Abstract

The purpose of this study is to provide sophisticated psychometric information for advanced interpretation of the Neuropsychological Assessment Battery (NAB) with older adults. This information includes the base rates of low scores, intellectual-cognitive discrepancy scores, and a method for determining change. The NAB contains 24 co-normed neurocognitive tests across five domains (i.e., Attention, Language, Memory, Spatial, and Executive Functions); provides 36 primary T-scores, five domain indexes, and a total index score; and was co-normed with a measure of intellectual abilities (Reynolds Intellectual Assessment Scales; Reynolds Intellectual Screening Test [RIST]). Participants for this study were 742 older adults from the NAB standardization sample (mean age = 68.1, SD = 6.9). From the standardization sample, 42 older adults (mean age = 67.3 years, SD = 8.3) were administered the NAB two times (mean retest interval = 6.7 months, SD = 0.7). The base rates of low index scores and low primary scores are presented for the entire sample, as well as stratified by the level of intellectual abilities. RIST–NAB discrepancy scores are presented for the entire sample and for the different levels of intellectual abilities. Finally, information needed to interpret change in test performance on serial assessments is provided.

Introduction

There is, and will be, increasing pressure to improve our ability to accurately detect mild cognitive impairment (MCI) and prodromal forms of degenerative dementias in older adults. Census information indicates that the number of people over the age of 55 years is steadily increasing over time. In the United States, the number of people 55 years and older increased by over 6 million people between 1990 and 2000 (U.S. Census Bureau, 2001). In Canada, the proportion of the population 55 years and older increased from 20.96% in 1996 to 22.52% in 2001 and to 25.3% in 2006 (Statistics Canada, 1999, 2002, 2007). The Office of National Statistics (2007) estimated that the number of people 55 years and older in the United Kingdom will increase by nearly 4 million by 2020. As increased numbers of people make up the “older adult” group, increased demand will be placed on health-care services that are designated for that segment of the population. In neuropsychology, this will result in a greater need to accurately and efficiently identify cognitive problems.

Neuropsychological theory and practice with older adults has become more sophisticated. In conjunction with the improvement of tests and normative data used for the assessment of cognition, there has also been an improvement in the methods employed for interpreting the performance on those neuropsychological measures. These more advanced approaches to interpreting neuropsychological test data are important for (a) reducing misinterpretation of isolated low test scores that result from normal human variability, (b) decreasing the likelihood that conclusions will be based on clinician bias, and (c) improving the sensitivity and specificity of the measures that we use in clinical and research settings.

The first advanced approach to interpreting neuropsychological data involves considering performance across the entire battery of co-normed tests using a multivariate approach. A thorough assessment of cognitive abilities often involves administering numerous tests and looking for low scores (i.e., deficit measurement). However, clinicians often fail to fully appreciate that healthy people can obtain low test scores and the probability of obtaining “impaired” scores goes up dramatically when multiple tests are given. Information on the prevalence of low scores in healthy adults and older adults on other co-normed batteries of tests has been the focus of several recent studies (seeBinder, Iverson, & Brooks, 2009, for a review).

Iverson and colleagues (Iverson, Brooks, & Holdnack, 2008; Iverson, Brooks, White, & Stern, 2008) have presented the prevalence of low scores in healthy adults on the Neuropsychological Assessment Battery (NAB; Stern & White, 2003a), the Wechsler Adult Intelligence Scale-III (WAIS-III; Wechsler, 1997a), and the Wechsler Memory Scale-III (WMS-III; Wechsler, 1997b). In concordance with other studies on the prevalence of low scores, Iverson and colleagues illustrated that the probability of making a Type I error (i.e., identifying a score as “impaired” when it is not) increases with (a) the more tests that are given and (b) more “lenient” cutoff scores. Moreover, they also illustrated that the likelihood of obtaining low scores “increases dramatically” with lesser intellectual abilities.

It is critical to be informed of the base rates of low scores across a battery of co-normed tests because clinicians administer multiple tests and the results are interpreted in combination. The goal, of course, is to minimize the likelihood of over-pathologizing a low score that is attributable to normal human variability or to measurement error. For example, clinicians and researchers interested in identifying cognitive deficits associated with MCI and early dementia have drawn attention to the prevalence of low memory scores in healthy older adults (Brooks, Iverson, Holdnack, & Feldman, 2008; Brooks, Iverson, & White, 2007; Palmer, Boone, Lesser, & Wohl, 1998) and the implications of over-interpreting low memory scores that might be more in line with normal human variability or measurement error than with an acquired impairment (seede Rotrou et al., 2005, for a discussion of “Accidental MCI”).

The use of psychometrically based information for supplementing clinical interpretation of tests is an important component of neuropsychology. Not all neuropsychological measures provide sophisticated interpretive methods for determining change in serial testing. As a result, clinicians often rely on clinical judgment, which can be subject to bias (seeIverson, Brooks, & Holdnack, 2008, for a review). The purpose of this descriptive paper is to provide an advanced interpretive guide for use with the NAB (Stern & White, 2003a) when assessing the cognition of older adults. Using the NAB normative sample, we provide psychometrically derived and clinically useful tables for: (a) simultaneously interpreting a battery of test scores from multiple cognitive domains (e.g., attention, concentration, and processing speed; receptive and expressive language; learning and memory; visual-spatial skills; and executive functioning); (b) interpreting the discrepancy between intellectual functioning and cognitive abilities; and (c) interpreting change in cognitive abilities when retested over time.

Materials and Methods

Participants

The older adults in this study (N = 742), ranging in age from 55 to 79 years (M = 68.1, SD = 6.9), were selected from the NAB standardization sample. As part of the NAB standardization, a subsample of the healthy older adults (n = 42; mean age = 67.3 years, SD = 8.3) was administered the NAB on two occasions (i.e., participants were administered the same form of the NAB at Time 1 and at Time 2). The mean retest interval was 6.7 months (SD = 0.71). Participants in the oldest age group, between the ages of 80 and 97 years, were not included in this study. Normative data in very old participants can have different distributional characteristics and are more likely, statistically, to be influenced by early forms of degenerative diseases that are not detected through self-report medical screening methods. Moreover, regression-based procedures for developing normative data across the lifespan are more likely to have distributional variability in the tails. Therefore, the oldest age cohort will be analyzed and presented in a separate study.

All of the older adults in this study were healthy, community-dwelling individuals from the United States. Exclusion criteria were employed in order to prevent the inclusion of persons who could have a neurological disease, acquired injury, psychiatric illness, treatment/medication, or physical impairment that would negatively impact test performance. Participants for the standardization sample were recruited and tested at five sites across the country, including Rhode Island Hospital, University of Florida Health Sciences Center, Indiana University, University of California at Los Angeles School of Medicine, and the Psychological Assessment Resources (PAR) offices in Lutz, Florida. These sites were chosen to represent the four geographical regions of the country (Northeast, Midwest, West, and South). (For additional information about the recruitment procedure of the NAB standardization sample and its generalizability to the U.S. population, refer to White and Stern, 2003.)

Measures

The NAB is comprised the following five modules: (a) Attention, (b) Language, (c) Memory, (d) Spatial, and (e) Executive Functions (the free-standing NAB Screening Module was not included in this study). The full NAB consists of 24 individual tests that provide 36 demographically corrected T-scores (mean = 50, SD = 10). (For descriptions of these tests, refer Stern and White, 2003b, and White and Stern, 2003.) Multiple tests comprise the five modules and are used to create five domain index scores and a total index score (mean = 100, SD = 15). Descriptions of the index scores are presented below.

