Disease susceptibility and risk of dementia in patients with chronic kidney disease: a prospective cohort study based on UK Biobank

Patients with chronic kidney disease (CKD) face an elevated risk of dementia, but multidomain determinants remain poorly defined. To identify modifiable psychosocial, lifestyle, and clinical factors associated with dementia risk in CKD and to examine subgroup-specific differences. We prospectively analyzed 16 083 CKD participants from the UK Biobank. A wide range of lifestyle, medical, social, electronic device use, environmental, and sensory factors were assessed. Cox proportional hazards models with inverse probability weighting (IPW) were applied, complemented by accelerated failure time models, phosphate-adjusted sensitivity analyses, Fine-Gray competing-risk models with death treated as a competing event, and model diagnostics. Stratified analyses were conducted by age, sex, kidney failure status, and genetic susceptibility in the genetic subcohort. Over a mean follow-up of 13.80 years, 806 dementia cases were identified. In nominal analyses, higher physical activity, greater educational attainment, and moderate computer use were associated with lower dementia risk, whereas hypertension, diabetes, depression, hearing loss, and limited social engagement were associated with higher risk. After FDR adjustment, moderate physical activity and moderate computer use remained associated with lower dementia risk, whereas hypertension, diabetes, depression, hearing loss, and lower leisure/social activity remained associated with higher risk. No nominally significant associations were detected for body mass index, low-density lipoprotein cholesterol, air pollution exposure, or visual impairment in this CKD cohort. Exploratory subgroup analyses, based on nominal P values, suggested that associations may differ by kidney function and genetic susceptibility. These findings support multidomain and individualized dementia-prevention strategies in CKD, while recognizing the need for further studies with non-CKD comparators and external validation.