Risk of Dementia Among Patients With Diabetes in a Multidisciplinary, Primary Care Management Program

Importance Although poorly controlled diabetes is associated with a higher incidence of dementia, few studies have examined the association of diabetes management interventions with dementia incidence. Objective To examine the association of receiving a multidisciplinary diabetes management program (the Risk Assessment and Management Program-Diabetes Mellitus [RAMP-DM]) that enables better glycemic control with subsequent risk of dementia incidence and the association of dementia with glycemic control. Design, Setting, and Participants This territory-wide, retrospective, matched cohort study with more than 8 years of follow-up was conducted using electronic health care records from all the patients who used public health care services in Hong Kong from 2011 to 2019. Eligible participants included all patients with type 2 diabetes (T2D) who were managed in primary care settings. Patients who received RAMP-DM were matched in a 1:1 ratio with patients who received usual care only. Data analysis occurred from April 2023 to July 2023. Exposures Diagnosis of T2D, hemoglobin A 1C (HbA 1C ) level, and attendance at a general outpatient clinic or family medicine clinic. Patients received either RAMP-DM or usual care. Main Outcomes and Measures Incidence of all-cause dementia and subtypes of dementia were compared between the RAMP-DM and usual care participants using a Cox proportional hazard model with other baseline characteristics, biomarkers, and medication history adjusted. HbA 1C levels were measured as a secondary outcome. Results Among the 55 618 matched participants (mean [SD] age, 62.28 [11.90] years; 28 561 female [51.4%]; 27 057 male [48.6%]), including the 27 809 patients in the RAMP-DM group and 27 809 patients in the usual care group, patients had been diagnosed with T2D for a mean (SD) of 5.90 (4.20) years. During a median (IQR) follow-up period of 8.4 (6.8-8.8) years, 1938 patients in the RAMP-DM group (6.97%) and 2728 patients in the usual care group (9.81%) received a diagnosis of dementia. Compared with those receiving usual care, RAMP-DM participants had a lower risk of developing all-cause dementia (adjusted hazard ratio [aHR], 0.72; 95% CI, 0.68-0.77; P < .001), Alzheimer disease (aHR, 0.85; 95% CI, 0.76-0.96; P = .009), vascular dementia (aHR, 0.61; 95% CI, 0.51-0.73; P < .001), and other or unspecified dementia (aHR, 0.71; 95% CI, 0.66-0.77; P < .001). Compared with having a mean HbA 1C level during the first 3 years after cohort entry between 6.5% and 7.5%, a higher risk of dementia incidence was detected for patients with a 3-year mean HbA 1C level greater than 8.5% (aHR, 1.54; 95% CI, 1.31-1.80]), between 7.5% and 8.5% (aHR, 1.33; 95% CI, 1.19-1.48), between 6% and 6.5% (aHR, 1.17; 95% CI, 1.07-1.29), and 6% or less (aHR, 1.39; 95% CI, 1.24-1.57). Conclusions and Relevance In this cohort study of patients with T2D, the findings strengthened evidence of an association of glycemic control with dementia incidence, and revealed that a multidisciplinary primary care diabetes management program was associated with beneficial outcomes for T2D patients against dementia and its major subtypes. A moderate glycemic control target of HbA 1C between 6.5% and 7.5% was associated with lower dementia incidence.