THE EFFECTS OF CAFFEINE ON ANXIETY
A high-sensitivity electronic search and which underwent peer review according to the PRESS Guide was carried out in March 2021 in the following databases as recommended by the Cochrane Handbook: Cochrane Library, MEDLINE via PUBMED, LILACS via VHL, APA PsycNet, EMBASE, Scielo, Scopus, Web Of Science and Cinahl. The eligibility criteria was Randomized clinical trials complete and available that evaluated the direct influence of caffeine through capsules or caffeinated beverages on anxiety and/or anxiety disorders in samples of healthy individuals of any age and sex. By excluding 2726 duplicates, a detailed screening of 4273 titles and abstracts was performed. The full texts of the 91 remaining studies were evaluated, of which only 27 were chosen for the qualitative synthesis. The final references were exported to a Google Sheets file, where two review authors (AN and CS) used a form to extract the characteristics of the studies: article data (authors year of publication, country of study and duration), sociodemographic data population (sex, age group, caffeine consumption habits and comorbidities) and methodological data (study design, anxiety scale instrument used to measure anxiety, scores before and after the intervention, intervention dose, time interval between measurement, intervention and new measurement and, activity performed between the intervention and measurement). The quality assesssment was conducted independently by two review authors (AN and CS), in which disagreements were resolved through discussions between the evaluators. All articles included had their methodological quality assessed, but without using it as an exclusion criterion. To this end, the Revised Cochrane Risk-of-bias Tool for Randomized Trials (Rob2) was used, a Cochrane tool recommended for analyzing the risk of bias in randomized clinical trials, in versions for parallel design trials and also for crossovers.