|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 26, No 25 (September 1), 2008: pp. 4138-4143 © 2008 American Society of Clinical Oncology. DOI: 10.1200/JCO.2008.16.8864 Screening for Neurocognitive Impairment in Pediatric Cancer Long-Term Survivors
From the Department of Child Psychology, Texas Children's Hospital; Department of Pediatrics, Baylor College of Medicine; Texas Children's Cancer Center, Houston, TX; and Division of AIDS and Health and Behavior Research, National Institute of Mental Health, Rockville, MD Corresponding author: Kevin R. Krull, PhD, Department of Epidemiology and Cancer Control, St Jude Children's Research Hospital, 332 N Lauderdale St, MS 735, Memphis, TN 38105-2794; e-mail: kevin.krull{at}stjude.org
Purpose Recent studies suggest that up to 40% of childhood cancer survivors may experience neurocognitive problems, a finding that has led the Children's Oncology Group to recommend regular evaluation. However, for a variety of reasons, including costs, time restraints, health insurance, and access to professional resources, these guidelines are often difficult to implement. We report reliability and validity data on a brief neurocognitive screening method that could be used to routinely screen patients in need of comprehensive follow-up. Patients and Methods Two hundred forty consecutive patients were screened during their annual visits to a long-term survivor clinic using standard neurocognitive measures and brief parent rating. From this total, 48 patients had a second screening, and 52 patients had a comprehensive follow-up evaluation. Test-retest reliability and predictive and discriminative validity were examined. Results Good test-retest reliability was demonstrated, with an overall r = 0.72 and all individual subtest correlations greater than r = 0.40. Although means tended to improve from first to second testing, no significant changes were detected (all P > .10). The screen accurately predicted global intellect (F6,45 = 11.81, P < .0001), reading skills (F6,45 = 4.74, P < .001), and mathematics (F6,45 = 3.35, P < .008). Parent rating was a marginal indicator of global intellect only. Conclusion The brief neurocognitive screening was a better predictor of child functioning than specific parent rating. This brief measure, which can be completed in 30 minutes, is a practical and reliable method to identify cancer survivors in need of further neurocognitive follow-up.
Today in the United States, approximately 80% of more than 12,000 children diagnosed with cancer each year can expect to become long-term survivors (ie, alive 5 years after initial diagnosis).1 One of every 640 adults age 20 to 39 years will be a childhood cancer survivor.2 Given the increased survival, much attention has turned to quality of life and functional outcome after treatment, including neurocognitive and neurobehavioral development. Studies have suggested that up to 40% of childhood cancer survivors may experience neurocognitive impairment in one or more specific domains (eg, processing speed, attention, memory).3,4 Impairment in these specific domains can impede the learning of new information and interfere with the maintenance of previously learned information, which may ultimately lead to declines in global intellect. This, in turn, can result in poor academic and vocational success, low self-esteem, and behavioral or emotional disorders. The fact that neurocognitive problems are a relatively common long-term outcome of childhood cancer diagnosis and/or therapy has led the Children's Oncology Group to issue a recommendation for regular evaluation to monitor development after cranial radiation therapy and/or antimetabolite chemotherapy.5 However, for a variety of reasons, including limited health insurance coverage and limited access to professional resources, these guidelines are often difficult to implement. Full neurocognitive evaluations take an average of 8 hours to complete and routinely cost in excess of $1,500.6 Routine screening for common neurocognitive problems could provide an alternative to full evaluations for all patients. Should such a method become available, routine low-cost screening could be used to identify patients in greatest need of more comprehensive and more expensive neurocognitive evaluation and follow-up. We report the development and validation of a brief neurocognitive screening method in long-term survivors of childhood cancer. By using highly sensitive measures of commonly affected neurocognitive skills, we hoped to identify a process by which patients could be routinely screened. Our intent was that screening could be used to identify patients in greatest need of further, more comprehensive evaluation. Such neurocognitive screening has been successfully used in adult populations to identify impairment related to dementia,7-10 substance abuse,11,12 multiple sclerosis,13,14 and HIV.15 Screening has also been used in young children who are at risk for neurodevelopmental disorders.16 Furthermore, positive results have been reported using computerized screening for neurocognitive dysfunction in adolescents with systemic lupus erythematosus.17 However, to our knowledge, no screening procedure has been demonstrated to be reliable and valid in a population of pediatric cancer survivors.
