Profile of mild behavioral impairment and factor structure of the Mild Behavioral Impairment Checklist in cognitively normal older adults

Abstract

Objectives: In this large population study we set out to examine the profile of Mild Behavioral Impairment using the Mild Behavioral Impairment Checklist (MBI-C), and explore its factor structure when employed as a self-report and informant rated tool. Design: Population based cohort study. Setting: Online testing via the PROTECT study (http://www.protectstudy.org.uk) Participants: 5,742 participant-informant dyads. Measurements: Both participants and informants completed the MBI-C. The factor structure of the MBI-C was evaluated by exploratory factor analysis (EFA). Results: The most common MBI-C items as rated by self-report and informants related to affective dysregulation (mood/anxiety symptoms), being present in 34% and 38% of the sample respectively. The least common were items relating to abnormal thoughts and perception (psychotic symptoms) (present in 3 and 6% of the sample respectively). There were only weak correlations between self-report and informant-report MBI-C responses. EFA for both sets of respondent answers indicated a five-factor solution for the MBI-C was appropriate, reflecting the hypothesized structure of the MBI-C. Conclusion: This is the largest and most detailed report on the frequency of MBI symptoms in a non-dementia sample. The full spectrum of MBI symptoms was present in our sample, whether rated by self-report or informant report. However, we show that the MBI-C performs differently in self-report versus informant-report situations, which may have important implications for the use of the questionnaire in clinic and research.