The “Attention Index” is a composite score representing diverse neurocognitive abilities such as attentional capacity, working memory, psychomotor speed, selective attention, divided attention, distractibility, sustained attention, attention to detail, and information processing speed. The “Language Index” is a composite score of overall language functioning. It is based on tests measuring a variety of basic language abilities such as oral production, confrontation naming, auditory comprehension, writing, and bill payment. The “Memory Index” is a composite score for the four learning and memory tests. Tests include list learning, shape learning, story learning, and daily living (medications instructions and name, address, and phone number). Immediate and delayed recall scores are included for each test. The “Spatial Index” score is a composite score of the four tests measuring visual-perceptual skills, attention to visual detail, visual-construction, right–left orientation, map reading, and visual scanning. The “Executive Functions Index” is a composite score based on four tests. Several neurocognitive abilities, including planning, judgment, conceptualization, cognitive response set, mental flexibility, verbal fluency, and generativity, are measured by these tests. The “Total NAB Index” is based on the sum of the five full module index scores. It is meant to be an omnibus measure of neuropsychological functioning in the domains of attention, language, memory, spatial, and executive functions. (For additional information regarding the tests, the domain scores, and the psychometric properties of the NAB, refer to the NAB manuals [Stern & White, 2003b; White & Stern, 2003] or to Iverson, Brooks, White, et al., 2008.)

Intellectual abilities were estimated using the Reynolds Intellectual Screening Test (RIST; Reynolds & Kamphaus, 2003). The RIST is an abbreviated administration of the Reynolds Intellectual Assessment Scales (RIAS). It is normed across the lifespan (i.e., ages 3 through 94). The RIST provides a quick overall estimate of general intellectual abilities. The RIST consists of one verbal (i.e., Guess What, which assesses verbal reasoning, vocabulary, language development, and one's general fund of knowledge) and one nonverbal (i.e., Odd Item Out, which measures nonverbal reasoning, spatial ability, and visual imagery) test. The T-scores of the verbal and nonverbal tests are summed to produce a composite index score (mean 100, SD of 15). The mean RIST composite index score for the older adults, which was used for this study, was 104.4 (SD = 13.0).

The RIST normative sample (N = 2,438) was recruited from 41 states and stratified to represent the United States population based on age, gender, ethnicity, education, and region. Individuals were excluded from the standardized sample if they had potentially confounding factors such as color blindness, uncorrected hearing loss, or visual impairments, were currently involved in drug or alcohol treatment or abused substances, had a history significant for 5 or more minutes of unconsciousness, were currently taking psychiatric medication, or had prior ECT treatment. (For information regarding the reliability and validity of the RIST, refer to the technical manual [Reynolds & Kamphaus, 2003].) The RIST was co-administered to the NAB normative sample. However, discrepancy score comparisons between the RIST and the NAB were not presented in the NAB manuals (Stern & White, 2003b; White & Stern, 2003).

Analyses

Prevalence of low test scores.

The prevalence of low test scores on the NAB in healthy older adults was calculated by “simultaneously” examining test performance across either all five index scores or all 36 primary test scores. The cutoff scores, which were selected to reflect those that might be routinely used in clinical practice or in research, included: (a) below the 16th percentile (i.e., more than 1 SD below the mean), (b) below the 10th percentile, (c) at or below the 5th percentile, and (d) below the 2nd percentile (i.e., more than 2 SD below the mean). For NAB Index scores, these cutoffs correspond to (a) <85, (b) <81, (c) ≤76, and (d) <70, respectively. For the NAB primary test T-scores, these cutoffs correspond to (a) <40, (b) <37, (c) ≤34, and (d) <30, respectively.

The prevalence of low scores was examined for the entire sample of healthy older adults, as well as for various levels of intellectual abilities on the RIST. The sample was stratified into the following groups: (a) low average intellectual abilities (RIST = 80–89); (b) average intellectual abilities (RIST = 90–109); (c) high average intellectual abilities (RIST = 110–119); and (d) superior/very superior intellectual abilities (RIST = 120+).

The number of low scores below each cutoff was examined using frequency distributions. The frequency distributions for the prevalence rates of low index scores are presented as cumulative percentages. Using the frequency distributions for the primary test scores, we identified the number of low test scores found in <2%, 2%–9%, 10%–24%, 25%–75%, 76%–90%, and >90% of healthy older adults.

RIST–NAB discrepancy scores.

Discrepancy scores were calculated by subtracting each one of the NAB Index scores (e.g., Attention, Language, Memory, Spatial, Executive Functions, and Total) from the RIST Index score. The difference scores for each NAB Index score were examined using frequency distributions, with analyses being conducted for both the entire sample and stratified by intellectual abilities. Using the frequency distributions, the following cutoff scores (based on cumulative percentages of the sample) were identified for the discrepancy scores: <20%, <15%, <10%, <5%, and <1%.

Frequency distribution of change scores.

One method for interpreting change in test performance, which simply relies on considering the prevalence of change scores, is to examine the frequency distribution of difference scores and determine various cutoff scores for both decline and improvement. Difference scores (Time 2 − Time 1) were calculated for the five domain indexes, the total index, and the 36 primary T-scores. Frequency distributions were produced for each difference score, and cutoffs for interpreting change in test performance were determined separately for decline and improvement using cumulative percentile ranks corresponding to ≤20th percentile and ≤10th percentile.

Results

Prevalence of Low Test Scores

The base rates of low index scores on the NAB in the older adults sample are presented in Table 1. When the five index scores are considered simultaneously, 35.6% of healthy older adults had one or more scores below the 16th percentile, 22.9% had one or more index scores below the 10th percentile, 12.2% had one or more scores at or below the 5th percentile, and 5.2% had one or more scores below the 2nd percentile.

Table 1.

Prevalence of low NAB Index scores in older adults

Number of index scores below cutoff Total sample Intellectual abilities (RIST)
 
  Low average Average High average Superior/very superior 
<25th percentile 
 5 low scores 3.9 17.9 2.1 – – 
 4 or more 8.8 40.5 6.3 0.6 – 
 3 or more 17.1 59.5 14.5 7.3 2.0 
 2 or more 30.5 75.0 32.5 13.3 8.1 
 1 or more 52.8 85.7 58.6 35.8 27.3 
 0 low scores 47.2 14.3 41.4 64.2 72.7 
<16th percentile 
 5 low scores 1.6 9.5 0.3 – – 
 4 or more 4.7 25.0 2.4 – – 
 3 or more 8.9 39.3 6.1 1.2 1.0 
 2 or more 18.0 56.0 16.7 6.7 4.0 
 1 or more 35.6 79.8 38.9 16.4 13.1 
 0 low scores 64.4 20.2 61.2 83.6 86.9 
<10th percentile 
 5 low scores 1.0 7.1 – – – 
 4 or more 2.5 15.4 0.5 – – 
 3 or more 4.8 28.5 1.3 – 1.0 
 2 or more 10.0 39.2 7.1 1.8 3.0 
 1 or more 22.9 57.1 24.5 8.5 5.0 
 0 low scores 77.1 42.9 75.5 91.5 94.9 
≤5th percentile 
 5 low scores 0.5 3.6 – – – 
 4 or more 1.2 9.6 – – – 
 3 or more 2.7 19.1 0.3 – – 
 2 or more 6.2 31.0 3.5 – 1.0 
 1 or more 12.2 44.1 10.9 1.8 2.0 
 0 low scores 87.8 56.0 89.2 98.2 98.0 
<2nd percentile 
 5 low scores 0.3 1.2 – – – 
 4 or more 0.4 2.4 – – – 
 3 or more 0.5 3.6 – – – 
 2 or more 1.5 9.6 – – – 
 1 or more 5.2 20.3 3.4 0.6 1.0 
 0 low scores 94.8 79.8 96.6 99.4 99.0 
Number of index scores below cutoff Total sample Intellectual abilities (RIST)
 
