Diagnostic accuracy of dementia screening tools in the Chinese population: a systematic review and meta-analysis of 167 diagnostic studies

Background The rate of undetected dementia is high in China. However, the performance of dementia screening tools may differ in the Chinese population due to the lower education level and cultural diversity. This study aimed to evaluate the diagnostic accuracy of dementia screening tools in the Chinese population. Methods Eleven electronic databases were searched for studies evaluating the diagnostic accuracy of dementia screening tools in older Chinese adults. The overall diagnostic accuracy was estimated using bivariate random-effects models, and the area under the summary receiver operating characteristic curve was presented. Results One hundred sixty-seven studies including 81 screening tools were identified. Only 134 studies qualified for the meta-analysis. The Mini-Mental State Examination (MMSE) was the most commonly studied tool, with a combined sensitivity (SENS) and specificity (SPEC) of 0.87 (95%CI 0.85–0.90) and 0.89 (95%CI 0.86–0.91), respectively. The Addenbrooke’s Cognitive Examination-Revised (ACE-R) (SENS: 0.96, 95%CI 0.89–0.99; SPEC: 0.96, 95%CI 0.89–0.98) and Montreal Cognitive Assessment (MoCA) (SENS: 0.93, 95%CI 0.88–0.96; SPEC: 0.90, 95%CI 0.86–0.93) showed the highest performance. The General Practitioner Assessment of Cognition (GPCOG), Hasegawa’s Dementia Scale and Cognitive Abilities Screening Instrument had performances comparable to that of the MMSE. The cut-off scores ranged widely across studies, especially for the MMSE (range: 15–27) and MoCA (range: 14–26). Conclusions A number of dementia screening tools were validated in the Chinese population after cultural and linguistical adaptations. The ACE-R and MoCA had the best diagnostic accuracy, whereas the GPCOG, with an administration time < 5 minutes, could be considered as a rapid screening tool.