Importance: Type 2 diabetes prevalence is increasing, and currently 24% of young adults have prediabetes. Little is known regarding timing and progression from normoglycemia to diabetes starting in young adulthood.
Objective: To examine glycemic trajectories from young adulthood through middle age.
Design, setting, and participants: This cohort study used sequence analysis to identify patterns of glycemic trajectories across 30 years in the Coronary Artery Risk Development in Young Adults (CARDIA) study, a large, diverse, population-based cohort of young adults from 4 national community sites with repeated measures of fasting plasma glucose (FPG). Sequence analysis reduces the various glycemic trajectories that individuals may follow over time into common trajectory clusters. The CARDIA study enrolled adults aged 18 to 30 years from 1985 to 1986, with groups balanced on Black and White race, sex, age, and educational level. Data through year 30 (2015-2016) were used, excluding participants with baseline diabetes, those missing the first 2 FPG values, and those missing 6 or more FPG values. Data analysis was performed from July 2023 to October 2024.
Main outcomes and measures: Clusters of common glycemic trajectory patterns.
Results: Among the 4684 study participants (2553 [54.5%] female), 1278 unique glycemic trajectories were identified across 30 years. A total of 3420 participants (73.0%) had trajectories classifiable into 9 common patterns of glycemic trajectory: stable normoglycemia; 5 patterns of impaired fasting glucose (IFG) not progressing to diabetes, including a brief period of low IFG regressing to normoglycemia at younger (mean [SD] age at IFG, 35.9 [6.6] years) and older ages (mean [SD] age at IFG, 46.8 [3.7] years), younger-onset sustained IFG (mean [SD] age at IFG, 38.6 [5.0] years), older-onset sustained low IFG (mean [SD] age at IFG, 51.7 [5.1] years), and older-onset sustained high IFG (mean [SD] age at IFG, 41.9 [8.8] years); and 3 patterns of diabetes, including young onset (mean [SD] age at diabetes, 34.8 [3.6] years), middle-age onset (mean [SD] age at diabetes, 43.6 [4.4] years), and older onset (mean [SD] age at diabetes, 52.4 [3.8] years).
Conclusions and relevance: This cohort study revealed substantial heterogeneity in glycemic trajectories across 30 years of follow-up. Clusters that represented IFG during young adulthood did not demonstrate progression to diabetes, and clusters that represented diabetes lacked prior IFG during young adulthood. These data suggest the need for strategies beyond simply screening for IFG during young adulthood to identify and target diabetes prevention for young adults.