  Low average Average High average Superior/very superior 
<25th percentile 
 5 low scores 3.9 17.9 2.1 – – 
 4 or more 8.8 40.5 6.3 0.6 – 
 3 or more 17.1 59.5 14.5 7.3 2.0 
 2 or more 30.5 75.0 32.5 13.3 8.1 
 1 or more 52.8 85.7 58.6 35.8 27.3 
 0 low scores 47.2 14.3 41.4 64.2 72.7 
<16th percentile 
 5 low scores 1.6 9.5 0.3 – – 
 4 or more 4.7 25.0 2.4 – – 
 3 or more 8.9 39.3 6.1 1.2 1.0 
 2 or more 18.0 56.0 16.7 6.7 4.0 
 1 or more 35.6 79.8 38.9 16.4 13.1 
 0 low scores 64.4 20.2 61.2 83.6 86.9 
<10th percentile 
 5 low scores 1.0 7.1 – – – 
 4 or more 2.5 15.4 0.5 – – 
 3 or more 4.8 28.5 1.3 – 1.0 
 2 or more 10.0 39.2 7.1 1.8 3.0 
 1 or more 22.9 57.1 24.5 8.5 5.0 
 0 low scores 77.1 42.9 75.5 91.5 94.9 
≤5th percentile 
 5 low scores 0.5 3.6 – – – 
 4 or more 1.2 9.6 – – – 
 3 or more 2.7 19.1 0.3 – – 
 2 or more 6.2 31.0 3.5 – 1.0 
 1 or more 12.2 44.1 10.9 1.8 2.0 
 0 low scores 87.8 56.0 89.2 98.2 98.0 
<2nd percentile 
 5 low scores 0.3 1.2 – – – 
 4 or more 0.4 2.4 – – – 
 3 or more 0.5 3.6 – – – 
 2 or more 1.5 9.6 – – – 
 1 or more 5.2 20.3 3.4 0.6 1.0 
 0 low scores 94.8 79.8 96.6 99.4 99.0 

Notes: RIST = Reynolds Intellectual Screening Test; N = 735 because of some missing values. Values represent cumulative percentages of the sample with low scores. There are slight variations due to rounding. There are 5 index test scores that were considered simultaneously for these analyses. Intellectual abilities are based on the RIST Index and comprise the following scores: Low average, RIST = 80–89 (n = 84); average, RIST = 90–109 (n = 379); high average, RIST = 110–119 (n = 165); superior/very superior, RIST = 120+ (n = 99). Scores for adults with RIST scores <80 are not presented due to small sample sizes. Produced by special permission of the Publisher, Psychological Assessment Resources, Inc., 16204 North Florida Avenue, Lutz, FL 33549, USA, from the standardization data presented in the Neuropsychological Assessment Battery Psychometric and Technical Manual by TW and Robert A. Stern, Ph.D. Copyright 2001, 2003 by PAR, Inc. Further reproduction is prohibited without permission from PAR, Inc.

The base rate of low index scores in older adults decreases systematically as level of intelligence increases. Notice that only 20.2% of older adults with RIST scores in the low average range (i.e., RIST = 80–89) have “no index scores below the 16th percentile” compared with 86.9% of adults with RIST scores in the superior or very superior range (i.e., RIST = 120+). Using a cutoff score of 1 SD, 56% of adults with low average RIST scores have two or more low index scores and 16.7% of adults with average RIST scores (i.e., 90–109) have two or more low index scores. Low index scores were generally uncommon in those with high average, superior, or very superior intellectual abilities.

The prevalence of low primary scores on the NAB in healthy older adults is presented in Table 2. When considering the 36 T-scores simultaneously, it was “common” (i.e., found in 25%–75% of healthy older adults) to obtain between 3 and 7 scores below 1 SD. Similar to the index scores, the number of low test scores decreased with higher levels of intellectual abilities. For example, in healthy older adults with low average intellectual abilities, it was “common” to obtain up to 15 low scores (i.e., <1 SD). However, in those with high average, superior, or very superior intellectual abilities, obtaining 15 low scores <1 SD would be “very uncommon” (i.e., found in fewer than 2% of healthy older adults).

Table 2.

Cumulative percentage of low NAB primary test scores as a function of level of intellectual ability

Groups and cutoff scores Cumulative percentage of sample with number of low primary NAB scores
 
 >90% 76%–90% 25%–75% 10%–24% 2%–9% <2% 
All older adults 
 <25th percentile 0–1 2–3 4–12 13–18 19–25 26+ 
 <16th percentile 1–2 3–7 8–11 12–18 19+ 
 <10th percentile – 1–4 5–7 8–14 15+ 
 ≤5th percentile – 1–3 4–6 7–11 12+ 
 <2nd percentile – – 0–1 3–6 7+ 
Low average intellectual abilities 
 <25th percentile 0–4 5–9 10–21 22–26 27–28 29+ 
 <16th percentile 0–2 3–5 6–15 16–20 21–26 27+ 
 <10th percentile 1–3 4–11 12–14 15–22 23+ 
 ≤5th percentile 1–2 3–8 9–12 13–18 19+ 
 <2nd percentile – – 0–3 4–6 7–11 12+ 
Average intellectual abilities 
 <25th percentile 0–1 2–4 5–12 13–16 17–21 22+ 
 <16th percentile 1–2 3–7 8–10 11–16 17+ 
 <10th percentile – 0–1 2–4 5–7 8–11 12+ 
 ≤5th percentile – 1–3 4–5 6–10 11+ 
 <2nd percentile – – 0–1 3–4 5+ 
High average intellectual abilities 
 <25th percentile 1–2 3–9 10–12 13–18 19+ 
 <16th percentile – 0–1 2–5 6–7 8–12 13+ 
 <10th percentile – 1–3 4–5 6–7 8+ 
 ≤5th percentile – – 0–2 4–6 7+ 
 <2nd percentile – – 0–1 – 3+ 
Superior/very superior intellectual abilities 
 <25th percentile 1–2 3–7 8–11 12–16 17+ 
 <16th percentile – 0–1 2–3 4–6 7–10 11+ 
 <10th percentile – 1–2 4–8 9+ 
 ≤5th percentile – – 0–1 2–3 4–6 7+ 
 <2nd percentile – – 2–3 4+ 
Groups and cutoff scores Cumulative percentage of sample with number of low primary NAB scores
 
