Sex-Specific Dementia Risk Factors Across 19 Countries, 2010-2020

39,439 participants aged 65+ from four longitudinal aging studies across 19 countries were analyzed for incident late-onset dementia. The study, published on figshare in 2026, includes 3,548 recorded dementia cases and uses feature selection models like LASSO and Boruta to identify sex-specific modifiable risk factors.

Use Cases

• Modeling sex-specific dementia risk based on longitudinal cohort data from 19 countries.
• Analyzing the interaction between alcohol consumption and age as a risk factor for dementia.
• Comparing modifiable risk factors for dementia across different geographic regions and cohorts.
• Training predictive models for late-onset dementia incidence using selected features like age and drinking status.

Strengths

• Data aggregates 39,439 participants from four major longitudinal aging studies.
• Analysis spans 19 countries and a 10-year period (2010-2020).
• Employs rigorous feature selection methods (LASSO, Boruta) and Cox proportional hazards models.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count for the underlying analytical dataset is unknown, which may limit suitability assessment.
• The primary file format is PDF, which may require extraction for computational analysis.

Provenance

Source

Aggregated from four cohort studies: CHARLS (China), SHARE (Europe), HRS (USA), and ELSA (England).

Collection Method

Longitudinal cohort analysis using feature selection and Cox proportional hazards models.

Time Range

2010 to 2020

Freshness

Last updated 2026-04 07:55:11; freshness should be verified.

Geography

19 countries across North America, Europe, and China.