Diabetes Care and Dementia Among Older Adults: A Nationwide 3-Year Longitudinal Study.

OBJECTIVES

To compare diabetes monitoring and the incidence of acute diabetic complications between patients with and without incident Alzheimer's Disease and Related Syndromes (ADRS).

DESIGN

Longitudinal observational study from 2010 to 2014.

SETTING

Data from the French national health system database.

PARTICIPANTS

The France-Démence cohort: individuals aged 65 years or older suffering from incident ADRS, based on long-term disease registry, hospitalization for dementia, or antidementia drug delivery. They were matched (1:1) to a pair free of ADRS on age, sex, residence area, and insurance scheme. This study included France-Démence population with known diabetes for at least 2 years.

MEASUREMENTS: Data related to diabetes control and complications: biological monitoring such as glycated hemoglobin A1c (HbA1c, ≥1/y, ≥2/y), lipid profile, microalbuminuria; eye examination; hospitalization for diabetes-related complications such as coma with ketoacidosis; and hospitalization for hypoglycemia were studied between the year prior to ADRS identification (Y-1) and the 2 following years (Y0; Y1). Incidences between the 2 groups (ADRS/non-ADRS) were compared using age-standardized incidence ratios (SIR).

RESULTS

The studied population included 87,816 individuals. HbA1c determination was less frequent in ADRS group, no matter the study period and the minimal annual threshold used. Respectively, 82.6% and 88.5% of ADRS and non-ADRS group had at least 1 HbA1c testing during Y-1 [SIR = 0.94, 95% confidence interval (CI) 0.93-0.95], 73.4% and 89.0% during Y0 (SIR = 0.83, 95% CI 0.82-0.84), and 75.4% and 89.3% during Y1 (SIR = 0.85, 95% CI 0.83‒0.86). Subjects with ADRS were also consistently more hospitalized than non-ADRS peers. The gap was maximal in the year following the diagnosis, as observed for hospitalizations for any cause related to diabetes (SIR Y-1: 2.04, Y0: 3.14, Y1: 1.67), diabetes mellitus with coma (SIR Y-1: 3.84, Y0: 9.30, Y1: 3.06), and hypoglycemia (SIR Y-1: 4.20, Y0: 5.25, Y1: 2.27).

CONCLUSIONS

Incident ADRS is associated with a less frequent diabetes monitoring and an increased risk of diabetes complications compared with older people without ADRS. Our study questions healthcare quality offered to participants with ADRS in comorbidity control. Further investigations are required to explain the mechanisms underlying our results and to propose actions to improve care of patients with ADRS.