 >90% 76%–90% 25%–75% 10%–24% 2%–9% <2% 
All older adults 
 <25th percentile 0–1 2–3 4–12 13–18 19–25 26+ 
 <16th percentile 1–2 3–7 8–11 12–18 19+ 
 <10th percentile – 1–4 5–7 8–14 15+ 
 ≤5th percentile – 1–3 4–6 7–11 12+ 
 <2nd percentile – – 0–1 3–6 7+ 
Low average intellectual abilities 
 <25th percentile 0–4 5–9 10–21 22–26 27–28 29+ 
 <16th percentile 0–2 3–5 6–15 16–20 21–26 27+ 
 <10th percentile 1–3 4–11 12–14 15–22 23+ 
 ≤5th percentile 1–2 3–8 9–12 13–18 19+ 
 <2nd percentile – – 0–3 4–6 7–11 12+ 
Average intellectual abilities 
 <25th percentile 0–1 2–4 5–12 13–16 17–21 22+ 
 <16th percentile 1–2 3–7 8–10 11–16 17+ 
 <10th percentile – 0–1 2–4 5–7 8–11 12+ 
 ≤5th percentile – 1–3 4–5 6–10 11+ 
 <2nd percentile – – 0–1 3–4 5+ 
High average intellectual abilities 
 <25th percentile 1–2 3–9 10–12 13–18 19+ 
 <16th percentile – 0–1 2–5 6–7 8–12 13+ 
 <10th percentile – 1–3 4–5 6–7 8+ 
 ≤5th percentile – – 0–2 4–6 7+ 
 <2nd percentile – – 0–1 – 3+ 
Superior/very superior intellectual abilities 
 <25th percentile 1–2 3–7 8–11 12–16 17+ 
 <16th percentile – 0–1 2–3 4–6 7–10 11+ 
 <10th percentile – 1–2 4–8 9+ 
 ≤5th percentile – – 0–1 2–3 4–6 7+ 
 <2nd percentile – – 2–3 4+ 

Notes: N = 730 because of some missing values. Values represent the number of low NAB subtest scores below various cutoff scores. There are 36 primary T-scores that were considered simultaneously for these analyses. Descriptive categories are based on the prevalence of the sample obtaining the number of low scores below the various cutoff scores. Intellectual abilities are based on the RIST Index and comprise the following scores: Low average, RIST = 80–89 (n = 84); average, RIST = 90–109 (n = 379); high average, RIST = 110–119 (n = 165); superior/very superior, RIST = 120+ (n = 99). Scores for adults with RIST scores <80 are not presented due to small sample sizes. Produced by special permission of the Publisher, Psychological Assessment Resources, Inc., 16204 North Florida Avenue, Lutz, FL 33549, USA, from the standardization data presented in the Neuropsychological Assessment Battery Psychometric and Technical Manual by TW and Robert A. Stern, Ph.D. Copyright 2001, 2003 by PAR, Inc. Further reproduction is prohibited without permission from PAR, Inc.

RIST–NAB Discrepancy Scores

RIST–NAB discrepancy scores are presented in Table 3. In the entire sample, fewer than 20% had NAB Index scores more than 16–18 points below the RIST Index scores. It was rare (<5% of sample) to have NAB Index scores more than ∼26–32 points lower than RIST Index scores. As seen in Table 3, stratifying the sample by the level of intellectual abilities yielded different results. The amount of discrepancy between NAB and RIST scores that would be considered clinically meaningful in healthy older adults increased with higher levels of intellectual abilities. For example, the amount of discrepancy between the RIST and the NAB Memory Index score that was found in fewer than 10% of the sample was as follows: Low average intellectual abilities = 15; average intellectual abilities = 18; high average intellectual abilities = 27; and superior or very superior intellectual abilities = 36.

Table 3.

Commonness of RIST–NAB discrepancy score in older adults as a function of intellectual ability

Groups and discrepancy scores Cumulative percentage of sample with RIST–NAB discrepancy scores
 
 <20% <15% <10% <5% <1% 
All older adults 
 RIST-Attention Index 18–21 22–24 25–31 32–41 42+ 
 RIST-Language Index 17–19 20–23 24–27 28–41 42+ 
 RIST-Memory Index 17–19 20–23 24–28 29–37 38+ 
 RIST-Spatial Index 18–20 21–23 24–28 29–36 37+ 
 RIST-Executive Functions Index 17–19 20–23 24–27 28–35 36+ 
 RIST-Total Index 16–18 19–22 23–25 26–33 34+ 
Low average intellectual abilities 
 RIST-Attention Index 13–14 15–16 17–18 19–23 24+ 
 RIST-Language Index 11 12–14 15–18 19–34 35+ 
 RIST-Memory Index 12–13 14 15 16–19 20+ 
 RIST-Spatial Index 12–13 14–15 16–17 18–23 24+ 
 RIST-Executive Functions Index 11–12 13–14 15–16 17–21 22+ 
 RIST-Total Index 11–13 14–16 17–18 19–26 27+ 
Average intellectual abilities 
 RIST-Attention Index 12–14 15–17 18–21 22–32 33+ 
 RIST-Language Index 12–13 14–16 17–21 22–32 33+ 
 RIST-Memory Index 12–14 15–17 18–19 20–30 31+ 
 RIST-Spatial Index 12–13 14–17 18–21 22–32 33+ 
 RIST-Executive Functions Index 12–15 16–17 18–22 23–29 30+ 
 RIST-Total Index 12–13 14–15 16–21 22–25 26+ 
High average intellectual abilities 
 RIST-Attention Index 24–26 27–29 30–32 33–43 44+ 
 RIST-Language Index 20–22 23–25 26 27–35 36+ 
 RIST-Memory Index 22–24 25–26 27–29 30–40 41+ 
 RIST-Spatial Index 22–25 26–28 29–30 31–36 37+ 
 RIST-Executive Functions Index 21–22 23–26 27–30 31–36 37+ 
 RIST-Total Index 21–22 23–24 25–27 28–34 35+ 
Superior/very superior intellectual abilities 
 RIST-Attention Index 32 33–39 40–44 45–49 50+ 
 RIST-Language Index 34 35–37 38–43 44–62 63+ 
 RIST-Memory Index 30–32 33–35 36–40 41–49 50+ 
 RIST-Spatial Index 28–30 31–34 35–37 38–47 48+ 
 RIST-Executive Functions Index 28–29 30–32 33–36 37–49 50+ 
 RIST-Total Index 26–28 29–30 31–36 37–49 50+ 
Groups and discrepancy scores Cumulative percentage of sample with RIST–NAB discrepancy scores
 
 <20% <15% <10% <5% <1% 
All older adults 
 RIST-Attention Index 18–21 22–24 25–31 32–41 42+ 
 RIST-Language Index 17–19 20–23 24–27 28–41 42+ 
 RIST-Memory Index 17–19 20–23 24–28 29–37 38+ 
 RIST-Spatial Index 18–20 21–23 24–28 29–36 37+ 
 RIST-Executive Functions Index 17–19 20–23 24–27 28–35 36+ 
 RIST-Total Index 16–18 19–22 23–25 26–33 34+ 
Low average intellectual abilities 
 RIST-Attention Index 13–14 15–16 17–18 19–23 24+ 
 RIST-Language Index 11 12–14 15–18 19–34 35+ 
 RIST-Memory Index 12–13 14 15 16–19 20+ 
 RIST-Spatial Index 12–13 14–15 16–17 18–23 24+ 
 RIST-Executive Functions Index 11–12 13–14 15–16 17–21 22+ 
 RIST-Total Index 11–13 14–16 17–18 19–26 27+ 
Average intellectual abilities 
 RIST-Attention Index 12–14 15–17 18–21 22–32 33+ 
 RIST-Language Index 12–13 14–16 17–21 22–32 33+ 
 RIST-Memory Index 12–14 15–17 18–19 20–30 31+ 
 RIST-Spatial Index 12–13 14–17 18–21 22–32 33+ 
 RIST-Executive Functions Index 12–15 16–17 18–22 23–29 30+ 
 RIST-Total Index 12–13 14–15 16–21 22–25 26+ 
High average intellectual abilities 
 RIST-Attention Index 24–26 27–29 30–32 33–43 44+ 
 RIST-Language Index 20–22 23–25 26 27–35 36+ 
 RIST-Memory Index 22–24 25–26 27–29 30–40 41+ 
 RIST-Spatial Index 22–25 26–28 29–30 31–36 37+ 
 RIST-Executive Functions Index 21–22 23–26 27–30 31–36 37+ 
 RIST-Total Index 21–22 23–24 25–27 28–34 35+ 
Superior/very superior intellectual abilities 
 RIST-Attention Index 32 33–39 40–44 45–49 50+ 
 RIST-Language Index 34 35–37 38–43 44–62 63+ 
 RIST-Memory Index 30–32 33–35 36–40 41–49 50+ 
 RIST-Spatial Index 28–30 31–34 35–37 38–47 48+ 
 RIST-Executive Functions Index 28–29 30–32 33–36 37–49 50+ 
 RIST-Total Index 26–28 29–30 31–36 37–49 50+ 

