# Validity and reliability of the Italian version of the Mild Behavioral Impairment Checklist in cognitively unimpaired and mild cognitive impairment individuals

## TL;DR

The study confirms that the Italian version of the Mild Behavioral Impairment Checklist is reliable and valid for assessing behavioral symptoms in older adults without dementia.

## Contribution

The study provides validation of the Italian version of the MBI-C for use in dementia-free populations.

## Key findings

- The MBI-C showed high internal consistency (α = 0.867) and strong inter-domain correlations.
- The MBI-C had an AUC of 0.937 for detecting clinical MBI with a cut-off of 5.5.
- The MBI-C total score strongly correlated with the NPI-Q total score (ρ = 0.820).

## Abstract

The Mild Behavioral Impairment Checklist (MBI-C) is a tool for detecting MBI, a neurobehavioral syndrome associated with an increased dementia risk.

This study aimed to evaluate the reliability and validity of the Italian version of the informant-rated MBI-C in an outpatient sample of dementia-free individuals.

A cross-sectional study was conducted on 72 older people without dementia (n = 47, mild cognitive impairment; n = 25, cognitively unimpaired). During the visit, physicians administrated the MBI-C and Neuropsychiatric Inventory Questionnaire (NPI-Q) to the informant. Internal consistency of MBI-C was measured by the Cronbach's coefficient alpha and inter-domain correlation coefficients. Diagnostic performance of MBI-C for clinically identified MBI by ISTAART criteria was assessed through ROC analysis, identifying the optimal cut-off based on the Youden Index. Spearman's correlations were used to evaluate the concurrent validity of MBI-C with the NPI-Q, Mini-Mental State Examination (MMSE), Instrumental Activity of Daily Living (IADL) and 3-item UCLA Loneliness Scale.

MBI-C showed high internal consistency (
α
 = 0.867) and strong inter-domain correlation (
ρ
 = 0.760
∼
0.859, p < 0.001). The Area Under the Curve (AUC) for detecting clinical MBI was 0.937 (95%CI: 0.865–0.972), with an optimal cut-off of 5.5 (sensitivity = 0.849, specificity = 0.876). The MBI-C total score strongly correlated with the NPI-Q total score (
ρ
 = 0.820, p < 0.001). Only the MBI-C total score significantly correlated with the 3-item UCLA (
ρ
 = 0.236, p = 0.046); no significant correlations were found with MMSE and IADL scores.

The Italian version of MBI-C demonstrated strong reliability, validity, and diagnostic performance. Therefore, MBI-C may be a suitable tool for assessing behavioral symptoms in dementia-free individuals.