The detrimental effect of increased variability in glycated Haemoglobin(HbA1c) on cardiovascular disease(CVD) and mortality risk in patients with diabetes remains unclear. The aim of this study was to investigate their associations.This prospective cohort study included 147 811 patients aged 45-84 years with type2 diabetes mellitus, without CVD and attained at least three HbA1c records before baseline within 2008-2010. HbA1c variability was obtained using a mixed effects model to reduce regression dilution bias. Age-specific associations(45-54;55-64;65-74;75-84 years) between HbA1c variability and risk of CVD and mortality were assessed by Cox regression adjusted for patient characteristics and usual HbA1c.After a median follow-up of 7.4 years(1.02 million person-years), an overall incidence of 40 785 events including CVD(27 793 incidence) and all-cause mortalities(23 175 incidence) were identified. Positive log-linear associations between HbA1c variability and CVD and mortality were identified in all age groups. The hazard ratios (HRs) for the composite of CVD and all-cause mortality revealed age was inversely associated with HbA1c variability, with a 28% higher risk per 1% increase in HbA1c variability in the 45-54 age group[all composite outcomes HR:1.28(1.21,1.35)], whereas only a 14% higher risk in the 75-84 age group[all composite outcomes HR:1.14(1.11,1.17)]. Subgroup analysis showed the risk in patients with usual HbA1c < 7% was about eight times more than those with usual HbA1c ≥ 8%.HbA1c variability was strongly related to CVD and mortality in patients with diabetes across all age groups. Whilst pursuing optimal HbA1c target, attention should be given to patients with high HbA1c variability especially those young ones with good HbA1c control. This article is protected by copyright. All rights reserved.