Association of Triglyceride to HDL-Cholesterol ratio with Cognitive impairment among elderly people: A cross-sectional study in Southwestern Uganda

This dataset was generated as part of a community-based cross-sectional study conducted in Rukiga District, Southwestern Uganda, from February to March 2025. The study aimed to investigate the association between the triglyceride to high-density lipoprotein cholesterol (TG/HDL-C) ratio and cognitive impairment among elderly individuals aged 60 years and above. Our central hypothesis was that elevated TG/HDL-C ratio, a marker of metabolic dysregulation, is independently associated with cognitive impairment in older adults.

Data were collected from 272 participants selected via multistage cluster sampling. Each participant underwent a standardized cognitive assessment using the Mini-Mental State Examination (MMSE), with scores below 24 indicating cognitive impairment. In total, 183 individuals (67.3%) screened positive, with 95 having mild and 88 severe impairment.

The dataset includes variables across multiple domains: sociodemographic factors (e.g., age, gender, education level), clinical conditions (e.g., diabetes, hypertension, depression), anthropometric data (e.g., waist circumference, BMI), lifestyle behaviors (e.g., smoking, alcohol use, physical activity), and laboratory-measured serum lipid profiles (TG, HDL-C, LDL-C, TC), collected after 8–12 hours of fasting. TG/HDL-C ratio and other lipoprotein ratios were computed. Blood samples were processed using enzymatic methods on a HumaStar 200 analyzer with appropriate quality controls.

Statistical analysis included descriptive statistics, nonparametric tests, and logistic regression. Notably, a high TG/HDL-C ratio was independently associated with cognitive impairment after adjusting for confounders (aOR for highest tertile = 3.38; p = 0.031). Age, lower educational attainment, and depression were also significant predictors.

This dataset provides crucial insight into the metabolic and psychosocial determinants of cognitive impairment in a low-resource setting. It is particularly valuable for researchers exploring blood-based biomarkers for early detection of neurocognitive disorders and can inform public health interventions targeting aging populations in sub-Saharan Africa. The dataset is structured to enable both univariate and multivariate analyses and is suitable for epidemiological, clinical, or machine learning-based investigations into geriatric cognitive health.