Administration duration for the Wechsler Adult Intelligence Scale-III and Wechsler Memory Scale-III

Transcription

1 Archives of Clinical Neuropsychology 16 (2001) 293±301 Administration duration for the Wechsler Adult Intelligence Scale-III and Wechsler Memory Scale-III Bradley N. Axelrod* Psychology Section (116B), John D. Dingell Department of Veterans Affairs Medical Center, 4646 John R. Street, Detroit, MI , USA Accepted 31 December 1999 Abstract The administration times for each of the subtests from the Wechsler Adult Intelligence Scale-III (WAIS-III) and Wechsler Memory Scale-III (WMS-III) were recorded for a clinical sample of 81 patients. The findings revealed that the time needed to administer the WAIS-III subtests to generate the summary scores, index scores, and both scores were 58, 51, and 65 min, respectively. The time required to complete the primary subtests on the WMS-III was 21, 15, and 6 min for Immediate Memory, General Memory, and Working Memory, respectively, resulting in a total administration time of 42 min. The time necessary to administer most of the subtests was unrelated to age, education, or performance level. These data demonstrate a shorter than expected administration time for the WAIS-III and a longer than anticipated administration for the WMS-III. Results for other clinical settings will be impacted by examiner familiarity and patient composition. D 2001 National Academy of Neuropsychology. Published by Elsevier Science Ltd. Keywords: Administration duration; Wechsler Adult Intelligence Scale-III; Wechsler Memory Scale-III 1. Introduction The Psychological Corporation simultaneously revised, normed, and released the Wechsler Adult Intelligence Scale-III (WAIS-III; Wechsler, 1997b) and Wechsler Memory Scale-III (WMS-III; Wechsler, 1997c) in As with the predecessors of these measures, they have quickly been incorporated into the arsenals for psychological and * Tel.: ext address: bradley.axelrod@med.va.gov (B.N. Axelrod) /01/$ ± see front matter D 2001 National Academy of Neuropsychology. PII: S (00)

2 294 B.N. Axelrod / Archives of Clinical Neuropsychology 16 (2001) 293± 301 neuropsychological assessment. Unfortunately, there has been some discussion in the field regarding the length of administration time for these measures in light of restrictions placed on clinicians by insurance and managed care companies. Specifically, the concern that lengthy administration time of these two tasks might limit the inclusion of additional measures in a neuropsychological evaluation. As a result, psychologists might select other measures of intellectual and memory functioning that are more brief to administer. The WAIS-III is the fifth version of the intelligence scale, which began with the Wechsler±Bellevue (Wechsler, 1939), and replaces the WAIS-Revised (Wechsler, 1981). The WAIS-III is composed of 14 subtests. Eleven of the subtests produce the three summary scores of Verbal IQ, Performance IQ, and Full Scale IQ. The verbal subtests include Vocabulary, Similarities, Arithmetic, Digit Span, Information, and Comprehension. The performance subtests include Picture Completion, Digit Symbol (Coding), Block Design, Matrix Reasoning, and Picture Arrangement. Two additional subtests, Letter±Number Sequencing and Symbol Search, are supplemental and included in index scores. The index scores of Verbal Comprehension (Vocabulary, Similarities, and Information), Perceptual Organization (Picture Completion, Block Design, and Matrix Reasoning), Working Memory (Arithmetic, Digit Span, and Letter±Number Sequencing), and Processing Speed (Digit Symbol and Symbol Search) were derived via factor analytic studies (Wechsler, 1997a). According to the WAIS-III manual (Wechsler, 1997b), administration of the 11 subtests, which comprise the summary scores averages 75 min, with a range of 60 to 90 min. The 11 subtests that comprise the index scores average 60 min, ranging from 45 to 75 min, to administer. The time needed to administer the 13 subtests required to generate all of the summary and index scores is 80 min, with a range of 65 to 95 min. Ryan et al. (1998) quantified administration time for the WAIS-III in a sample of 62 clinical patients who had a mean Full Scale IQ of 94.3 (SD = 11.7). Administration time to generate the summary scores, index scores, and both sets of scores averaged 91, 77, and 100 min, respectively. The results from Ryan et al. demonstrated administration time increases of 22% to 28% relative to the standards reported in the manual, and all times fell just at or above the high end of the range reported in the manual. In comparison to the previous version of the WAIS, administration time for the summary scores on the WAIS-R (Wechsler, 1981) was found to average 69 min for a clinical sample with Full Scale IQ scores in the borderline deficient range (Ward et al., 1987) and 91 min for a clinical sample with an average Full Scale IQ of 91 (Ryan & Rosenberg, 1984). Performance of those clinical cases fell in the high end of the 60 to 90 min range for administration presented in the manual (Wechsler, 1981). The WMS-III (Wechsler, 1997c) is a significant revision over both the Wechsler Memory Scale-Revised (Wechsler, 1987) and the original Wechsler Memory Scale (Wechsler, 1945). The WMS-III has three primary indexes, namely, Immediate Memory, General (i.e., delayed) Memory, and Working Memory. Immediate memory is composed of four subtests which include the immediate recall scores from Logical Memory, Faces, Verbal Paired Associates, and Family Pictures. The General Memory Index uses the delayed results from each of the same four subtests plus Auditory Recognition Delay. Finally, Working Memory

