Associations of diabetes subgroups and life-style with incident dementia: insights from unsupervised phenotype clustering evaluation

Abstract

Background
To examine heterogeneity in dementia threat throughout diabetes subgroups and look at whether or not life-style mitigates this threat.
Methods
We did an unsupervised clustering evaluation (self-organizing maps) in sufferers with diabetes from the Health and Retirement Study (HRS). Clustering variables included age at diabetes onset, physique mass index (BMI), hemoglobin A1c (HbA1c), whole ldl cholesterol (TC), high-density lipoprotein ldl cholesterol (HDL-c), systolic blood strain (SBP), estimated glomerular filtration charge (eGFR), and C-reactive protein (CRP). Healthy life-style elements included by no means smoking, low to average alcohol consumption, and prime tertile of bodily exercise. Incident dementia was outlined as a self-reported prognosis or Telephone Interview for Cognitive Status (TICS) scores ≤6.
Results
We included 10,705 individuals (40.2% males and imply age 67.6 years) on this evaluation. Three subgroups have been recognized amongst diabetes: subgroup 1 (excessive BMI and elevated inflammatory marker), subgroup 2 (late-onset diabetes, hypertension, and low eGFR), and subgroup 3 (early-onset diabetes, excessive HbA1c, and excessive levels of cholesterol). Compared to normoglycemia, subgroup 3 of diabetes confirmed a considerably greater threat of dementia (hazard ratio [HR] 1.43, 95% confidence interval [CI] 1.15-1.77), whereas the opposite subgroups didn’t. However, adopting a positive life-style was related to a diminished threat of dementia in individuals with normoglycemia (HR 0.70, 95% CI 0.57-0.87), prediabetes (HR 0.67, 95% CI 0.50-0.90), and subgroup 1 of diabetes (HR 0.33, 95% CI 0.14-0.79).
Conclusions
Diabetes-related dementia threat is heterogeneous, and advantages of life-style might differ by diabetes subgroups.