Notes: RIST = Reynolds Intellectual Screening Test; NAB = Neuropsychological Assessment Battery. Values in parentheses represent frequency in sample. The difference scores were created by subtracting the NAB Index score from the RIST score. Total sample size, N = 742. Intellectual abilities are based on the RIST Index and comprise the following scores: Low average, RIST = 80–89 (n = 84); average, RIST = 90–109 (n = 379); high average, RIST = 110–119 (n = 165); superior/very superior, RIST = 120+ (n = 99). Scores for adults with RIST scores <80 are not presented due to small sample sizes. Produced by special permission of the Publisher, Psychological Assessment Resources, Inc., 16204 North Florida Avenue, Lutz, FL 33549, USA, from the Neuropsychological Assessment Battery Psychometric and Technical Manual by TW and Robert A. Stern, Ph.D. Copyright 2001, 2003 by PAR, Inc. Further reproduction is prohibited without permission from PAR, Inc.

Frequency Distribution of Change

The frequency distribution of change scores for determining decline or improvement on the NAB in older adults is presented in Table 4. If uncommon declines are defined as occurring in less than or equal to 20% of healthy older adults, then the cutoff scores for interpreting decline on the indexes are as follows: Attention = 6, Language = 5, Memory = 4, Spatial = 13, Executive Functions = 5, and Total Score = 5. If uncommon improvements are defined as occurring in less than or equal to 20% of healthy adults, then the cutoff scores for interpreting improvement on the indexes are as follows: Attention = 10, Language = 12, Memory = 13, Spatial = 7, Executive Functions = 17, and Total Score = 11.

Table 4.

Interpreting change on the NAB in older adults based on the frequency distribution of difference scores (Time 2 − Time 1)

NAB Indexes and subtests Mean change score Percentage of sample with decline on retest
 
Percentage of sample with improvement on retest
 
  ≤20% ≤10% ≤20% ≤10% 
Indexes 
 Attention Index 2.6 6–7 8+ 10–12 13+ 
 Language Index 4.8 5–10 11+ 12–20 21+ 
 Memory Index 4.8 4–11 12+ 13–17 18+ 
 Spatial Index −3.3 13–18 19+ 7–11 12+ 
 Executive Functions Index 5.4 5–7 8+ 17–20 21+ 
 Total Index 3.7 5–6 7+ 11–15 16+ 
Attention module subtests 
 Digits Forward −0.4 8–12 13+ 8–12 13+ 
 Digits Backward 1.6 5–10 11+ 7–10 11+ 
 Dots 1.9 8–11 12+ 12–15 16+ 
 N & L Part A Speed 0.1 6–7 8+ 6–7 8+ 
 N & L Part A Errors −0.3 8–14 15+ 8–12 13+ 
 N & L Part A Efficiency −0.5 6–7 8+ 6+ 
 N & L Part B Efficiency −0.1 8–12 13+ 5–9 10+ 
 N & L Part C Efficiency 2.8 3–6 7+ 8–11 12+ 
 N & L Part D Efficiency 1.5 6–9 10+ 10–12 13+ 
 N & L Part D Disruption 3.0 6–8 9+ 12–16 17+ 
 Driving Scenes 2.0 7–8 9+ 11–14 15+ 
Language module subtests 
 Oral Production 4.4 6–8 9+ 14–18 19+ 
 Auditory Comprehension 3.2 1–3 4+ 11–18 19+ 
 Naming 0.3 1–10 11+ 2–9 10+ 
 Writing 1.8 3–13 14+ 10–21 22+ 
 Bill Payment -0.8 2–19 20+ 2–11 12+ 
Memory module subtests 
 List Learning A Immediate Recall 1.6 7–14 15+ 11–13 14+ 
 List Learning B Immediate Recall -0.9 9–13 14+ 8–13 14+ 
 List Learning A Short Delayed 1.8 6–9 10+ 10–14 15+ 
 List Learning A Long Delayed 3.3 4–5 6+ 10–13 14+ 
 Shape Learning Immediate Recall 3.9 8+ 12–17 18+ 
 Shape Learning Delayed 1.4 8–14 15+ 13–14 15+ 
 Story Learning Immediate Recall 0.5 8–9 10+ 7–11 12+ 
 Story Learning Delayed 0.5 7–8 9+ 7–13 14+ 
 Daily Living Memory Immediate Recall 4.2 5–9 10+ 12–13 14+ 
 Daily Living Memory Delayed 1.5 6–9 10+ 10–12 13+ 
Spatial Module Subtests 
 Visual Discrimination 1.1 10–19 20+ 8–13 14+ 
 Design Construction 2.5 4–8 9+ 11–13 14+ 
 Figure Drawing Copy −7.8 20–24 25+ 4–10 11+ 
 Figure Drawing Organization −0.6 10–13 14+ 8–9 10+ 
 Figure Drawing Immediate Recall −2.1 9–10 11+ 5–11 12+ 
 Map Reading −0.2 10–14 15+ 7–10 11+ 
Executive functions module subtests 
 Mazes 2.0 6–8 9+ 10–12 13+ 
 Judgment −0.2 12–14 15+ 12–13 14+ 
 Categories 5.9 2–6 7+ 14–16 17+ 
 Word Generation 2.2 5–6 7+ 9–11 12+ 
NAB Indexes and subtests Mean change score Percentage of sample with decline on retest
 
