There is concern that the increasing prevalence of type 2 diabetes may diminish improving trends in life expectancy. This study aimed to determine whether the mortality of type 2 diabetes, relative to mortality in the general population, is remaining constant. The study included a cohort of 48,556 subjects with type 2 diabetes first diagnosed between 1996 and 2006, drawn from 197 family practices in the United Kingdom General Practice Research Database. There were 6,630 deaths observed. Expected mortality was estimated from United Kingdom life tables. Relative mortality was modeled using Poisson regression. In men with diabetes, from 1996 to 2006, the age-standardized all-cause mortality rate decreased by 0.82/1,000 per year (95% confidence interval (CI): 0.36, 1.27) and by 0.49 (95% CI: 0.29, 0.68) in women with diabetes. After adjustment for age, sex, and diabetes duration, there was a consistent decrease in relative mortality during the period of study. Relative mortality for subjects diagnosed in 1996 was 13% (95% CI: 2, 25) higher than that in 2001; for subjects diagnosed in 2006, relative mortality was 26% (95% CI: 8, 40) lower than that in 2001. Relative mortality of type 2 diabetes appears to be decreasing in men and women in the United Kingdom.