Whilst aging is well recognized to be the most significant risk factor for the development of atrial fibrillation (AF),AF also affects young person in whom it is frequently symptomatic and troublesome.The relationship between age and prevalence of AF among patients with hypertension remains uncertain.Methods A retrospective cross-sectional study was conducted in consecutively enrolled hypertensive patients aging more than 18 years old from January 1,2013,to December 31,2013,at the Guangdong community in China.AF was systematically screened and diagnosed by rest 12-lead electrocardiogram (ECG) or self-reported.Univariate and multivariate logistic regression,two-piecewise linear regression model and generalized additive model were performed to evaluate the relationship between age and AF.Results There were 7808 participants including 3678 (47.1％) men and 4130 (52.9％) women (mean age of 62.3 years old) and 78 cases of AF were identified,including 42 men (53.8％).Both univariate and multivariate logistic regression illustrated that age was positively correlated with AF whether age was analyzed as a continuous (all P ＜ 0.0001) or categorical (P for trend ＜ 0.001)variable.A non-linear correlation was detected after adjusting for potential confounders and the inflection points were 40 and 60 years.Conclusions The relationship between age and AF is non-linear.Age ＜ 40 years is a protective factor for AF,while age ＞ 60 years is significantly correlated with a higher risk of AF.