Developing and testing a South African Brief Cognitive Score in literate and illiterate people of mixed language groups

BACKGROUND: The Folstein mini mental state exam (MMSE) is the most commonly administered assessment of cognitive functioning; however, its utility in illiterate individuals is limited. In South Africa, more than eight million adults are considered functionally illiterate and cognitive evaluation using standard scales is inaccurate. Other countries have developed adapted MMSE scales for their local purposes. AIM: The first aim of this study was to develop a South African Brief Cognitive Score (SA BCS) for use in minimally literate or illiterate individuals. The second aim was to test this SA BCS against the original Folstein MMSE in patients with memory problems. SETTING: The study was conducted in Tshilamba, Tshiombo, Tshifudi, Dzimauli and Pile in Venda as well as Rethabiseng and Zithobeni in Bronkhorstspruit for the illiterate study group, and Steve Biko Academic Hospital for the literate study groups. METHODS: The SA BCS was developed considering our local requirements and substituting questions needing literacy with items that did not. Both the original Folstein MMSE and the SA BCS were administered to groups of literate and illiterate normal individuals. Thereafter, the tests were repeated in groups of literate and illiterate patients with cognitive decline. RESULTS: Firstly, 33 illiterate and 31 literate subjects were assessed. The average original Folstein score was 29.29/30 for the literate subjects, and for the SA BCS 29.80. For the illiterate subjects, the average score for the original Folstein MMSE was 21.24/30 and for the SA BCS 27.45. Kruskall–Wallis equality of population rank test confirmed a significant improvement in the scores of the SA BCS in the illiterate group when compared to the original Folstein (p = 0.0001). In the second part of the study, 20 literate and 20 illiterate patients were assessed as before. In the literate group, the average original Folstein MMSE score was 20.5, while the average score for the South African BCS was 22.5. In the illiterate group, the average Folstein MMSE was 18.9; and the average score in the South African BCS was 22.8. The Kruskal–Wallis equality of population rank test showed a significant difference (p = 0.008) between the scores of the illiterate versus literate patients when the Folstein MMSE was used to assess cognition. With the SA BCS, no significant difference was found between the groups (p = 0.79). CONCLUSION: The SA BCS appears to have potential to be a more reliable scale when assessing cognition in illiterate or minimally literate subjects when compared to the original Folstein MMSE.