Most previous studies on cardiovascular health (CVH) and incident type 2 diabetes (T2D) have used a single measure of CVH, and none have investigated the association with impaired fasting glucose (IFG). We examined the association between changes in CVH and incident T2D and IFG. Within the Whitehall II study, CVH was examined every 5 years from 1991/93 until 2015/16. Subjects with 0-2, 3-4, and 5-6 ideal metrics of CVH from the American Heart Association were categorized as having low, moderate, or high CVH, respectively. There were 6,234 participants (mean age 49.8 ± 6.0 years, 70% male) without prior cardiovascular disease and T2D, including 5,015 who were additionally free from IFG at baseline. Over a median follow-up of 24.8 (interquartile range 24.0-25.2) years, 895 and 1,703 incident cases of T2D and IFG occurred, respectively. Change in CVH between 1991/93 and 2002/04 was calculated among 4,464 participants free from CVD and T2D and among 2,795 participants additionally free from IFG. In multivariate analysis, compared with those with stable low CVH, risk of T2D was lower in those with initially high CVH (hazard ratio [HR] 0.21; 95% CI 0.09, 0.51), those who had persistently moderate CVH or changed from moderate to high CVH (moderate-moderate/high; HR 0.53; 95% CI 0.41, 0.69), low-moderate/high (HR 0.62; 95% CI 0.45, 0.86), and moderate-low (HR 0.74; 95% CI 0.56, 0.98). Results were similar for IFG, but the effect sizes were smaller. Compared with stable low CVH, other patterns of change in CVH were associated with lower risk of T2D and IFG. © 2019 by the American Diabetes Association.