Emerging evidence suggests that the built environment (BE) and social vulnerability of communities play a critical role in the Alzheimer's disease (AD) dementia burden. However, to what extent they are predictive of dementia burden nationally and regionally remains understudied.
Objective
This study investigates the cross-sectional associations between county-level BE and social vulnerability index (SVI) variables and AD dementia prevalence across the US and identifies the key variables within each region.
Methods
We applied eXtreme Gradient Boosting (XGBoost) and Random Forest (RF) models nationally and regionally using a combined set of BE and SVI variables. Variable importance was assessed using SHapley Additive Explanations values.
Results
Significant regional differences in AD dementia prevalence were observed (F = 126.56, p < 0.05). XGBoost consistently outperformed RF when using SVI and BE variables, with the highest predictive accuracy observed in the Northeast (R² = 0.92), followed by the South (R² = 0.73), West (R² = 0.72), and Midwest (R² = 0.72). In the Northeast, BE variables alone explained most of the variance, with grocery store density and walkability emerging as key variables. In contrast, the BE-only model for the South performed poorly but improved substantially with the inclusion of SVI variables. In the Midwest and West regions, both BE and SVI variables contributed more evenly to the model predictions.
Conclusions
While BE variables are more influential in urbanized regions with increased infrastructure (Northeast), SVI variables play a stronger role in more socioeconomically disadvantaged regions (South), underscoring the need for region-specific interventions.
ClarkePJWeuveJBarnesL, et al.Cognitive decline and the neighborhood environment. Ann Epidemiol2015; 25: 849–854.
2.
SylversDLHickenMEspositoM, et al.Walkable neighborhoods and cognition: implications for the design of health promoting communities. J Aging Health2022; 34: 893–904.
3.
MollaloAKnoxSMengJ, et al.Geospatial analysis of the association between Medicaid expansion, minimum wage policies, and Alzheimer’s disease dementia prevalence in the United States. Information (Basel)2024; 15: 688.
4.
CicioraDVásquezEValachovicE, et al.Social and behavior factors of Alzheimer's disease and related dementias: a national study in the U.S. Am J Prev Med2024; 66: 573–581.
5.
Centers for Disease Control and Prevention. Place and Health - Geospatial Research, Analysis, and Services Program (GRASP). 2025 [cited 2025 Mar 1], https://www.atsdr.cdc.gov/place-health/php/svi/
6.
DesaiPBondJDhanaA, et al.The social vulnerability index and incidence of Alzheimer disease in a population-based sample of older adults. Neurology2025; 104: e213464.
7.
PowerMCAdarSDYanoskyJD, et al.Exposure to air pollution as a potential contributor to cognitive function, cognitive decline, brain imaging, and dementia: a systematic review of epidemiologic research. Neurotoxicology2016; 56: 235–253.
8.
CacciottoloMWangXDriscollI, et al.Particulate air pollutants, APOE alleles and their contributions to cognitive impairment in older women and to amyloidogenesis in experimental models. Transl Psychiatry2017; 7: e1022.
9.
GuanCTanJHallB, et al.The effect of the built environment on the COVID-19 pandemic at the initial stage: a county-level study of the USA. Sustainability2022; 14: 3417.
10.
HuangHShiHZordanM, et al.Investigating the spatiotemporal relationship between the built environment and COVID-19 transmission. ISPRS Int J Geo-Inf2023; 12: 390.
11.
SturgeJNordinSPatilDS, et al.Features of the social and built environment that contribute to the well-being of people with dementia who live at home: a scoping review. Health Place2021; 67: 102483.
12.
WardRKeadyJVernooij-DassenM, et al.The lived neighborhood: understanding how people with dementia engage with their local environment. Int Psychogeriatr2018; 30: 867–880.
13.
MollaloAGrekousisGFlorezH, et al.Alzheimer's disease dementia prevalence in the United States: a county-level spatial machine learning analysis. Am J Alzheimers Dis Other Demen2025; 40: 15333175251335570.
14.
MurdochWJSinghCKumbierK, et al.Definitions, methods, and applications in interpretable machine learning. Proc Natl Acad Sci U S A2019; 116: 22071–22080.
15.
NoriVSHaneCACrownWH, et al.Machine learning models to predict onset of dementia: a label learning approach. Alzheimers Dement (N Y)2019; 5: 433–441.
16.
DaduATaMTustisonNJ, et al.Prediction, prognosis and monitoring of neurodegeneration at biobank-scale via machine learning and imaging. medRxiv 2024. DOI: 10.1101/2024.10.27.24316215[Preprint]. Posted October 28, 2024.
17.
ThompsonPMRajT. Artificial intelligence and machine learning (AIML) approaches to empower genomics, drug, and biomarker discovery in ADRD. Alzheimers Dement2023; 19: 1284–1296.
18.
LundbergSLeeS. A unified approach to interpreting model predictions. arXiv Prepr 2017; arXiv:1705.07874.
19.
DinticaCSBahorikAXiaF, et al.Dementia risk and disadvantaged neighborhoods. Arch Neurol (Chicago)2023; 80: 944–950.
MarquardtGBueterKMotzekT. Impact of the design of the built environment on people with dementia: an evidence-based review. HERD2014; 8: 127–157.
22.
KolakMBhattJParkYH, et al.Quantification of neighborhood-level social determinants of health in the continental United States. JAMA Netw Open2020; 3: e1919928.
23.
