Cassandra Germain, Ashley Dunham, Oksana Makeevaemail, Valentina Markova, Irina Zhukova, Zarui Melikyan, Natalia Zhukova, Larisa Minaycheva, Stepan Buikin, Marina Abushaeva, Svetlana Burlutskaya, Ekaterina Karas, Elena Starinskaya, Yuka Maruyama, Heather Romero, Brenda Plassman, Kathleen Welsh-Bohmer, Kathleen Hayden: Variations in cognitive performance, demographics and health across community-based registries, from North Carolina to Tomsk, Russia // Alzheimer's & Dementia: The Journal of the Alzheimer's Association. Volume 9, Issue 4, Supplement, Page P322, July 2013
Regional differences in cognitive performance are important to understand for the interpretation of data in large multinational clinical trials. Variation in cognitive performance may be due to differences in years of education, health conditions, cultural differences, and testing experience.
Data were collected from three community studies (n =2343) building participant registries for a primary prevention study of Alzheimer's disease: Duke University, Durham NC, the Murdock Study, Kannapolis and Cabarrus County, NC, and the Nebbiolo Study, Tomsk, Russia. Using the same protocols, each site administered identical tests including: Montreal Cognitive Assessment (MoCA), CERAD Word List Memory Test delayed recall (WLM), Trail Making Test Part B (Trails B), and the self-report ADCS Mail-In Cognitive Function Screening Instrument (MCFSI). Site differences were evaluated with GLM or χ 2 as appropriate. Multilevel modeling was used to measure the variance explained by each site and predictors of cognitive performance.
There were significant site differences on most demographic and health variables, cognitive measures, and the impact of demographic and health variables on cognitive measures. The average age of the sample was 68.5(7.1). Tomsk had the highest mean age (70 years), the highest prevalence of high blood pressure (80.7%) and cardiovascular disease (CVD) (42.7%); Kannapolis/Cabarrus had the highest prevalence of diabetes (DM) (19.1%). Site differences accounted for 36% of variation on Trails B; 33% on the MoCA; 17.7% on MCFSI scores; and 3.7% the variation on WLM. Covariates including demographic and health variables (age, sex, education, CVD, DM, blood pressure, obesity, and stroke) accounted for 21.1% of site variation on Trails B; 23% of site variation on the MoCA; 29% variation on MCFSI scores; and 46% of site variation on the WLM delayed recall.
Sources of variation in cognitive test performance need to be investigated in order to guide interpretation of test performance in international research contexts. Potential differences include test wiseness, prevalence of vascular disease and other health conditions, cultural differences in exposure to test stimuli, and attitudes towards time and timed tests. Results point to the need for local normative data for different countries to better interpret performance in clinical contexts.