OBJECTIVE: To investigate whether adults with atopic eczema are at an increased risk of cardiovascular disease and whether the risk varies by atopic eczema severity and condition activity over time. DESIGN: Population based matched cohort study. SETTING: UK electronic health records from the Clinical Practice Research Datalink, Hospital Episode Statistics, and data from the Office for National Statistics, 1998-2015. PARTICIPANTS: Adults with a diagnosis of atopic eczema, matched (on age, sex, general practice, and calendar time) to up to five patients without atopic eczema. MAIN OUTCOME MEASURES: Cardiovascular outcomes (myocardial infarction, unstable angina, heart failure, atrial fibrillation, stroke, and cardiovascular death). RESULTS: 387 439 patients with atopic eczema were matched to 1 528 477 patients without atopic eczema. The median age was 43 at cohort entry and 66% were female. Median follow-up was 5.1 years. Evidence of a 10% to 20% increased hazard for the non-fatal primary outcomes for patients with atopic eczema was found by using Cox regression stratified by matched set. There was a strong dose-response relation with severity of atopic eczema. Patients with severe atopic eczema had a 20% increase in the risk of stroke (hazard ratio 1.22, 99% confidence interval 1.01 to 1.48), 40% to 50% increase in the risk of myocardial infarction, unstable angina, atrial fibrillation, and cardiovascular death, and 70% increase in the risk of heart failure (hazard ratio 1.69, 99% confidence interval 1.38 to 2.06). Patients with the most active atopic eczema (active >50% of follow-up) were also at a greater risk of cardiovascular outcomes. Additional adjustment for cardiovascular risk factors as potential mediators partially attenuated the point estimates, though associations persisted for severe atopic eczema. CONCLUSIONS: Severe and predominantly active atopic eczema are associated with an increased risk of cardiovascular outcomes. Targeting cardiovascular disease prevention strategies among these patients should be considered.