BACKGROUND AND AIM: Living in greener areas is associated with slower cognitive decline among the elderly, but few studies have explored the link with dementia-related mortality. This study examines the relationship between residential surrounding greenness and dementia-related mortality in the five largest urban areas in Belgium. METHODS: We used data from the 2001 Belgian census linked to mortality register data during 2001-2014. We included individuals aged 60 years or older and living in the agglomerations of Brussels, Antwerp, Ghent, Charleroi and Liege in 2001. Exposure to residential surrounding greenness was assessed using the Normalised Difference Vegetation Index (NDVI) within 500-m buffer from the residential address registered in 2001. We used ICD-10 codes to identify general dementia-related mortality including any neurodegenerative diseases (F00-F03, G12.2, G20-G22, G30-G31 and G35), and four types of dementia-related mortality: Alzheimer’s disease (G30); vascular dementia (F01); unspecified dementia (F03); and Parkinson’s disease (G20-G22). To define our cases, we considered any mention to dementia in the death certificate (i.e., as original, immediate, additional or intermediate cause of death). Cox proportional hazard models were conducted to obtain hazard ratios (HR) for dementia-related mortality for an IQR increase in surrounding greenness. Models were adjusted for age, gender, migrant background, education, housing tenure and PM2.5. RESULTS:In total, 1,089,215 individuals were included at baseline (2001). 6.1% died from neurodegenerative diseases during follow-up (1.9% Alzheimer’s disease, 0.5% vascular dementia, 0.4% Parkinson’s disease and 2.8% unspecified dementia). Surrounding greenness was inversely associated with all outcomes of dementia-related mortality, excepting Parkinson’s disease. These relationships remained significant after adjustment for all the covariates considered, but not with unspecified dementia after including PM2.5. The strongest beneficial association after complete adjustment was found for vascular dementia (HR 0.91, 95%CI: 0.86, 0.96). CONCLUSIONS:Our findings suggest that living close to green spaces in urban areas might reduce dementia-related mortality risk among the elderly. KEYWORDS: Green space, Neurodegenerative outcomes, Mortality