ABSTRACT
Introduction: Clinical notes, biomarkers, and neuroimaging have been proven valuable in dementia prediction models. Whether commonly available structured clinical data can predict dementia is an emerging area of research. We aimed to predict Alzheimer's disease (AD) and Alzheimer's disease related dementias (ADRD) in a well-phenotyped, population-based cohort using a machine learning approach. Methods: Administrative healthcare data (k=163 diagnostic features), in addition to Census/vital record sociodemographic data (k = 6 features), were linked to the Cache County Study (CCS, 1995-2008). Results: Among successfully linked UPDB-CCS participants (n=4206), 522 (12.4%) had incident AD/ADRD as per the CCS "gold standard" assessments. Random Forest models, with a 1-year prediction window, achieved the best performance with an Area Under the Curve (AUC) of 0.67. Accuracy declined for dementia subtypes: AD/ADRD (AUC = 0.65); ADRD (AUC = 0.49). DISCUSSION: Commonly available structured clinical data (without labs, notes, or prescription information) demonstrate modest ability to predict AD/ADRD, corroborated by prior research.
ABSTRACT
To address the numerous challenges associated with aging, increased attention has been given to the development of age-friendly cities and communities (AFCC) to promote healthy aging and social participation. However, limited evidence exists for addressing both health and social needs through the AFCC framework. We address this gap by conducting a scoping review of the interventions that target both health and social outcomes within the purview of the AFCC framework. The results showed that many of the successful interventions used a partnership model and behavioral change theories to inform the program design and implementation. The results also indicated that social participation and engagement played a key role in making the interventions successful. However, the results revealed that the literature is dominated by person-focused approaches. Future research should focus more on evaluating environmental-focused interventions and develop a holistic framework that combines both person- and environment-based approaches to healthy aging.
Subject(s)
Aging , Humans , CitiesABSTRACT
BACKGROUND: Prior research indicates that at least 35% of Alzheimer's disease and related dementia risk may be amenable to prevention. Subjective cognitive decline is often the first indication of preclinical dementia, with the risk of subsequent Alzheimer's disease in such individuals being greater in women than men. We wished to understand how modifiable factors are associated with subjective cognitive decline, and whether differences exist by sex. METHODS: Data were collected from men and women (45 years and older) who completed the U.S. Behavioral Risk Factor Surveillance System Cognitive Decline Module (2015-2018), n = 216,838. We calculated population-attributable fractions for subjective cognitive decline, stratified by sex, of the following factors: limited education, deafness, social isolation, depression, smoking, physical inactivity, obesity, hypertension, and diabetes. Our models were adjusted for age, race, income, employment, marital and Veteran status, and accounted for communality among risk factors. RESULTS: The final study sample included more women (53.7%) than men, but both had a similar prevalence of subjective cognitive decline (10.6% of women versus 11.2% of men). Women and men had nearly equivalent overall population-attributable fractions to explain subjective cognitive decline (39.7% for women versus 41.3% for men). The top three contributing risk factors were social isolation, depression, and hypertension, which explained three-quarters of the overall population-attributable fraction. CONCLUSIONS: While we did not identify any differences in modifiable factors between men and women contributing to subjective cognitive decline, other factors including reproductive or endocrinological health history or biological factors that interact with sex to modify risk warrant further research.
