I read with great interest the paper “Comparison of performance on three neuropsychological tests in healthy Turkish immigrants and Danish elderly” authored by Nielsen et al. (Reference Nielsen, Vogel and Waldemar2012) in International Psychogeriatrics. In their study, 73 Turkish with lower educational level and 36 Danish older subjects were assessed on the Recall of Pictures Test (RPT), Clock Reading Test (CRT), and supermarket fluency (SF). They found significant effects of age and years of education on the Turkish performance, and significant differences between the Turkish and Danish samples on the CRT and SF performances, but not on the RPT. Despite these differences in performance, the authors suggested that these three tests are important neuropsychological instruments for the assessment of dementia in ethnic minority people. As the elderly population is increasing globally, the need for reliable cross-cultural measures suitable to assess this population at risk to develop dementia is also growing. In particular, normative data need to be appropriated ethnically and in terms of educational level, for example, to determine valid cut-offs for mild cognitive impairment (MCI) and patients with dementia. I would like to provide evidence on further neuropsychological instruments sensitive to memory and language deficits for different ethnic groups of Portuguese-speaking Brazilian individuals within the large geographical area of Brazil. Although they all speak Portuguese, they are of different ethnic background and therefore differ in their educational levels significantly. We assessed 411 healthy older subjects (61 to 85 years old and 0 to 17 years of education) from different states in Brazil on the Hopkins Verbal Learning Test – Revised (HVLT-R; Brandt and Benedict, Reference Brandt and Benedict2001), a verbal episodic memory test, and on the Brief Visuospatial Memory Test – Revised (BVMT-R; Benedict, Reference Benedict1997), a visuospatial episodic memory test (Miotto et al., Reference Miotto, Campanholo, Rodrigues, Serrao, Lucia and Scaff2012). The results showed a significant influence of age on both tests, as well as in the original American normative data. Education also showed significant influence on the results, possibly due to the different levels of education of the sample, including illiterate subjects. Gender had no influence on subjects' tests results.

In another study, we assessed 416 healthy elderly Portuguese-speaking Brazilians (60 to 80 years old and 0 to 17 years of education) on the Boston Naming Test, a language test (Miotto et al., Reference Miotto, Sato, Lucia, Camargo and Scaff2010). The results showed an effect of gender, education, and age on the total scores. Those with lower educational level produced, in general, fewer correct responses. This can possibly be explained by the fact that subjects with higher educational levels are more exposed to information, reading material, and knowledge about cultural and linguistic features of Portuguese and foreign languages. These results are in line with those from Nielsen et al.'s (Reference Nielsen, Vogel and Waldemar2012) study, where age and years of education were negatively correlated with performance on most of the tests only in the Turkish sample who had lower educational level.

In conclusion, these findings indicate the importance of adequate normative studies when assessing different ethnic populations, particularly when including subjects with lower educational levels. Larger studies which incorporate different ethnic groups would need to (even if the same language is used) generate first their normative data in the same population to have valid cut-offs. Finally, these findings also suggest the cross-cultural applicability of a number of widely used neuropsychological tests to assess the presence of MCI and dementia in elderly patients from ethnic minority populations.