Patients The procedures used in this study were reviewed and approved by the Institutional Review Board at the Baylor College of Medicine and Texas Children's Hospital (Houston, TX). Consecutive patients from the Long-Term Survivors Clinic at Texas Children's Hospital were recruited at each clinic visit as part of a standard clinic research protocol. All patients were assessed with a neurocognitive screening battery during their regular follow-up visits. Testing was conducted by a trained research associate, and families incurred no cost for the evaluation. Inclusion criteria were diagnosis of childhood cancer, current age  6 years, and language fluency in English because of the nature and availability of the neurocognitive tests administered. Exclusion criteria included prior head injury, other genetic disorders, and unrelated neurologic conditions. Although detailed recruitment information was not collected, the overall rate was estimated based on clinic attendance, clinic demographics, and study inclusion/exclusion criteria. Approximately 345 patients met age criteria for inclusion. Of these patients, approximately 15% were ineligible because of primary language spoken. Of the 293 remaining patients estimated to be eligible for the study, 240 (82%) were successfully recruited. A total of 240 patients (ages 6 to 18 years) were identified as meeting established inclusion and exclusion criteria. From this total, 48 patients had a second neurocognitive screen within a 2-year period (test-retest reliability group). These patients were recruited during a subsequent clinic visit, similar to the original cohort. In addition, 52 patients had a comprehensive follow-up neurocognitive assessment (predictive validity group). This assessment was offered to patients who either demonstrated difficulty on the screen or whose parent expressed interest in additional testing. There was no overlap between patients in the test-retest group and the validity group. Table 1 lists the basic demographic data and diagnoses for these groups.
Measures The screening battery was designed to target specific domains that are judged to be sensitive to cancer and its treatment based on existing literature and experience.18-20 The battery was comprised of standardized clinical performance-based measures and included the Digit Span Test,21,22 the Verbal Fluency Test,23 the Grooved Pegboard Test,24 and the Trail Making Test25 (hereafter referred to as DIVERGT). Each patient was administered all measures in a fixed order. All raw scores were converted to age-adjusted standard scores based on population means (mean = 100, standard deviation = 15). In addition, caregivers completed the Child Symptom Inventory (CSI) to rate their child's academic performance on a 4-point Likert scale (1 = failing, 2 = below average, 3 = average, 4 = above average). The CSI is a standardized rating scale designed to screen for behavioral, emotional, academic, and cognitive symptoms, consistent with formal diagnostic criteria.26 For this study, only the global ratings for academic and cognitive problems were analyzed. For the test-retest and validity aspects of the study, the research associate who completed follow-up testing had no knowledge of performance results on the previous screen at the time of test administration and scoring. The full or comprehensive evaluations included standard clinical measures of intelligence (ie, Wechsler Intelligence Scale for Children–Third Edition or the Wechsler Abbreviated Scale of Intelligence)21 and measures of academic functioning (ie, Wide Range Achievement Test–Third Edition).27 Intellectual measures included vocabulary, verbal reasoning, visuospatial construction, visual reasoning, processing speed, and working memory. Academic measures included word reading and mathematical calculation.
Procedures To determine the sensitivity and specificity of the DIVERGT battery, performance of 52 patients was compared with their subsequent performance on a comprehensive follow-up neurocognitive assessment. An overall impairment index was determined, and standard scores were averaged to obtain a global mean performance. The impairment index and the global mean were then compared with outcome measures on the comprehensive follow-up neurocognitive assessment, including global intelligence quotient (IQ) and performance on formal measures of academic skills. The sensitivity and specificity of the parent report of academic problems on the CSI was then compared with the sensitivity and specificity of the DIVERGT.
Statistical Analyses To further evaluate clinical validity of the screen, comparisons were made between groups determined to be at high and low risk for neurocognitive problems. High risk was defined based on a history of cancer therapy using cranial irradiation as primary treatment, whereas low risk was defined based on the absence of CNS disease and a history of chemotherapy-only treatment.
Demographic data for the sample are listed in Table 1. A slightly higher percentage of males were present in the follow-up sample compared with the total sample (  2 = 4.45, P < .05). However, no other significant differences were present. Within the total sample, 26.2% of children were reported by parents to be receiving some form of special education services. Performance and parent rating data are listed in Table 2. The mean combined performance on the first DIVERGT administration (score = 93.1) was similar to mean performance on the second administration (score = 95.2; r48 = 0.72; P < .0001). The initial administration for this test-retest subgroup did not differ from the mean score of 93.4 for the group who did not receive additional testing (t238 = 0.70; P = .49). Although mean performance on the screening tended to improve from first to second testing, no significant changes were detected (all P > .10).
