Exposure to alcohol and risk of dementia: a systematic review and dose-response meta-analysis

Background: The relation between alcohol intake and cognitive function risk has been established, especially for high levels of exposure. Conversely, the effects at low-moderate levels is still debated and the aim of this study is to characterize the dose-response relation between different degree of alcohol intake and cognitive decline and dementia. Methods: We carried out a comprehensive literature search in PubMed and Embase online databases for cohort and case-control studies investigating the association between alcohol intake on cognitive function. For study identification and selection, we will use the Rayyan online open-source program. Whenever possible, we will conduct a dose-response meta-analysis using the new one stage cubic spline mixed-effects model in order to evaluate the nonlinear relation of such association. According to inclusion criteria, we selected studies on human population with no major comorbidities (e.g. CVD, diabetes, immunological deficits and cancer) and without a diagnosis of dementia and MCI at baseline. Further exclusion criteria were binge drinking and alcoholism, in addition to articles in which no exposure dose was reported. Results: Out of total 3676 articles, we excluded 951 duplicates and after title and abstract screening, we further excluded 2496 records to be subjected to a more accurate analysis. After full-text evaluation, of the 266 resulting articles, we identified 20 eligible studies, with alcohol intake ranging from 0 to 50.4 g alcohol/day and 0 to 4.8 drinks/day. Conclusions: The study is currently ongoing; data extraction and prediction of the database for the dose-response meta- analysis will be presented at the Congress.