Background: Microbial dysbiosis and translocation of microbial products across digestive tract mucosal surfaces are hypothesized risk factors for impaired glucose regulation. We specifically investigated whether measures of subgingival dysbiosis predicted longitudinal fasting plasma glucose (FPG) change. Methods: The Oral Infections, Glucose Intolerance and Insulin Resistance Study (ORIGINS) enrolled 300 diabetes-free adults (77% female) aged 20-55 years (mean±SD=34±10). Subgingival plaque samples were collected and analyzed using 16S rDNA sequencing (Illumina MiSeq; NexGEN) using the Human Oral Microbe Identification using Next Generation Sequencing (HOMINGS) protocol. FPG was measured at baseline and after two-years. Microbial community alpha-diversity was calculated using Simpson’s index. Multivariable linear regressions modeled 2-year glucose change on baseline relative abundance of 376 individual oral taxa (in separate models) with adjustment for age, sex, race/ethnicity, education, smoking status, BMI and baseline glucose levels. All taxa with a p-value<0.05 were used to construct a microbial dysbiosis-index (MD-index) defined as the ratio of taxa abundance in organisms positively vs. inversely associated with FPG change. The MD-index was subsequently used in regression models to predict glucose change. Statistical significance was based on a false discovery rate (FDR)<0.05 or a Bonferroni corrected p-value (1x10 -4 ). Results: Mean 2-year FPG change±SD was 1.5±8 mg/dl. Mean±SD values of MD-index were 2.2±2.4. Baseline levels of 23 taxa were associated with FPG change (p<0.05), 11 of which had inverse associations. Among these 23 taxa, Treponema HOT238, Leptotrichia HOT498 and Stomatobaculum HOT097 had an FDR<0.05. The MD-index was inversely correlated with alpha-diversity (r=-0.19, p<0.01). Mean FPG-change±SE in the 3 rd vs. 1 st tertile of MD-index were 4.5±0.9 vs. -1.6±0.9 (p<1x10 -4 ). Conclusion: Increased levels of microbial dysbiosis, reflecting decreased microbial diversity, were associated with increased 2-year glucose change among young, diabetes-free subjects.