Aims The aim of this prospective study was to examine CVD risk reduction in type 1 diabetes (1) for people with favourable cardiovascular health metrics and (2) by clustering of these metrics. Methods Data from 2313 participants from the EURODIAB Prospective Complications Study were analysed. All had type 1 diabetes (51% men, mean ± SD age 32 ± 9 years). Seven cardiovascular health metrics were studied—smoking, BMI, physical activity, a diet score, total cholesterol/HDL-cholesterol ratio, combined systolic and diastolic BP and HbA 1c —divided into favourable/less favourable categories. Cox proportional hazards models were used to calculate HRs (95% CIs) of incident CVD for each metric. Clusters were made by scoring each individual by the number of favourable metrics. Results A total of 163 people developed incident CVD during a mean ± SD follow-up of 7.2 ± 1.3 years. Participants with more favourable HbA 1c levels of <57 mmol/mol (<7.4%) had a 37% significantly lower CVD risk than those with a less favourable HbA 1c (HR 95% CI 0.63 0.44, 0.91), and participants with a more favourable BP (systolic BP <112 mmHg and diastolic BP <70 mmHg) had a 44% significantly lower CVD risk than participants in the less favourable BP group (HR 95% CI 0.56 0.34, 0.92). There was a dose–response relation with a lower HR observed with greater clustering of more favourable metrics: people with four or more favourable metrics had an HR of 0.37 (95% CI 0.18, 0.76), adjusted for sex and age at diabetes diagnosis, compared with those with no favourable metrics. Conclusions/interpretation Low HbA 1c and low BP were protective cardiovascular health metrics in our study of people with type 1 diabetes. Targeting all cardiovascular health metrics could be more effective in preventing CVD than targeting single metrics. Graphical abstract