Due to phenotypic heterogeneity in diabetes, it is important to examine different sub-phenotypes of diabetes based on biomarkers and evaluate their associations with all-cause mortality. We utilized the nationally representative sample in the NHANES III (The Third National Health and Nutrition Examination Survey, 1988-1994) to explore possible sub-phenotypes of diabetes determined based on tertile levels of 50 nutritional, environmental, and inflammatory biomarkers, respectively. Weighted logistic regression analyses were performed among 2,130 diabetic patients aged 20 to 90 years with adjustment for age, sex, race, BMI, smoking status, drinking status, physical activity, and educational background. We observed an increased risk of all-cause mortality associated with higher level of serum C-reactive protein (P for trend=.01), serum thyroid stimulating hormone (P for trend=.01), and urine albumin (P for trend<.0001), respectively. In contrast, higher levels of serum sodium (P for trend=.01) and alpha carotene (P for trend=.003) are associated with a decreased risk of all-cause mortality. In addition, these significant associations were not modified by age, sex, and race. In conclusion, our results indicate that some biomarkers are reliable predictors of diabetes-related mortality and have potential clinical values for improving risk stratification in diabetic patients, though further validation of their respective and joint predictive values is needed.