In reference to the predictive validity subgroup, mean performance on the initial DIVERGT administration (score = 90.1) was slightly lower than the mean for the 188 patients who did not complete follow-up testing (score = 94.0; t238 = 1.97; P < .05). This reduced performance was likely associated with parent incentive for the additional follow-up testing. However, it could be argued that the prediction of follow-up testing from the DIVERGT was statistically more conservative using this group because the lower performance reduced the variance and correspondingly increased the difficulty in reaching a level of significance. DIVERGT predictability would likely improve with the inclusion of children functioning at a higher level. Multivariate regression analysis for the prediction of global intellect was highly significant (F6,45 = 11.81, P < .0001) and explained 61% of the variance. Univariate analyses were significant for all DIVERGT subtests (with all P < .01). Table 3 lists correlations between the total screen performance, individual screen subtests, and global intellect. For the prediction of reading skills, multivariate regression analysis was again significant (F6,45 = 4.74, P < .001), and 39% of the variance was explained. With the exception of the Trail Making Test Part A, all DIVERGT subtests were significantly related to reading (all P < .05). For mathematics, the multivariate prediction was also significant (F6,45 = 3.35, P < .008) and accounted for 31% of the variance. Univariate analyses were significant for all DIVERGT subtests (with all P < .05).
Predictive validity was also explored via dichotomous classification of impairment based on DIVERGT performance and parent rating. These classifications were then compared with performance on the comprehensive follow-up evaluation, which was also dichotomized based on impairment. For DIVERGT performance, impairment was defined as either one standard score less than 70 (ie, < second percentile) or two standard scores less than 80 (ie, < 10th percentile). Within the overall group, 47.9% of survivors met this criterion for impairment. For parent report, impairment was defined as a rating of failing or below average. For the follow-up evaluation, impairment was defined as a global IQ score less than 85 (ie, < 16th percentile) or a score on measures of academics less than 85 (ie, < 16th percentile). Within the follow-up group, 32.7% of patients displayed impaired global IQ, 28.8% displayed impaired reading, and 26.7% displayed impaired mathematics. Impairment on the DIVERGT correctly predicted impaired global intellect ( 2 = 15.06, P < .0001) better than did parent rating of global problems (2 = 8.26, P < .01). Sensitivity and specificity of the DIVERGT for impaired global intellect were 94% and 63%, respectively; whereas for parent rating, they were 53% and 46%, respectively. The DIVERGT correctly predicted impaired reading (2 = 7.69, P < .03), whereas parent rating of reading did not (2 = 2.06, P > .10). Sensitivity and specificity of the DIVERGT for reading impairment were 87% and 57%, respectively. The DIVERGT also correctly predicted impaired mathematical skills (2 = 7.59, P < .03), whereas parent rating again did not (2 = 5.50, P > .10). Sensitivity and specificity of the DIVERGT for mathematical impairment were 100% and 50%, respectively. Thus, the DIVERGT missed few individuals who displayed impairment on full clinical evaluations. Discriminative validity was explored by comparing the performance of patients at high risk (ie, cranial irradiation) versus low risk for neurocognitive problems. Mean performance on the DIVERGT for the high-risk group was 86.9 (standard deviation = 12.80) compared with 94.5 (standard deviation = 10.84) for the low-risk group (F2,234 = 7.50, P < .001). Multivariate analysis of variance using the individual subtests from the DIVERGT was significant for risk (F6,229 = 2.43) and age at diagnosis (F6,229 = 4.49, P < .0001). Children who were younger at the time of diagnosis had decreased performance on a measure of shifting attention (F1,234 = 657, P < .02) and verbal fluency (F1,234 = 8.53, P < .005). Children in the high-risk group had significantly reduced scores on a measure of shifting attention (F1,234 = 9.15, P < .003) and fine motor dexterity (F1,234 = 6.81, P < .01). As depicted in Table 4, children diagnosed with CNS tumors displayed significantly lower performance. However, each of the diagnostic groups demonstrated increased rates of children falling in the clinically impaired range, as determined by a screen subtest score less than the 10th percentile of national norms.
