To assess the relationship between use of β-blockers and all-cause mortality in patients with and without diabetes. Using data from the US National Health and Nutrition Examination Survey 1999-2010, we conducted a prospective cohort study. The study participants were followed-up from the survey participation date until December 31, 2011. We used a Cox proportional hazards model for all-cause mortality analysis. The multivariate-adjusted hazard ratios (HRs) of the participants taking β-blockers were compared with those of the participants not taking β-blockers. This study included 2840 diabetic participants and 14,684 nondiabetic participants. Compared with diabetic participants not taking a β-blocker, all-cause mortality was significantly higher in diabetic participants taking any β-blocker (HR, 1.49; 95% CI, 1.09-2.04; P=.01), taking a β1-selective β-blocker (HR, 1.60; 95% CI, 1.13-2.24; P=.007), or taking a specific β-blocker (bisoprolol, metoprolol, and carvedilol) (HR, 1.55; 95% CI, 1.09-2.21; P=.01). In addition, all-cause mortality in diabetic participants with coronary heart disease (CHD) was significantly higher in those taking beta-blockers, compared with those not taking beta-blockers (HR, 1.64; 95% CI, 1.08-2.48; P=.02), whereas that in non-diabetic participants with CHD was significantly lower in those taking beta-blockers (HR, 0.68; 95% CI, 0.50-0.94; P=.02). A propensity score-matched Cox proportional hazards model yielded similar results. Use of β-blockers may be associated with an increased risk of mortality for patients with diabetes and among the subset who have CHD.