Association Between Anthropometry and Cognitive Impairment in Patients with Type 2 Diabetes: Secondary Analysis from...

With an ageing global population, the triad of type 2 diabetes (T2D), obesity and dementia poses a growing public health challenge. India harbours a notable proportion of younger patients with T2D with a non-obese or lean phenotype. However, data remain scarce on the impact of adiposity on cognition in T2D, particularly when assessed using a comprehensive, culturally and linguistically adaptable cognitive battery in a cohort free from major confounding factors. The aim of this study is to examine the relationship between various anthropometric indices and cognitive performance in patients with T2D. This cross-sectional observational study was conducted at the diabetic clinic of a tertiary hospital from 2022 to 2024. Eligible participants were patients with T2D aged 20-60 years with at least primary education. Exclusion criteria included non-T2D diagnosis, inability to communicate in Bengali and conditions known to impair cognition. A total of 125 patients with T2D were recruited. Demographics, diabetes-related variables and anthropometric measurements were recorded. Cognitive function was assessed using the Bengali version of the Addenbrooke's Cognitive Examination (ACE-III). Statistical analysis was performed using Jeffreys' Amazing Statistics Program (v.0.19). ACE-III total scores showed significant positive correlations with height, weight and neck circumference (NC) (p<0.001 for each). Attention was positively associated with height, weight, NC, neck-height ratio (NHR) and negatively with weight-adjusted waist index (WWI) (p<0.05 for all). Memory correlated positively with height and weight (p<0.05 for both). Language was positively related to height, weight and NC (p<0.05 for all) and negatively to WWI (p=0.011). Visuospatial ability positively correlated with height, weight, waist circumference (WC), hip circumference (HC), NC and NHR (p<0.05 for all). Lean patients with T2D had significantly lower visuospatial scores (p=0.040), lower ACE-III total scores (p=0.049) and a greater prevalence of cognitive impairment (p=0.032). In multiple linear regression, height (p=0.014) and HC (p=0.024) were independent predictors of ACE-III total score. This is the first Indian study to evaluate the association between anthropometric measures and cognition in T2D. Cognitive impairment and dementia were more prevalent in lean than in obese patients with T2D. Future studies incorporating imaging-based body composition analysis are warranted to identify modifiable anthropometric risk factors for cognitive decline in T2D.