Percentage of sample with improvement on retest
 
  ≤20% ≤10% ≤20% ≤10% 
Indexes 
 Attention Index 2.6 6–7 8+ 10–12 13+ 
 Language Index 4.8 5–10 11+ 12–20 21+ 
 Memory Index 4.8 4–11 12+ 13–17 18+ 
 Spatial Index −3.3 13–18 19+ 7–11 12+ 
 Executive Functions Index 5.4 5–7 8+ 17–20 21+ 
 Total Index 3.7 5–6 7+ 11–15 16+ 
Attention module subtests 
 Digits Forward −0.4 8–12 13+ 8–12 13+ 
 Digits Backward 1.6 5–10 11+ 7–10 11+ 
 Dots 1.9 8–11 12+ 12–15 16+ 
 N & L Part A Speed 0.1 6–7 8+ 6–7 8+ 
 N & L Part A Errors −0.3 8–14 15+ 8–12 13+ 
 N & L Part A Efficiency −0.5 6–7 8+ 6+ 
 N & L Part B Efficiency −0.1 8–12 13+ 5–9 10+ 
 N & L Part C Efficiency 2.8 3–6 7+ 8–11 12+ 
 N & L Part D Efficiency 1.5 6–9 10+ 10–12 13+ 
 N & L Part D Disruption 3.0 6–8 9+ 12–16 17+ 
 Driving Scenes 2.0 7–8 9+ 11–14 15+ 
Language module subtests 
 Oral Production 4.4 6–8 9+ 14–18 19+ 
 Auditory Comprehension 3.2 1–3 4+ 11–18 19+ 
 Naming 0.3 1–10 11+ 2–9 10+ 
 Writing 1.8 3–13 14+ 10–21 22+ 
 Bill Payment -0.8 2–19 20+ 2–11 12+ 
Memory module subtests 
 List Learning A Immediate Recall 1.6 7–14 15+ 11–13 14+ 
 List Learning B Immediate Recall -0.9 9–13 14+ 8–13 14+ 
 List Learning A Short Delayed 1.8 6–9 10+ 10–14 15+ 
 List Learning A Long Delayed 3.3 4–5 6+ 10–13 14+ 
 Shape Learning Immediate Recall 3.9 8+ 12–17 18+ 
 Shape Learning Delayed 1.4 8–14 15+ 13–14 15+ 
 Story Learning Immediate Recall 0.5 8–9 10+ 7–11 12+ 
 Story Learning Delayed 0.5 7–8 9+ 7–13 14+ 
 Daily Living Memory Immediate Recall 4.2 5–9 10+ 12–13 14+ 
 Daily Living Memory Delayed 1.5 6–9 10+ 10–12 13+ 
Spatial Module Subtests 
 Visual Discrimination 1.1 10–19 20+ 8–13 14+ 
 Design Construction 2.5 4–8 9+ 11–13 14+ 
 Figure Drawing Copy −7.8 20–24 25+ 4–10 11+ 
 Figure Drawing Organization −0.6 10–13 14+ 8–9 10+ 
 Figure Drawing Immediate Recall −2.1 9–10 11+ 5–11 12+ 
 Map Reading −0.2 10–14 15+ 7–10 11+ 
Executive functions module subtests 
 Mazes 2.0 6–8 9+ 10–12 13+ 
 Judgment −0.2 12–14 15+ 12–13 14+ 
 Categories 5.9 2–6 7+ 14–16 17+ 
 Word Generation 2.2 5–6 7+ 9–11 12+ 

Notes: NAB = Neuropsychological Assessment Battery. The cutoff scores were derived based on statistical considerations. The frequency distribution of difference scores for each variable was examined, and for the majority of scores the cutoff presented is less than the 20% or 10% percentile rank in each tail of the distribution. Index scores have a mean = 100 and SD = 15. Subtest scores have a mean = 50 and a SD = 10.

Across the 36 primary scores, cutoff scores for uncommon (≤20th percentile) declines were typically from 3 to 14 points on the Attention tests, 1 to 19 points on the Language tests, 4 to 14 points on the Memory tests, 4 to 24 points on the Spatial tests, and 2 to 14 points on the Executive Functions tests. Cutoff scores for uncommon improvements (≤20th percentile) on the Attention tests ranged from 5 to 16 points, on the Language tests ranged from 2 to 21 points, on the Memory tests ranged from 7 to 17 points, on the Spatial tests ranged from 4 to 13 points, and on the Executive Functions tests ranged from 9 to 16 points.

Discussion

The goal of this study was to capitalize on clinical and psychometric research over the past 25 years and to provide advanced psychometric data for interpreting a battery of neuropsychological tests in older adults. The psychometric information provided in this descriptive study extends the interpretive information presented in the NAB Psychometric and Technical Manual (White & Stern, 2003) and supplements other studies that aim to improve the clinician's ability to identify cognitive impairment using individual subtests from the NAB (e.g., Brown et al., 2005; Gavett et al., 2009). The tables presented in this article should facilitate clinical inferences using the NAB regarding isolated and widespread neuropsychological deficits, intelligence-cognition discrepancy scores, and change in cognition over time.

The base rates of low scores in the present study are consistent with the work of Reitan and Wolfson (1985, 1993), Heaton and colleagues (e.g., Heaton, Grant, & Matthews, 1991; Heaton, Miller, Taylor, & Grant 2004), and Schretlen and colleagues (e.g., Diaz-Asper, Schretlen, & Pearlson, 2004; Schretlen, Munro, Anthony, & Pearlson, 2003; Schretlen, Testa, Winicki, Pearlson, & Gordon, 2008). Understanding the base rates of low scores across the NAB is important when interpreting test performance in older adults. This is because neuropsychologists administer numerous tests and then interpret scores in combination, not in isolation. However, little is known about the base rates of low scores in healthy older adults across a battery of tests. Most neuropsychologists use flexible neuropsychological test batteries (Rabin, Barr, & Burton, 2005); the base rates of low scores across these batteries are unknowable unless the same battery is given to a normative sample or a large control sample (e.g., de Rotrou et al., 2005; Palmer et al., 1998). Alternatively, the base rates of low scores across a battery of tests can be estimated for a battery of tests if the intercorrelations for all the tests are known (Crawford, Garthwaite, & Gault, 2007). It is reasonable to assume that the general base rate principles illustrated in Tables 1 and 2 would apply to any fixed or flexible battery (for reviews, seeBinder et al., 2009; Iverson & Brooks, in press).

In this study, the number of low scores on the NAB varied by intelligence. It has been reported repeatedly in the literature that neuropsychological test performance is correlated with education (e.g., Heaton et al., 1991) and with intelligence (e.g., Dodrill, 1997, 1999). People with below average intelligence tend to perform worse on neuropsychological tests than people with average intelligence (e.g., Dodrill, 1999; Horton, 1999; Reitan, 1985), as seen in Tables 1 and 2. However, Tables 1 and 2 also indicate that it would be a mistake to assume that people with superior/very superior intellectual abilities should have no low neuropsychological test scores (i.e., this has been referred to as a “myth” of clinical neuropsychology; Dodrill, 1997, 1999).

Table 2 is very helpful in clinical practice if the entire NAB is administered. This table allows the clinician to make a statement about the patient's performance across the entire test battery. For example, if a 70-year-old man obtained 5 scores below 1 SD from the mean, and he was estimated to have average intelligence (current and premorbid), then his overall profile would be considered “normal” or “average.” Of course, this pertains to the base rate across the entire battery. For example, if those five low scores were all obtained on the Memory Module, then this would not be considered normal (seeBrooks et al., 2007) and might be suggestive of the presence of a memory problem. Given the relative importance of assessing memory in older adults, we suggest that clinicians also consider the prevalence of low scores on the memory module as part of their interpretive process (seeBrooks et al., 2007, for these tables).

Tables 1 and 2 are particularly helpful for identifying cognitive problems in high functioning older adults. Clinicians know that it can be very challenging to identifying a decline in cognitive functioning in well-educated older adults with high average or superior intelligence. Although the NAB normative data are adjusted for education, it is not adjusted for the level of intelligence. As seen in Table 1, having two index scores in the low average range (i.e., <25th percentile) is very uncommon in older adults with premorbid RIST scores in the high average or superior classification ranges. These base rate data might allow the clinician to identify a dementing condition sooner (i.e., when an index score drops below the 25th percentile) in older adults with above average intelligence as opposed to applying more traditional cutoffs (e.g., 1 SD, 5th percentile, or 2 SD).

In clinical practice, we are usually focused on deficit measurement in quest to identify cognitive diminishment or impairment. Tables 1 and 2 also allow the clinician to make a statement about how well, not just how poorly, an examinee performs across an entire battery of tests. Most people have isolated low scores when given a battery of tests. If a person of average intelligence on the RIST has zero or only one primary test score below the 25th percentile, that person's performance “across the entire battery” (i.e., their overall profile) would be considered “very good” (i.e., found in fewer than 10% of healthy older adults; see Table 2). That is, the absence of below average test scores is very uncommon—it suggests that the person performed consistently well across the battery.

