The Impact of Built and Social Environmental Characteristics on Diagnosed and Estimated Future Risk of Dementia

Article type: Research Article

Authors: Bagheri, Nasser^{a; b; *} | Mavoa, Suzanne^c | Tabatabaei-Jafari, Hossein^a | Knibbs, Luke D.^d | Coffee, Neil T.^b | Salvador-Carulla, Luis^{a; e} | Anstey, Kaarin J.^{f; g}

Affiliations: [a] Centre for Mental Health Research, the Research School of Population Health, the Australian National University, Australia | [b] The Australian Geospatial Health Lab, Health Research Institute, The University of Canberra, Australia | [c] Melbourne School of Population and Global Health, the University of Melbourne, Australia | [d] The School of Public Health, The University of Sydney, Australia | [e] Menzies Centre for Health Policy, Faculty of Medicine and Health, University of Sydney | [f] UNSW Ageing Futures Institute, the University of New South Wales, Australia | [g] Neuroscience Research Australia, Australia

Correspondence: [*] Correspondence to: Dr. Nasser Bagheri, 63 Eggleston Road, Acton, ACT 2601, Australia. Tel.: +61 2 6125 9564; E-mail: Nasser.bagheri@anu.edu.au.

Abstract: Background:Dementia is a major global health challenge and the impact of built and social environments’ characteristics on dementia risk have not yet been fully evaluated. Objective:To investigate associations between built and social environmental characteristics and diagnosed dementia cases and estimated dementia risk. Methods:We recruited 25,511 patients aged 65 and older from family physicians’ practices. We calculated a dementia risk score based on risk and protective factors for patients not diagnosed with dementia. Our exposure variables were estimated for each statistical area level 1: social fragmentation, nitrogen dioxide, public open spaces, walkability, socio-economic status, and the length of main roads. We performed a multilevel mixed effect linear regression analysis to allow for the hierarchical nature of the data. Results:We found that a one standard deviation (1-SD) increase in NO2 and walkability score was associated with 10% higher odds of any versus no dementia (95% CI: 1%, 21% for NO2 and 0%, 22% for walkability score). For estimated future risk of dementia, a 1-SD increase in social fragmentation and NO2 was associated with a 1% increase in dementia risk (95% CI: 0, 1%). 1-SD increases in public open space and socioeconomic status were associated with 3% (95% CI: 0.95, 0.98) and 1% decreases (95% CI: 0.98, 0.99) in dementia risk, respectively. There was spatial heterogeneity in the pattern of diagnosed dementia and the estimated future risk of dementia. Conclusion:Associations of neighborhood NO2 level, walkability, public open space, and social fragmentation with diagnosed dementia cases and estimated future risk of dementia were statistically significant, indicating the potential to reduce the risk through changes in built and social environments.

Keywords: Dementia, green spaces, social fragmentation, spatial analysis, walkability

DOI: 10.3233/JAD-210208

Journal: Journal of Alzheimer's Disease, vol. 84, no. 2, pp. 621-632, 2021