In this study, we examined the practical and clinical utility of using a brief neurocognitive screen to identify children at risk for future cognitive and academic difficulties. Regular monitoring of neurocognitive outcomes has been recommended for children exposed to cranial irradiation or antimetabolite chemotherapy during treatment of pediatric cancer.5 Ideally, all such children could undergo comprehensive neurocognitive evaluations to identify strengths and weaknesses, which could then be used to develop appropriate intervention plans. However, this approach is not practical at many institutions because of excessive cost, time requirements, and professional resources. In the managed care environment, fiscal responsibility and need for services are continually being assessed. A neurocognitive screen offers a practical and cost-effective option for identifying at-risk children. If the rate of neurocognitive problems in the survivor population was extremely low or extremely common, screening would be unjustified. With current estimates of the prevalence of neurocognitive problems in survivors of childhood cancer ranging from 20% to 40%,3,4 screening seems to be an appropriate method of reducing the number of unwarranted and expensive evaluations. Of note, the rate of functional problems in our sample ranged from 26.7% for mathematics to 32.7% for global intellect. The rate of more specific neurocognitive problems identified on the screen was 47.9%, which is consistent with previous reports.3,4 The DIVERGT battery demonstrated good test-retest reliability, as well as discriminative and predictive validity. The DIVERGT was a better predictor of child global cognitive and academic functioning than was specific parent report. In particular, nearly all children who demonstrated future general cognitive and academic problems were correctly identified. Furthermore, because the DIVERGT required only 20 to 30 minutes of testing time, it identified these at-risk children in an extremely efficient and cost-effective manner. The DIVERGT was not intended to replace full neurocognitive evaluations for those children at risk, but rather to identify children who would most likely benefit from the detailed analysis and targeted recommendations that such comprehensive evaluations would provide. In so doing, the screen would assist in proper allocation of limited financial and professional resources. In the long-term survivor setting, patients are commonly referred for comprehensive neurocognitive assessment when parents voice concerns about their child's academic performance to clinic staff. Although such patients may indeed be in need of further assessment, the results of this study suggest that parent report is not as sensitive or as specific in identifying the majority of at-risk patients. In this study, we quantified parent report by asking questions about general and specific academic performance, with the requirement that parents rate their child's performance on a Likert scale. Even when asked to rate specific academic areas (ie, reading and math), the parent report method did not reliably identify those children with specific academic difficulties. These results have significant implications for how we identify these at-risk children. Informal report of a child's academic functioning may result in a significant number of children who either go unidentified or experience a delay in identification and subsequent delivery of services. Early identification and intervention of cognitive and academic difficulties are extremely important. Untreated cognitive and academic difficulties have been associated with a host of negative outcomes pertaining to quality of life in cancer survivors.28-30 Children with untreated learning and attention problems are more likely to drop out of school, have a lower earning potential, and have higher incidence for development of emotional and behavioral problems.31,32 Consecutive yearly screening evaluation may also allow for ipsative comparisons that could potentially identify negative trends in neurocognitive performance before the threshold for impairment is reached. For example, if performance on the DIVERGT decreases from a standard score of 110 (ie, 75th percentile) to a standard score of 90 (25th percentile) over a 1-year period, the performance is still within the average range; however, the trend in performance could alert clinicians that further investigation may be prudent. The utility of applying the screening procedure in this way warrants further investigation. Although all children in the current study were screened in their primary language of English, the specific measures used in this screen have been translated and/or applied in numerous other countries and languages, including Arabic,33 Cantonese,34 Greek,35 Hebrew,36 and Spanish,37 among many others. In fact, many of the skills assessed as part of the screen seem to be universal constructs important for cognitive development and learning in many cultures throughout the world.38 Thus, with additional research that validates specific versions of the individual measures, international application of the DIVERGT to childhood cancer survivors may be possible. In conclusion, the DIVERGT battery is a practical tool designed to assist with identifying at-risk children for neurocognitive problems during long-term survivorship of childhood cancer. It is a cost-effective means by which to evaluate larger numbers of survivors to identify those at greatest risk for neurocognitive problems. Using this methodology, smaller groups of children identified to be at greatest risk can be referred for full evaluations. Because the DIVERGT assesses universal skills that are required for neurocognitive growth and development and that seem to generalize across cultures and languages, this process may be applicable to survivors of other chronic illnesses (eg, diabetes, lupus, organ transplantation) in many other countries.
The author(s) indicated no potential conflicts of interest.