3 B.N. Axelrod / Archives of Clinical Neuropsychology 16 (2001) 293± Table 1 Means and standard deviations of WAIS-III and WMS-III summary and index Variable Mean Standard deviation WAIS-III summary scores Verbal IQ Performance IQ Full Scale IQ WAIS-III index scores Verbal Comprehension Perceptual Organization Working Memory Processing Speed WMS-III index scores Immediate Memory Auditory Immediate Visual Immediate General Memory Auditory Delayed Visual Delayed Auditory Recognition Working Memory is composed of Letter±Number Sequencing and Spatial Span. There are additional subtests, including Orientation, Word List learning, and Visual Reproduction, that are supplemental. The WMS-III manual (Wechsler, 1997c) estimates that actual administration time for the primary subtests is 30 to 35 min, which takes into account the 30-min delay between Logical Memory I and II. The administration of all of the supplemental subtests was estimated to take 15 to 20 min of actual testing time. Together, administration of the complete WMS-III should take 45 to 55 min. The present study sought to provide information regarding the administration time for the primary subtests, summary scores, and index scores for the WAIS-III and WMS-III in a clinical sample. This project will serve as a possible replication of the study by Ryan et al. (1998) vis-a-vis the WAIS-III, but it is a unique opportunity to provide the initial administration time data for the WMS-III. 2. Method 2.1. Participants The sample was composed of 81 consecutively referred patients for neuropsychological evaluation at an urban Department of Veterans Affairs Medical Center. The sample was composed primarily of men (96%) whose average age was 48.7 (SD = 14.1) years and who

4 296 B.N. Axelrod / Archives of Clinical Neuropsychology 16 (2001) 293± 301 Table 2 Means and standard deviations of administration duration for WAIS-III subtests Variable Mean Standard deviation Verbal subtests Vocabulary Similarities Arithmetic Digit Span Information Comprehension Performance subtests Picture Completion Digit Symbol Block Design Matrix Reasoning Picture Arrangement Supplemental subtests Symbol Search Letter ± Number Sequencing had 12.1 (SD = 2.3) years of education. Slightly less than 54% of the sample was White (46% Black) and 88% were right-handed. The mean performance, as demonstrated by summary and index scores, for the WAIS-III and WMS-III appear in Table 1. As can be seen, the mean performance on the WAIS-III summary and index scores fell in the low average range. VIQ scores are significantly higher than PIQ scores (t(80) = 3.6, p = 0.001). Verbal Comprehension was significantly higher than Processing Speed, but no other differences were noted among the index scores (t(80) = 3.2, p = 0.002). On the WMS-III, the sample's score on Immediate Memory fell in the borderline range, whereas delayed General Memory was significantly higher (t(80) = 5.0, p 0.001), falling in the low end of the low average range. Working Memory on the WMS-III was comparable to WAIS-III performance, which was significantly higher than Immediate Memory performance (t(80) = 5.1, p 0.001) Procedure The WAIS-III and WMS-III were administered according to the standardized procedures outlined in their respective manuals (Wechsler, 1997b,c). The subtests were administered in order. Two subtests, Object Assembly from the WAIS-III and Visual Reproduction from the WMS-III were not included, as they were not administered as part of the clinical evaluation. The start and stop times for each subtest from the WAIS-III and WMS-III were recorded at the time of administration. Duration for each subtest was calculated from these times and rounded to the nearest min. Nearly all of the evaluations

