Abstract
Background
Survival after an Alzheimer's disease (AD) diagnosis is vital for patients, their families, caregivers, and healthcare providers. Hawaii, known for its diverse ethnic population, exhibits significant racial health disparities.
Objective
This study examined racial/ethnic and socioeconomic disparities in AD survival in Hawaii and developed machine learning models to predict overall survival using Hawaii Medicare data.
Methods
Nine years of Hawaii Medicare data were utilized to gather information on AD development after age 65, following patients to capture all-cause survival or until censoring. The study examined the effects of race/ethnicity and socioeconomic status (SES) on mortality risk. Cox regression analysis was conducted on overall survival, accounting for covariates. A Survival Random Forest was employed to model survival, incorporating K years of longitudinal health profiles.
Results
The study included 9393 AD subjects. Analysis revealed that Asian Americans (AA) had a later age at AD diagnosis (p < 0.001), with an average age of 85.9, compared to 82.7 and 83.3 years for whites and Native Hawaiians and Pacific Islanders (NHPI), respectively. Low SES showed a marginal increase in hazard (Hazard Ratio [HR] = 1.36, p < 0.001). After covariate adjustment, compared to AAs with better SES, increased hazards were found for their white counterpart (HR = 1.18, p < 0.001) and groups with low SES: AA (HR = 1.28, p < 0.001), white (HR = 1.51, p < 0.001), and NHPI (HR = 1.39, p < 0.001). The predictive model had a Concordance-Index of 0.82, showing reasonable predictability.
Conclusions
Racial/ethnic and SES disparities significantly influence AD onset and survival. Combined with longitudinal health status data, machine learning demonstrates reasonable predictability of survival.
Get full access to this article
View all access options for this article.
References
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.
