Frontal Assessment Battery in Early Cognitive Impairment: Psychometric Property and Factor Structure

Background: The Frontal Assessment Battery (FAB) is a reliable and valid bedside tool for testing executive function in dementia. Given the increasing interest in utility of FAB as a screening tool in early cognitive impairment (ECI), there is a surprising lack of studies evaluating its psychometric property and factor structure, nor the influence of factors such as age, education and gender, in ECI.

Objectives: This study aims to investigate the psychometric properties and factor structure of FAB in older adults with ECI, as well as the influence of age, gender and education.

Design, Setting And Participants: This is a retrospective, observational cross-sectional study with 300 community dwelling, predominantly Chinese older adults (14 normal, 130 mild cognitive impairment (MCI), and 156 mild dementia) who presented to Memory Clinic from January 2011 to December 2013. Measurements and Analysis: We collected data on demographic, cognitive, functional and behavioral evaluation. To examine the psychometric properties of FAB, we examined the concurrent, convergent, and discriminant validity; internal consistency by Cronbach's alpha; and factor structure by exploratory factor analysis. The influence of age, education and gender was examined using unadjusted and adjusted correlational analyses with CDR-SOB. We performed analysis for the whole group and for MCI subgroup.

Results: FAB total score decreases significantly from normal to dementia group attesting to concurrent validity. It correlated significantly with digit span backwards and Chinese Mini Mental State Examination (r=0.38 and 0.47 respectively, p<0.01) and poorly with Neuropsychiatric Inventory-Questionnaire and depression (r=0.004 and -0.02 respectively), supporting its convergent and discriminant validity. Factor analysis yielded a single-factor solution for FAB with fair Internal consistency (alpha=0.610). FAB is relatively unaffected by age, gender and education level. These good psychometric properties extend to MCI, albeit with greater influence by education level. FAB items of conceptualization and mental flexibility have good discriminatory ability between MCI and normal subjects.

Conclusion: FAB has good concurrent, convergent and discriminant validity with fair internal consistency in ECI that is premised on a one-factor structure. It is relatively unaffected by age, gender or education. Taken together, FAB is a useful bedside screening tool for executive function in ECI.