (i.e., Visual Memory, Verbal Memory, Reaction Time and Processing Speed). Subsequent research has generally supported that structure; however, there appears to be evidence for shared variance among Processing Speed and Reaction Time scores as well as Verbal Memory and Visual Memory scores. Given those findings, it made sense to offer a two-factor interpretive solution as an additional method of interpreting test results. The valid clinical association of providing a Two-Factor Score is supported in published literature by Schatz and Maerlender (2013), and Gerrard, Iverson, Atkins (2017).