Cognitive-behavioral correlates of proxy reports on cognitive capabilities in pediatric patients with epilepsy

• Published in: Seizure (London, England) Vol. 51; pp. 193 - 199
• Main Authors: Helmstaedter, Christoph, Pandolfo, Renata Vaz, Hoppe, Christian, Witt, Juri-Alexander
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 01.10.2017
• Subjects: Adolescent; Assessment; Child; Cognition Disorders - etiology; Epilepsy - psychology; Female; Humans; Male; Neuropsychology; Parents; Proxy; Proxy report; Psychometrics - instrumentation; Seizures; Surveys and Questionnaires; Validity; Validity; Neuropsychology; Assessment; Seizures; Proxy report
• ISSN: 1059-1311; 1532-2688
• DOI: 10.1016/j.seizure.2017.08.016

•Proxy reports on childrens and adolescents cognition provide a valid but rough estimate of objective test performance.•The KOPIJ ratings reflect behavioral problems of children as well.•Parent ratings underestimate cognitive and overestimate behavioral problems.•Cognitive subdomains rated by parents do not differentially correlate with neuropsychological tests.•Parent ratings may be taken as a surrogate measure when testing is not possible. To validate the Cognitive Problems in Children and Adolescents Questionnaire (KOPKIJ, German: Kognitive Probleme bei Kindern und Jugendlichen), a proxy report measure for recognizing cognitive problems in pediatric patients with epilepsy. Anonymized data sets from 279 pediatric epilepsy patients were standardized in regard to the KOPKIJ results of 352 healthy children and adolescents. The KOPKIJ was related to objective routine neuropsychological assessment (NPY), and to two subjective measures, the Child Behavior Checklist questionnaire (CBCL), a proxy rating by the parents, and a questionnaire for self-perceived health-related quality of life in children and adolescents (KINDL). Following principal component analysis of the KOPKIJ’s normative data, three scales “basic functions”, “academic skills”, and a “total score” were differentiated, which indicated problems in 35%, 33%, and 32% of the children. Low IQ was evident in 23%, objective impairments in at least one major cognitive domain in 64% of the patients. Behavior (CBCL) and quality of life (KINDL) were impaired in 40% and 21% of the patients. Separate regression analyses revealed that objective cognitive performance (IQ, language, visual-spatial functions) explained ∼30%, behavior (CBCL) and coping with the disease (KINDL) about ∼40%, and clinical features (age at onset) 5–8% of the variance of the KOPKIJ scales. The parents' impressions of children’s cognition obtained via the KOPKIJ only in part reflect the neuropsychological cognitive status of children and adolescents with epilepsy. They appear rather determined by the children’s behavioral problems, which in real life situations indeed often co-occur with cognitive impairments. Aspects of the epilepsy only marginally influence the parents’ ratings of their children’s cognition.