DhanaKBeckTDesaiP, et al.Prevalence of Alzheimer's disease dementia in the 50 US states and 3142 counties: a population estimate using the 2020 bridged-race postcensal from the National Center for Health Statistics. Alzheimers Dement2023; 19: 1988–1999.
PruittSLEberthJMMorrisES, et al.Rural-urban differences in late-stage breast cancer: do associations differ by rural-urban classification system?Tex Public Health J2015; 67: 12–18.
JamesGWittenDHastieT, et al.An introduction to statistical learning: with applications in R. 1st ed. New York: Springer, 2013.
32.
NoriVSHaneCASunY, et al.Deep neural network models for identifying incident dementia using claims and EHR datasets. PLoS One2020; 15: e0236400.
33.
YiFYangHChenD, et al.XGBoost-SHAP-based interpretable diagnostic framework for Alzheimer’s disease. BMC Med Inform Decis Mak2023; 23: 238.
34.
RamrajSUzirNRamanS, et al.Experimenting XGBoost algorithm for prediction and classification of different datasets. arXiv Prepr 2022.
35.
ChengLDe VosJZhaoP, et al.Examining non-linear built environment effects on elderly’s walking: a random forest approach. Transp Res D Transp Environ2020; 83: 102552.
36.
BreimanL. Random forests. Mach Learn2001; 45: 5–32.
LundbergSMNairBVavilalaMS, et al.Explainable machine-learning predictions for the prevention of hypoxaemia during surgery. Nat Biomed Eng2018; 2: 749–760.
39.
ChenTGuestrinC. XGBoost. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, New York, NY, 2016, pp.785–794: ACM.
40.
PedregosaFVaroquauxGGramfortA, et al.Scikit-learn: machine learning in Python. J Mach Learn Res2011; 12: 2825–2830.
41.
ProbstPWrightMNBoulesteixA. Hyperparameters and tuning strategies for random forest. WIRES Data Min Knowl Discov2019; 9: e1301.
42.
SommerJSarigiannisDParnellT. Learning to tune XGBoost with XGBoost. arXiv Prepr 2019; arXiv:1909.07218.
43.
GoodfellowIBengioYCourvilleA. Deep learning. Cambridge, MA: MIT Press, 2016.
44.
CavanaughJENeathAA. The Akaike information criterion: background, derivation, properties, application, interpretation, and refinements. WIRES Comput Stat2019; 11: e1460.
45.
KramerMCuttyMKnoxS, et al.Rural–urban disparities of Alzheimer's disease and related dementias: a scoping review. Alzheimers Dement (N Y)2025; 11: e70047.
46.
LiuAJMukherjeeAHuL, et al.Performance and interpretability comparisons of supervised machine learning algorithms: An empirical study. arXiv Prepr 2022; arXiv:2204.12868.
47.
MollaloAKramerMCuttyM, et al.Systematic review and meta-analysis of rural-urban disparities in Alzheimer’s disease dementia prevalence. J Prev Alzheimers Dis2025; 12: 234–242.
48.
Van CauwenbergJVan HolleVDe BourdeaudhuijI, et al.Neighborhood walkability and health outcomes among older adults: the mediating role of physical activity. Health Place2016; 37: 26–34.
49.
McCormackGRFrehlichLBlackstaffeA, et al.Active and fit communities: associations between neighborhood walkability and health-related fitness in adults. Int J Environ Res Public Health2020; 17: 1131.
50.
DarashagamN. Investigation of the impact of neighborhood environments on Alzheimer’s disease and Alzheimer’s disease-related dementias (AD/ADRD). Thesis, University of Texas at Arlington, 2024.
51.
WrightRSGerassimakisCBygraveD, et al.Dietary factors and cognitive function in poor urban settings. Curr Nutr Rep2017; 6: 32–40.
52.
HenneyAEGillespieCSAlamU, et al.High intake of ultra-processed food is associated with dementia in adults: a systematic review and meta-analysis of observational studies. J Neurol2024; 271: 1671–1684.
EvansIEMMartyrACollinsR, et al.Social isolation and cognitive function in later life: a systematic review and meta-analysis. J Alzheimers Dis2019; 70: S119–S144.
55.
Al-RousanTMRubensteinLMWallaceRB. Disability levels and correlates among older mobile home dwellers, an NHATS analysis. Disabil Health J2015; 8: 271–278.
56.
ChristopherASMcCormickDWoolhandlerS, et al.Access to care and chronic disease outcomes among Medicaid-insured persons versus the uninsured. Am J Public Health2016; 106: 63–69.
57.
ShiLStevensGD. The role of community health centers in delivering primary care to the underserved: experiences of the uninsured and Medicaid insured. J Ambul Care Manage2007; 30: 159–170.
58.
BartakMTothPGavurovaB. Alzheimer’s disease mortality according to socioeconomic factors: country study. Int J Alzheimers Dis2018; 2018: 8137464.
59.
JinRLShahCPSvobodaTJ. The impact of unemployment on health: a review of the evidence. CMAJ1995; 153: 529–540.
60.
NgCDZhangPKowalS. Validating the social vulnerability index for alternative geographies in the United States to explore trends in social determinants of health over time and geographic location. Front Public Health2025; 13: 1547946.
61.
RahmanMARahmanMSUddinMJ, et al.Emerging risk of environmental factors: insight mechanisms of Alzheimer’s diseases. Environ Sci Pollut Res Int2020; 27: 12295–12309.
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
