OBJECTIVE: To determine prevalence, trends, and subsequent outcomes of prediabetes defined by American Diabetes Association (ADA), World Health Organization (WHO), and International Expert Committee (IEC) criteria in the United States between 1999 and 2018. METHODS: Ten cycles of cross-sectional National Health and Nutrition Examination Survey data were included. Prediabetes was defined by ADA, WHO, and IEC criteria. Unadjusted or covariate adjusted prevalence and trends of prediabetes were estimated. Cox proportional regression model was performed to evaluate the association between prediabetes and all-cause, cardiovascular, or diabetes-related mortality. RESULTS: Among the 59 369 participants included (weighted mean age, 41.1 years; 48.7% male), the prevalence of prediabetes was 29.4% in ADA criteria, 16.3% in WHO criteria, and 5.0% in IEC criteria. The covariate adjusted prevalence of prediabetes defined by ADA criteria increased significantly in at least twofolds from 15.6% in 1999-2002 to 37.3% in 2015-2018 (P < .001). Similar significant increased trends were observed in WHO and IEC criteria (P < .001). Compared with normal glycemia, prediabetes participants had higher adjusted risk of diabetes-related mortality irrespective of the criteria used (ADA: hazard ratio [HR] 9.11 [95% CI, 5.83-14.22]; WHO: HR 5.35 [95% CI, 3.01-9.51]; IEC: HR 9.64 [95% CI, 5.92-15.71]). No significant associations between prediabetes and all-cause or cardiovascular mortality were observed in the adjusted models. CONCLUSION: In the United States, approximately 1 in 3 individuals have prediabetes according to ADA criteria. The prevalence of prediabetes has shown a significant and more than twofold increase over the past 2 decades, posing an elevated risk of diabetes-related mortality, regardless of the criteria applied.