Reliability and Validity of Composite Scores from the NIH Toolbox Cognition Battery in Adults

• Published in: Journal of the International Neuropsychological Society Vol. 20; no. 6; pp. 588 - 598
• Main Authors: Heaton, Robert K., Akshoomoff, Natacha, Tulsky, David, Mungas, Dan, Weintraub, Sandra, Dikmen, Sureyya, Beaumont, Jennifer, Casaletto, Kaitlin B., Conway, Kevin, Slotkin, Jerry, Gershon, Richard
• Format: Journal Article
• Language: English
• Published: New York, USA Cambridge University Press 01.07.2014
• Subjects: Adult; Adults; Age; Age Factors; Aged; Aged, 80 and over; Alzheimer's disease; Brain research; Cognition & reasoning; Cognition - physiology; Education; Ethnicity; Executive Function; Family income; Female; Humans; Language; Male; Medical screening; Memory; Memory, Episodic; Memory, Short-Term; Middle Aged; National Institutes of Health (U.S.) - standards; Neuropsychological Tests - standards; Psychometrics - methods; Psychometrics - standards; Quantitative psychology; Reaction Time; Reproducibility of Results; Special Series; Surveys and Questionnaires; United States; Young Adult; United States; Neuropsychological assessment; Cognitive screener; Executive Function; Memory; Attention; Cognitive assessment
• ISSN: 1355-6177; 1469-7661; 1469-7661
• DOI: 10.1017/S1355617714000241

This study describes psychometric properties of the NIH Toolbox Cognition Battery (NIHTB-CB) Composite Scores in an adult sample. The NIHTB-CB was designed for use in epidemiologic studies and clinical trials for ages 3 to 85. A total of 268 self-described healthy adults were recruited at four university-based sites, using stratified sampling guidelines to target demographic variability for age (20–85 years), gender, education, and ethnicity. The NIHTB-CB contains seven computer-based instruments assessing five cognitive sub-domains: Language, Executive Function, Episodic Memory, Processing Speed, and Working Memory. Participants completed the NIHTB-CB, corresponding gold standard validation measures selected to tap the same cognitive abilities, and sociodemographic questionnaires. Three Composite Scores were derived for both the NIHTB-CB and gold standard batteries: “Crystallized Cognition Composite,” “Fluid Cognition Composite,” and “Total Cognition Composite” scores. NIHTB Composite Scores showed acceptable internal consistency (Cronbach’s alphas=0.84 Crystallized, 0.83 Fluid, 0.77 Total), excellent test–retest reliability (r: 0.86–0.92), strong convergent (r: 0.78–0.90) and discriminant (r: 0.19–0.39) validities versus gold standard composites, and expected age effects (r=0.18 crystallized, r=−0.68 fluid, r=−0.26 total). Significant relationships with self-reported prior school difficulties and current health status, employment, and presence of a disability provided evidence of external validity. The NIH Toolbox Cognition Battery Composite Scores have excellent reliability and validity, suggesting they can be used effectively in epidemiologic and clinical studies. (JINS, 2014, 20, 1–11)