Conception and design: Kevin R. Krull, M. Fatih Okcu, ZoAnn Dreyer, Pim Brouwers Financial support: Kevin R. Krull, Pim Brouwers Administrative support: Kevin R. Krull, M. Fatih Okcu, ZoAnn Dreyer, Pim Brouwers Provision of study materials or patients: Kevin R. Krull, M. Fatih Okcu, ZoAnn Dreyer, Kala Kamdar Collection and assembly of data: Kevin R. Krull, Brian Potter, Neelam Jain, Pim Brouwers Data analysis and interpretation: Kevin R. Krull, M. Fatih Okcu, Brian Potter, Neelam Jain, Kala Kamdar, Pim Brouwers Manuscript writing: Kevin R. Krull, M. Fatih Okcu, Brian Potter, Neelam Jain, ZoAnn Dreyer, Kala Kamdar, Pim Brouwers Final approval of manuscript: Kevin R. Krull, M. Fatih Okcu, Brian Potter, Neelam Jain, ZoAnn Dreyer, Kala Kamdar, Pim Brouwers
Authors’ disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Ries LAG, Melbert D, Krapcho M, et al: SEER cancer statistics review, 1975-2004. http://seer.cancer.gov/csr/1975_2004/ 2. Hewitt M, Weiner SL, Simone JV: Childhood Cancer Survivorship: Improving Care and Quality of Life. Washington, DC, National Academy of Sciences, 2003 3. Moleski M: Neuropsychological, neuroanatomical, and neurophysiological consequences of CNS chemotherapy for acute lymphoblastic leukemia. Arch Clin Neuropsychol 15:603-630, 2000[CrossRef][Medline] 4. Mulhern RK, Fairclough D, Ochs J: A prospective comparison of neuropsychologic performance of children surviving leukemia who received 18-Gy, 24-Gy, or no cranial irradiation. J Clin Oncol 9:1348-1356, 1991[Abstract] 5. Nathan PC, Patel SK, Dilley K, et al: Guidelines for identification of, advocacy for, and intervention in neurocognitive problems in survivors of childhood cancer: A report from the Children's Oncology Group. Arch Pediatr Adolesc Med 161:798-806, 2007 6. Sweet JJ, Nelson NW, Moberg PJ: The TCN/AACN 2005 "salary survey": Professional practices, beliefs, and incomes of U.S. neuropsychologists. Clin Neuropsychol 20:325-364, 2006[CrossRef][Medline] 7. Loewenstein DA, Acevedo A, Agron J, et al: Stability of neurocognitive impairment in different subtypes of mild cognitive impairment. Dement Geriatr Cogn Disord 23:82-86, 2007[Medline] 8. Cullum CM, Weiner MF, Gehrmann HR, et al: Feasibility of telecognitive assessment in dementia. Assessment 13:385-390, 2006 9. Knox MR, Lacritz LH, Chandler MJ, et al: Association between Dementia Rating Scale performance and neurocognitive domains in Alzheimer's disease. Clin Neuropsychol 17:216-219, 2003[Medline] 10. Meulen EF, Schmand B, van Campen JP, et al: The seven minute screen: A neurocognitive screening test highly sensitive to various types of dementia. J Neurol Neurosurg Psychiatry 75:700-705, 2004 11. Grohman K, Fals-Stewart W: The detection of cognitive impairment among substance-abusing patients: The accuracy of the neuropsychological assessment battery-screening module. Exp Clin Psychopharmacol 12:200-207, 2004[Medline] 12. Zinn S, Bosworth HB, Edwards CL, et al: Performance of recently detoxified patients with alcoholism on a neuropsychological screening test. Addict Behav 28:837-849, 2003[CrossRef][Medline] 13. Benedict RH, Zivadinov R: Reliability and validity of neuropsychological screening and assessment strategies in MS. J Neurol 254:II22-II25, 2007 (suppl 2)[CrossRef][Medline] 14. Benedict RH, Cox D, Thompson LL, et al: Reliable screening for neuropsychological impairment in multiple sclerosis. Mult Scler 10:675-678, 2004 15. Levine AJ, Hinkin CH, Miller EN, et al: The generalizability of neurocognitive test/retest data derived from a nonclinical sample for detecting change among two HIV+ cohorts. J Clin Exp Neuropsychol 29:669-678, 2007[CrossRef][Medline] 16. Robins DL, Fein D, Barton ML, et al: The Modified Checklist for Autism in Toddlers: An initial study investigating the early