Ten Year Patterns of Change in HbA1C and Risk of Dementia in a Diverse Cohort of Adults with Type 2 Diabetes

Abstract Background Though high HbA 1c (marker for glycemic control) is associated with increased dementia risk, the influence of longitudinal changes in HbA 1c is unclear. We examined the association between 10‐year trajectories of HbA 1c and dementia risk in a large, diverse sample of people in Northern California. Method In a cohort of 32, 914 patients with Type 2 diabetes aged ≥ 50 years, we obtained repeated measures of HbA 1c , dementia diagnoses, and comorbidities from electronic medical records (EMR) from 1996‐20021. We identified 5 latent trajectories of yearly mean HbA 1c over a 10‐year period (Figure 1): Group 1‐HbA 1c consistently <7; Group 2‐ HbA 1c consistently close to 7; Group 3‐moderate HbA 1c peaking late between 7‐9; Group 4‐moderate HbA 1c peaking early between 7‐9; Group 5‐ HbA 1c consistently between 9‐11. Cox proportional hazards models estimated the association between HbA 1c trajectories and incident dementia ≥5 years after the end of the exposure period adjusting for age (as timescale) and prevalent health conditions (Table 1) at the beginning of the trajectory period. Stratified models examined possible effect modification by sex or race/ethnicity. Result Approximately half of the sample were women (Table 1). The most common trajectory was Group 2‐consistently close to 7 (47%), followed by Group 1‐consistently <7 (35%), Group 3‐moderate peaking late (9%), Group 4‐moderate peaking early (7%), and Group 5‐ consistently high (3%). Compared to Group 1, Groups 3‐5 had elevated risk of dementia (Figure 2); risk was comparable in Groups 3 and 4 (HR Group3 = 1.30 (1.14, 1.48) and HR Group4 = 1.28 (1.10, 1.49)) and was highest in Group 5 (HR = 1.78 (1.42, 2.26). Group 2 was not at elevated dementia risk compared to Group 1 (HR = 1.02 (0.94, 1.09). These associations did not vary by sex or race/ethnicity. Conclusion Over a 10‐ year period, consistently elevated HbA 1c levels ≥ 9 are associated with an 80% elevated risk of dementia compared to those with a trajectory of HbA 1c 6‐7. Those whose trajectory included HbA 1c levels of 8‐9 either early (Group 4) or late (Group 3) in the10‐year period had approximately 30% increased risk. These results suggest that there are long‐term consequences of elevated HbA 1c even if levels subsequently decrease.