Validation and normative data for the Multilingual Naming Test (MINT) in French-speaking Quebec adults over 40

Picture naming tests are critical tools for assessing language and semantic memory deficits in dementia, but must be adapted to local cultural context. This study aimed to develop and validate a Quebec French version of the Multilingual Naming Test (MINT), including determining norms and assessing diagnostic accuracy of mild cognitive impairment (MCI) and dementia. Quebec French-speaking participants (n = 224) were drawn from the TRIAD Montreal cohort and the DEVOCS study and stratified into three subgroups using a double-threshold of quantitative (Montreal Cognitive Assessment) and qualitative (consensus diagnosis) assessments. Linear models were used to create norms and compare diagnostic groups; correspondence with the Boston Naming Test (BNT) used the equipercentile method. Item-specific analysis revealed that naming of three items was sex-dependent. We found significant effects of sex and education, and an interaction trend, on uncued MINT scores. The MINT showed high comparability to the BNT, and Receiver Operating Characteristic curves corroborated their similar diagnostic ability. The MINT performed better at discriminating dementia from cognitively unimpaired participants than at identifying MCI. Correcting for education and sex was necessary to obtain accurate scores. Our results show that the MINT is adequately valid and effective to replace the BNT in neuropsychological assessment in the Quebec French population.