The development and validation of a new cross‐cultural naming test

Abstract

AbstractBackgroundTraditional naming tests such as the Boston Naming Test are unsuitable to assess naming impairment in diverse populations, given the large influence of culture, language, and education on performance. A more fair naming test is urgently needed. The goal of this study was therefore to develop and validate a new cross‐cultural naming test to diagnose naming impairment in Alzheimer’s disease in diverse populations.MethodWe carried out a multistage pilot study. First, we generated a list of 149 potentially cross‐cultural items, and selected the items with a homogeneous age of acquisition and word frequency across multiple languages. For the 74 items that remained, we selected several colored photographs (examples provided in Figure 1), out of which 194 controls selected the best matching photograph. The 74‐item test was piloted in 15 diverse healthy controls, after which 14 items were removed. The final 60‐item test was validated in 39 controls and 61 culturally, educationally, and linguistically diverse patients – mainly with dementia (30%), psychiatric illness (20%), MCI (16%), or subjective memory impairment (15%). Participants were mostly from North Africa, Suriname, Eastern Europe, and the Middle East, and the majority of participants had an education level of primary school or lower (51%).ResultThe new cross‐cultural naming test was well able to discriminate between patients with Alzheimer’s/mixed dementia (n=11, median score: 48/60) and the rest of the sample (median: 58/60, AUC=0.87). Preliminary item analyses indicated that colors, as well as animals (ant, worm, snake) and occupations with a later age of acquisition/lower word frequency were the most sensitive. There were robust correlations with other measures of language (Recall of Pictures Test naming ρ=0.26, p=.04, animal fluency ρ=0.66, p<.001, foods fluency ρ=0.57, p<.001), as well as a culture‐sensitive cognitive screener (ρ=0.69, p<.001). The scores were not correlated with education level, age, or number of years of residence in the Netherlands (analysis in controls).ConclusionThis new naming test is an important step towards a more fair neuropsychological assessment and more accurate dementia diagnosis in culturally, linguistically, and educationally diverse individuals.