5 B.N. Axelrod / Archives of Clinical Neuropsychology 16 (2001) 293± Table 3 Means and standard deviations of administration duration for WMS-III subtests Variable Mean Standard deviation Immediate memory subtests Logical Memory I Faces I Verbal Paired Assoc I Family Pictures I General memory substests Logical Memory I Faces II Verbal Paired Assoc II Family Pictures II Auditory Recognition Delay Working memory subtests Letter ± Number Sequencing (From WAIS-III) Spatial Span Supplemental subtests Orientation Word Lists I Word Lists II were performed by a full-time psychology technician. This Master's Level psychologist had been employed as a neuropsychology technician for 4 years prior to the beginning of this project. The use of an experienced psychometrist was intentional, as it served to examine optimal administration time in a clinical setting. The administration time of technicians, interns, or psychologists new to the WAIS-III and WMS-III would overestimate the actual time required to complete these tasks. 3. Results The mean administration time for each of the WAIS-III and WMS-III subtests appears in Tables 2 and 3, respectively. Administration times for the summary and index scores for both WAIS-III and WMS-III appear in Table 4. On the WAIS-III, the Block Design subtest took the longest to administer; while Picture Arrangement and Comprehension were the next longest subtests to administer. Total administration time of the subtests included in VIQ did not differ from that of PIQ. On the Index scores, Perceptual Organization was the longest factor subtest grouping to administer, while Processing Speed was the shortest. Verbal Comprehension and Working Memory indexes were of comparable time length. The time needed for complete administration of Immediate Memory was approximately 21 min, with the time for auditory and visual tasks evenly split. Administration duration for the

6 298 B.N. Axelrod / Archives of Clinical Neuropsychology 16 (2001) 293± 301 Table 4 Means and standard deviations of administration duration of WAIS-III and WMS-III summary and index scores Variable Mean Standard deviation WAIS-III summary scores Verbal IQ Performance IQ Full Scale IQ WAI-III index scores Verbal Comprehension Perceptual Organization Working Memory Processing Speed WAIS-III summary and index subtests WMS-III index scores Immediate Memory Auditory Immediate Visual Immediate General Memory Auditory Delayed Visual Delayed Auditory Recognition Working Memory WMS-III total of primary subtests primary subtests of the WMS-III averaged 42 min, while the addition of Orientation and Word List learning added almost another 10 min. The frequency distribution for administration times for the primary index and summary scores from the WAIS-III and WMS-III appear in Table 5. (More detailed frequency distributions are available from the author.) As expected, the distribution demonstrates a positive skewing in which longer administration times resulted in a greater time difference from the mean than shorter administration times. For example, the time difference from the mean time for FSIQ is 29 min longer for the 95th percentile, but 15 min shorter for the 5th percentile. The patient characteristics of age, years of education, WAIS-III summary scores, WAIS-III index scores, and WMS-III index scores were correlated with administration time for each of the individual subtests and cluster scores. Given the likelihood of Type I errors when performing over 700 analyses, a p-value of <0.001 was adopted to insure significance. This p-value corresponds to a Pearson correlation of and accounts for 13.4% of the variance. The overall administration time for the WAIS-III was unrelated to test performance. However, Verbal Comprehension and Perceptual Organization index scores were positively correlated with administration duration. Two subtests from the WAIS-III stood out as having significant relationships with performance or patient

7 B.N. Axelrod / Archives of Clinical Neuropsychology 16 (2001) 293± Table 5 Frequency distribution of administration duration (in min) for WAIS-III and WMS-III summary and index scores Percentile Variable 5th 10th 25th 50th 75th 90th 95th WAIS-III summary scores Verbal IQ Performance IQ Full Scale IQ WAIS-III index scores Verbal Comprehension Perceptual Organization Working Memory Processing Speed WAIS-III summary and index subtests WMS-III index scores Immediate Memory General Memory Working Memory WMS-III total of primary subtests characteristics. The time to administer Matrix Reasoning was positively correlated with all of the WAIS-III index and summary scores. This subtest's administration time was also negatively related to age. Digit Span administration time was positively correlated with the Verbal Comprehension index and all three summary scores, as well as the Working Memory index score. The findings for the WMS-III revealed little relationship between administration time and performance. The only outstanding finding was a significant positive correlation between Working Memory administration time and Working Memory performance on the WMS-III. 4. Discussion The findings from the clinical sample presented in this study revealed administration durations of the WAIS-III and WMS-III that differ slightly from expectations and from prior research. On the WAIS-III, administration times of the summary scores, index scores, and total of all 13 included subtests fell lower than the expectations of the test publisher (Wechsler, 1997b). In fact, our administration times all fell at or below the low end of the ranges presented for the WAIS-III. In comparison to the only other study of administration time of the WAIS-III (Ryan et al., 1998), the length of administration was shorter for all subtests. The time to administer Vocabulary, Similarities, and Comprehension for the present

