Résumé
Abstract:INTRODUCTION Subjective cognitive decline (SCD), i.e. a self-experienced decline in cognitive ability in the absence of objective cognitive impairment, is recognized as the first notable cognitive syndrome in the preclinical stage of Alzheimer’s disease (AD) and other dementias. However, estimates on the prevalence of SCD are scarce. Therefore, we aimed to estimate SCD prevalence based on consensus research criteria for SCD across international cohort studies of ageing.METHODS Analyses were based on the combined baseline data for 16 international cohort studies from 15 countries. All studies were members of COSMIC (Cohort Studies of Memory in an International Consortium). Qualitative/semantic and quantitative (Item Response Theory/IRT) approaches were used to a) harmonize SCD items across studies and b) derive SCD prevalence estimates, applying a uniform operationalization algorithm based on current SCD research criteria and implementation guidelines.RESULTS Analyses included data from 39,387 cognitively unimpaired individuals at least 60 years of age (mean age: M = 73.1 years; SD = 7.1 years); 57.7% were women. Variety of SCD items was high between studies; however, qualitative harmonization (QH) and IRT both robustly suggested a total SCD prevalence of around one quarter (QH: 23.8%, 95%CI = 23.3%-24.4%; IRT: 25.6%, 95%CI = 25.1%-26.1%) across uniform criteria. SCD prevalence was lower in a) women compared to men, b) individuals with educational levels above primary, c) high-income countries compared to countries with lower income levels, and d) in White people compared to Asian people.DISCUSSION SCD occurs relatively frequently in ageing populations around the globe, as estimated across 16 international cohorts. However, SCD measurement techniques vary in the absence of standardized instruments, and therefore, our prevalence estimates may be associated with someinaccuracy. Nevertheless, with almost a quarter of all elderly reporting SCD, further investigation of its significance as the first cognitive syndrome in preclinical AD and other dementia is needed. This may lead to improved strategies for early identification and prevention of dementia.