Table 3 further illustrates this concept using the discrepancy scores between intellectual (RIST) and cognitive (NAB Index scores) abilities. In persons with low average intellectual abilities, fewer than 20% of the sample had RIST–NAB discrepancy scores of 12 or more points. However, there is nearly a two-fold increase in the amount of discrepancy needed to be “uncommon” for persons with high average intellectual abilities and nearly a three-fold increase in discrepancy needed to be considered “uncommon” for those with superior/very superior intellectual abilities. Although it is true that people with high average or superior intelligence score higher on many (not all) neuropsychological tests, as intelligence increases there appears to be a leveling off in the relation between intelligence and many neuropsychological abilities (e.g., Dodrill, 1999). This leveling off creates a greater spread of test scores in people with superior intelligence, for example, than in people with low average intelligence (as illustrated in Table 3). From a practical perspective, one should expect to see quite a few low neuropsychological test scores in healthy adults with low average intelligence and far fewer (but still some) low test scores in healthy adults with superior intelligence.

The results of the present study are consistent with previous research reporting that healthy adults with high average or superior intelligence are not expected to have comparably high scores across an entire battery of neuropsychological tests. On average, people with high average or superior intellectual abilities perform somewhat lower on neuropsychological tests. Similar findings have been reported for the relation between WAIS-III (Wechsler, 1997a) and WMS-III (Wechsler, 1997b) scores.

A common clinical premise upon which intelligence-cognition discrepancy score analysis is based is that the intelligence score is relatively robust and minimally affected by the clinical condition, whereas other abilities, such as attention and memory, might be compromised. Therefore, when interpreting RIST–NAB discrepancy scores, it is helpful to estimate the patient's premorbid level of intellectual functioning. A patient's current RIST score likely provides a reasonable estimate of premorbid functioning if the person has a relatively minor neurological problem that is unlikely to have a pronounced effect on intelligence. However, if a person has a neurological or psychiatric problem that likely has resulted in a substantial lowering of his or her RIST score, then the discrepancy scores are likely less meaningful.

The RIST–NAB discrepancy score data presented in Table 3 can be used to supplement the clinical interpretation of the NAB. Research is needed to determine if RIST–NAB discrepancies differ between specific clinical groups and the healthy adult population. Moreover, additional research is needed to determine the diagnostic validity of specific RIST–NAB discrepancy scores and the incremental validity of these discrepancy scores over interpreting the index scores alone.

In addition to providing supplemental methods for interpreting test performance based on a single assessment, this study also provided preliminary information to help determine whether change in person's cognitive abilities over time would be considered uncommon. Interpreting change in neurocognitive test performance is an essential component of many neuropsychological evaluations. The determination of change in functioning can be a complex clinical process that involves multiple sources of data. It is also important to stress that retest interpretive methods are meant to supplement, not replace, clinical judgment. Table 4 presents information based on the frequency distributions of change scores (i.e., Time 2 − Time 1). The clinician should be aware that these data are based on a healthy sample, not a clinical sample, and that practice effects over the 6-month interval were present for many tests. The presence of practice effects is why the tails in the difference score distributions are asymmetric (i.e., larger improvement scores are considered uncommon on a test with practice effects). Additional research is needed to better inform our clinical interpretation of serial testing in clinical samples and in samples tested more than twice over longer retest intervals.

There are some limitations to this study. First, despite a strong correlation between the RIST and the WAIS-III (r = .75), it is important to recognize that RIST intelligence scores might not necessarily translate into the same scores on other tests of intelligence. For example, Umphress (2008) reported that scores on the RIAS (Reynolds & Kamphaus, 2003) in a small sample of 20 people with low intelligence (i.e., FSIQ scores were below 80) were on average five IQ points higher compared with the WAIS-III (Wechsler, 1997a). The overall difference appeared to be mostly attributable to differences in nonverbal intelligence. Therefore, in order to confidently and precisely use the RIST-stratified interpretation tables, the clinician must administer the RIST (not substitute another intelligence test battery). Another limitation pertains to the sample size of the retest sample (n = 42). Unfortunately, the older adult retest sample was part of the larger NAB retest sample (n = 95) that also contained adults between 18 and 54 years of age. The initial intent of the retest sample was not to divide it into age groups. Although this relatively small sample size limits the types of analyses that can be conducted and potentially limits the generalizability of the findings, this paper still represents the first study on interpreting change on the NAB in older adults. Further, the retest data involved some participants receiving Form 1 and some participants receiving Form 2 of the NAB. However, it is important to note that the same form was given at Time 1 and Time 2.

The results presented in this descriptive study illustrate psychometrically what many practitioners already know clinically—some low scores can be common. This information on the base rates of low scores has important implications for clinicians using fixed or flexible batteries. Several studies have illustrated these implications using batteries of co-normed memory measures and have placed the results in the context of assessing for MCI (Brooks et al., 2007, 2008; de Rotrou et al., 2005; Palmer et al., 1998). For example, when considering several scores from a battery of memory tests, nearly one out of three “healthy” older adults will have an isolated low memory score and will meet psychometric criteria for MCI. However, when considering the prevalence of low memory scores in those with lesser intellectual abilities, the percentage of healthy older adults who would meet psychometric criteria for MCI increases to over 50% (Brooks et al., 2007, 2008). Additional research is needed to improve, psychometrically, our methods for identifying MCI in adults across the lifespan. The tables presented here should facilitate clinical inferences regarded isolated and widespread neuropsychological deficits on the NAB, intelligence-cognition discrepancy scores, and change in functioning over time in evaluations of older adults.

Conflict of Interest

BLB has no known, perceived, or actual conflict of interest with this research. Research funding has been provided to GLI from Psychological Assessment Resources, Inc. for other NAB research projects (this project was, however, unfunded). TW is the Vice President of Research and Development with Psychological Assessment Resources, Inc.

Acknowledgements

The authors thank Ms. Jennifer Bernardo for assistance with manuscript preparation.

References

Binder
L. M.
Iverson
G. L.
Brooks
B. L.
To Err is human: “Abnormal” neuropsychological scores and variability are common in healthy adults
Archives of Clinical Neuropsychology
 , 
2009
, vol. 
24
 (pg. 
31
-
46
)
Brooks
B. L.
Iverson
G. L.
Holdnack
J. A.
Feldman
H. H.
The potential for misclassification of mild cognitive impairment: A study of memory scores on the Wechsler Memory Scale-III in healthy older adults
Journal of the International Neuropsychological Society
 , 
2008
, vol. 
14
 
3
(pg. 
463
-
478
)
Brooks
B. L.
Iverson
G. L.
White
T.
Substantial risk of “Accidental MCI” in healthy older adults: Base rates of low memory scores in neuropsychological assessment
Journal of the International Neuropsychological Society
 , 
2007
, vol. 
13
 
3
(pg. 
490
-
500
)
Brown
L. B.
Stern
R. A.
Cahn-Weiner
D. A.
Rogers
B.
Messer
M. A.
Lannon
M. C.
, et al.  . 
Driving scenes test of the Neuropsychological Assessment Battery (NAB) and on-road driving performance in aging and very mild dementia
Archives of Clinical Neuropsychology
 , 
2005
, vol. 
20
 (pg. 
2090
-
2215
)
Crawford
J. R.
Garthwaite
P. H.
Gault
C. B.
Estimating the percentage of the population with abnormally low scores (or abnormally large score differences) on standardized neuropsychological test batteries: A generic method with applications
Neuropsychology
 , 
2007
, vol. 
21
 
