AIMS: To examine the association between diabetes, and ageing 12 years later. METHODS: Subjects reporting angina, cancer, heart attack or stroke at baseline (1990-1994) were excluded. Diabetes at baseline was identified by self-report or elevated plasma glucose (≥ 7.0 mmol/l fasting or ≥ 11.1 mmol/l non-fasting). 6431 eligible men and women had survived to age 70 years at follow-up (2003-2007), and 5704 with complete data were included in these analyses. Those without ischemic heart disease, stroke, and cancer, no perceived major difficulty with physical functioning, and no evidence of psychological distress were considered to have aged successfully, irrespective of diabetes at follow-up (n=1271). Logistic regression was used to examine the independent association between baseline diabetes and successful ageing. RESULTS: At baseline 216 eligible people were identified with diabetes. This was inversely associated with successful ageing at follow-up independent of smoking, physical activity, alcohol use and obesity (OR 0.65, 95% 0.42-0.99), Of the people with diabetes at follow-up, only 12.5% did not have another condition characteristic of usual ageing, compared with 22.7% of people who did not have diabetes. CONCLUSIONS: Diabetes is strongly associated with factors characterising less successful ageing, suggesting that the impact of diabetes may be larger than currently estimated.