detection of autism and pervasive developmental disorders. J Autism Dev Disord 31:131-144, 2001[CrossRef][Medline] 17. Brunner HI, Ruth NM, German A, et al: Initial validation of the Pediatric Automated Neuropsychological Assessment Metrics for childhood-onset systemic lupus erythematosus. Arthritis Rheum 57:1174-1182, 2007[CrossRef][Medline] 18. Brown RT, Madan-Swain A: Cognitive, neuropsychological, and academic sequelae in children with leukemia. J Learn Disabil 26:74-90, 1993 19. Cousens P, Ungerer JA, Crawford JA, et al: Cognitive effects of childhood leukemia therapy: A case for four specific deficits. J Pediatr Psychol 16:475-488, 1991 20. Brouwers P, Poplack D: Memory and learning sequelae in long-term survivors of acute lymphoblastic leukemia: Association with attention deficits. Am J Pediatr Hematol Oncol 12:174-181, 1990[Medline] 21. Wechsler D: Wechsler Intelligence Scale for Children (ed 3). San Antonio, TX, Psychological Corporation, 1991 22. Wechsler D: Wechsler Adult Intelligence Scale (ed 3). San Antonio, TX, Psychological Corporation, 1997 23. Benton AL, Hamsher K, Sivan AB: Multilingual Aphasi Examination (ed 3). Iowa City, IA, AJA Associates, 1983 24. Trites RL: Neuropsychological Test Manual. Ottawa, Ontario, Canada, Royal Ottawa Hospital, 1977 25. Reitan R: The Halstead-Reitan Neuropsychological Test Battery: Theory and Clinical Interpretation (ed 2). Tucson, AZ, Neuropsychology Press, 1993 26. Gadow KD, Sprafkin J: Child Symptom Inventory. Stony Brook, NY Checkmate Plus, Ltd, 1997 27. Wilkinson GS: Wide Range Achievement Test 3. Wilmington, DE, Wide Range, Inc, 1993 28. Mulhern RK, Friedman AG, Stone PA: Neuropsychological status of children with acute lymphoblastic leukemia treated for central nervous system relapse. Am J Pediatr Hematol Oncol 11:106-113, 1989[Medline] 29. Schultz KA, Ness KK, Whitton J, et al: Behavioral and social outcomes in adolescent survivors of childhood cancer: A report from the childhood cancer survivor study. J Clin Oncol 25:3649-3656, 2007 30. Robinson KE, Gerhardt CA, Vannatta K, et al: Parent and family factors associated with child adjustment to pediatric cancer. J Pediatr Psychol 32:400-410, 2007 31. Kessler RC, Adler L, Barkley R, et al: The prevalence and correlates of adult ADHD in the United States: Results from the National Comorbidity Survey Replication. Am J Psychiatry 163:716-723, 2006 32. Torgesen JK: The prevention of reading difficulties. J School Psychol 40:7-26, 2002[CrossRef] 33. Stanczak DE, Stanczak EM, Awadalla AW: Development and initial validation of an Arabic version of the Expanded Trail Making Test: Implications for cross-cultural assessment. Arch Clin Neuropsychol 16:141-149, 2001[CrossRef][Medline] 34. Lu L, Bigler ED: Performance on original and a Chinese version of Trail Making Test Part B: A normative bilingual sample. Appl Neuropsychol 7:243-246, 2000[CrossRef][Medline] 35. Kosmidis MH, Vlahou CH, Panagiotaki P, et al: The verbal fluency task in the Greek population: Normative data, and clustering and switching strategies. J Int Neuropsychol Soc 10:164-172, 2004[Medline] 36. Kave G: Phonemic fluency, semantic fluency, and difference scores: Normative data for adult Hebrew speakers. J Clin Exp Neuropsychol 27:690-699, 2005[CrossRef][Medline] 37. Ostrosky-Solis F, Ardila A, Rosselli M: NEUROPSI: A brief neuropsychological test battery in Spanish with norms by age and educational level. J Int Neuropsychol Soc 5:413-433, 1999[Medline] 38. Georgas J, Weiss LG, Van de Vijver FJR, et al: Culture and Children's Intelligence: Cross-Cultural Analysis of the WISC-III. San Diego, CA, Academic Press, 2003 Submitted February 21, 2008; accepted May 15, 2008.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2008 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