8 300 B.N. Axelrod / Archives of Clinical Neuropsychology 16 (2001) 293± 301 sample was less than half that observed in the previous study. Interestingly, the findings for overall administration time on the WAIS-III from the present sample fell at the opposite end of the time range observed by Ryan et al. (1998). Be that as it may, the findings of the prior and current study seem to agree that the length of administration of the WAIS-III for the computation of the summary scores are comparable to the time it takes when using the WAIS-R. In contrast to the manual's estimation that the WMS-III should take 30 to 35 min to administer (Wechsler, 1997c), the present study revealed times that exceeded the estimation. Of the present clinical sample, 88% took longer than 35 min to complete the primary subtests required for the WMS-III, while only 12% fell within that range. Together, it appears that the WAIS-III and WMS-III, including all of the supplemental tasks, can be administered within at least 2 h. In the current climate of managed care and insurance coverage, time constraints are placed on neuropsychological evaluations. The information from this study might be useful in creating an appropriate assessment battery that maximizes clinical data while meets whatever time limitations exist for an assessment. Furthermore, the data for each of the subtests from the WAIS-III and WMS-III can be used in the evaluation of administration duration for short forms of these scales. One interesting finding was the general lack of relationship between test performance and administration length on the tasks. For most of the subtests, the WAIS-III requires the examiner to discontinue a subtest after a specific number of consecutive failures. Despite the reduced number of items obtained by patients who discontinued early, the overall time to administer the WAIS-III is not impacted by the fewer items. Clinically, we observe patients who perform more poorly on the WAIS-III to take longer to respond to easier items within a subtest than patients who perform overall. It appears that there is an ``evening out'' of time per subtest, as patients of higher functioning respond to more items more quickly than lower functioning individuals. The exceptions to this pattern were that Matrix Reasoning takes longer for patients who perform better, and longer Digit Span administration was related to higher overall performance on a number of measures in which it is included. In contrast to informal opinions that Matrix Reasoning takes a relatively long time to administer, on the average it took one-half of the time that was required to administer Block Design. It is not surprising that administration times in a clinical setting might differ from guidelines presented in the test manuals. First, the times presented for both the WAIS-III and WMS-III are based on nonpatient cases included from the standardization sample. The performance of those individuals may or may not be analogous to patients seen in a clinical setting. More importantly, both the WAIS-III and WMS-III are made of subtests and items that were extracted from longer versions of the standardization tests. The projection of administration times for individual subtests might be affected by the administration of other, and later deleted, subtests that were included for standardization. The generalizablility of the present findings to other clinical samples is contingent on some obvious factors. The relevance of the present administration times is limited by the expertise of the examiner and the examiner's facility with the materials. It is certainly possible that the quality of the technician used in our facility is exceptional and her skills may not be observed in other settings. If, indeed the time of administration is less than that of psychology interns, fellows, and clinicians, the findings would bolster support for the use of dedicated

9 B.N. Axelrod / Archives of Clinical Neuropsychology 16 (2001) 293± psychology technicians in clinical practice. In contrast to the commonly observed finding of the less experienced taking decreasing time to administer a task as familiarity increases (e.g., Axelrod et al., 1994), for the examiner used in this study, there was no relationship between number of times administered and length of administration time. The second disclaimer for this study is that these findings will only apply to similar populations. The clinical utility and practicality of using the complete WAIS-III and WMS-III with all of the supplemental scales will need to be evaluated for each clinical setting. Acknowledgments The author wishes to express his gratitude to Elizabeth Lopez for her painstaking test administration and accurate time keeping for the purpose of this publication. I also thank Rodney D. Vanderploeg for his careful review and thoughtful comments regarding an earlier draft of this manuscript. References Axelrod, B. N., Greve, K. W., & Goldman, R. S. (1994). Comparison of Four Wisconsin Card Sorting Test scoring instructions. Assessment 1, 115±121. Ryan, J. J., Lopez, S. J., & Werth, T. R. (1998). Administration time estimates for WAIS-III subtests, scales, and short forms in a clinical sample. J Psychoeduc Assess 16, 315±323. Ryan, J. J., & Rosenberg, S. J. (1984). Administration time estimates for WAIS-R subtests and short forms in a clinical sample. J Psychoeduc Assess 2, 125±129. Ward, L. C., Selby, R. B., & Clark, B. L. (1987). Subtest administration times and short forms of the Wechsler Adult Intelligence Scale-Revised. J Clin Psychol 18, 276±278. Wechsler, D. (1939). Wechsler ±Bellevue Intelligence Scale. New York: Psychological Corporation. Wechsler, D. (1945). A standardized memory scale for clinical use. J Psychol 19, 87±95. Wechsler, D. (1981). Wechsler Adult Intelligence Scale-Revised. San Antonio, TX: Psychological Corporation. Wechsler, D. (1987). Wechsler Memory Scale-Revised. San Antonio, TX: Psychological Corporation. Wechsler, D. (1997). WAIS-III WMS-III Technical Manual. San Antonio, TX: Psychological Corporation. Wechsler, D. (1997). Wechsler Adult Intelligence Scale (3rd edn.). San Antonio, TX: Psychological Corporation. Wechsler, D. (1997). Wechsler Memory Scale (3rd edn.). San Antonio, TX: Psychological Corporation.