4
(pg. 
419
-
430
)
de Rotrou
J.
Wenisch
E.
Chausson
C.
Dray
F.
Faucounau
V.
Rigaud
A. S.
Accidental MCI in healthy subjects: A prospective longitudinal study
European Journal of Neurology
 , 
2005
, vol. 
12
 
11
(pg. 
879
-
885
)
Diaz-Asper
C. M.
Schretlen
D. J.
Pearlson
G. D.
How well does IQ predict neuropsychological test performance in normal adults?
Journal of the International Neuropsychological Society
 , 
2004
, vol. 
10
 
1
(pg. 
82
-
90
)
Dodrill
C. B.
Myths of neuropsychology
The Clinical Neuropsychologist
 , 
1997
, vol. 
11
 (pg. 
1
-
17
)
Dodrill
C. B.
Myths of neuropsychology: Further considerations
The Clinical Neuropsychologist
 , 
1999
, vol. 
13
 
4
(pg. 
562
-
572
)
Gavett
B. E.
Poon
S. J.
Ozonoff
A.
Jefferson
A. L.
Nair
A. K.
Green
R. C.
, et al.  . 
Diagnostic utility of the NAB List Learning test in Alzheimer's disease and amnestic mild cognitive impairment
Journal of the International Neuropsychological Society
 , 
2009
, vol. 
15
 (pg. 
121
-
129
)
Heaton
R. K.
Grant
I.
Matthews
C. G.
Comprehensive norms for an extended Halstead–Reitan Battery: Demographic corrections, research findings, and clinical applications
 , 
1991
Odessa, FL
Psychological Assessment Resources, Inc
Heaton
R. K.
Miller
S. W.
Taylor
M. J.
Grant
I.
Revised comprehensive norms for an expanded Halstead–Reitan Battery: Demographically adjusted neuropsychological norms for African American and Caucasian adults professional manual
 , 
2004
Lutz, FL
Psychological Assessment Resources, Inc
Horton
A. M.
Jr
Above-average intelligence and neuropsychological test score performance
International Journal of Neuroscience
 , 
1999
, vol. 
99
 
1–4
(pg. 
221
-
231
)
Iverson
G. L.
Brooks
B. L.
Schoenberg
M. R.
Scott
J. G.
Improving accuracy for identifying cognitive impairment
The black book of neuropsychology: A syndrome-based approach
 
New York
Springer Publishing Inc
 
(in press)
Iverson
G. L.
Brooks
B. L.
Holdnack
J. A.
Heilbronner
R. L.
Misdiagnosis of cognitive impairment in forensic neuropsychology
Neuropsychology in the courtroom: Expert analysis of reports and testimony
 , 
2008
New York
Guilford Press
(pg. 
243
-
266
)
Iverson
G. L.
Brooks
B. L.
White
T.
Stern
R. A.
Horton
A. M.
Jr.
Wedding
D.
Neuropsychological Assessment Battery (NAB): Introduction and advanced interpretation
The neuropsychology handbook
 , 
2008
3rd ed.
New York
Springer Publishing Inc
(pg. 
279
-
343
)
Office of National Statistics
Age structure of England and Whales
 , 
2007
 
Palmer
B. W.
Boone
K. B.
Lesser
I. M.
Wohl
M. A.
Base rates of “impaired” neuropsychological test performance among healthy older adults
Archives of Clinical Neuropsychology
 , 
1998
, vol. 
13
 
6
(pg. 
503
-
511
)
Rabin
L. A.
Barr
W. B.
Burton
L. A.
Assessment practices of clinical neuropsychologists in the United States and Canada: A survey of INS, NAN, and APA Division 40 members
Archives of Clinical Neuropsychology
 , 
2005
, vol. 
20
 
1
(pg. 
33
-
65
)
Reitan
R. M.
Relationships between measures of brain functions and general intelligence
Journal of Clinical Psychology
 , 
1985
, vol. 
41
 
2
(pg. 
245
-
253
)
Reitan
R. M.
Wolfson
D.
The Halstead–Reitan neuropsychological test battery: Theory and clinical interpretation
 , 
1985
Tucson, AZ
Neuropsychology Press
Reitan
R. M.
Wolfson
D.
The Halstead–Reitan neuropsychological test battery: Theory and clinical interpretation
 , 
1993
2nd ed.
Tucson, AZ
Neuropsychology Press
Reynolds
C. R.
Kamphaus
R. W.
Reynolds Intellectual Assessment Scales and Reynolds Intellectual Screening Test professional manual
 , 
2003
Lutz, FL
Psychological Assessment Resources
Schretlen
D. J.
Munro
C. A.
Anthony
J. C.
Pearlson
G. D.
Examining the range of normal intraindividual variability in neuropsychological test performance
Journal of the International Neuropsychological Society
 , 
2003
, vol. 
9
 
6
(pg. 
864
-
870
)
Schretlen
D. J.
Testa
S. M.
Winicki
J. M.
Pearlson
G. D.
Gordon
B.
Frequency and bases of abnormal performance by healthy adults on neuropsychological testing
Journal of the International Neuropsychological Society
 , 
2008
, vol. 
14
 
3
(pg. 
436
-
445
)
Statistics Canada
1996 Census Catalogue—Final Edition (Catalogue No. 92-350-UIE)
 , 
1999
 
Statistics Canada
Age (122) and Sex (3) for Population, for Canada, Provinces, Territories, Census Divisions, Census Subdivisions and Dissemination Areas, 2001 Census—100% Data (Catalogue No. 95F0300XCB2001001)
 , 
2002
 
Statistics Canada
Age Groups (13) and Sex (3) for the Population of Canada, Provinces and Territories, 1921 to 2006 Censuses—100% Data (Catalogue number: 97-551-X2006005)
 , 
2007
 
Stern
R. A.
White
T.
Neuropsychological assessment battery
 , 
2003
Lutz, FL
Psychological Assessment Resources
Stern
R. A.
White
T.
Neuropsychological assessment battery: Administration, scoring, and interpretation manual
 , 
2003
Lutz, FL
Psychological Assessment Resources
Umphress
T. B.
A comparison of low IQ scores from the Reynolds Intellectual Assessment Scales and the Wechsler Adult Intelligence Scale-Third Edition
Intellectual and Developmental Disabilities
 , 
2008
, vol. 
46
 (pg. 
229
-
233
)
United States Census Bureau
Age: 2000 (C2KBR/01-12)
 , 
2001
 
Wechsler
D.
Wechsler Adult Intelligence Scale—Third edition
 , 
1997
San Antonio, TX
The Psychological Corporation
Wechsler
D.
Wechsler Memory Scale—Third edition
 , 
1997
San Antonio, TX
The Psychological Corporation
White
T.
Stern
R. A.
Neuropsychological assessment battery: Psychometric and technical manual
 , 
2003
Lutz, FL
Psychological Assessment Resources

Author notes

Tables 1–4 contain original new data produced by special permission of the Publisher, Psychological Assessment Resources, Inc. (PAR, Inc.), 16204 North Florida Avenue, Lutz, FL 33549, from the standardization data presented in the Neuropsychological Assessment Battery Psychometric and Technical Manual by TW and Robert A. Stern, Ph.D. Copyright 2001, 2003 by PAR, Inc. Further reproduction is prohibited without permission from PAR, Inc. Portions of this study were presented at the National Academy of Neuropsychology (2006) and International Neuropsychological Society (2007) conferences.