Social Isolation and Mortality in U.S. Stroke Survivors, 1999-2018 NHANES Data

1,284 participants aged 40+ with a self-reported stroke history from the 1999–2018 National Health and Nutrition Examination Survey. This retrospective cohort study by Ying Zhou investigates the association between social isolation and all-cause/CVD mortality, with mediation analysis on systemic inflammation. The dataset includes hazard ratios, confidence intervals, and population-attributable fractions derived from Cox proportional hazards models.

Use Cases

• Modeling mortality risk based on social isolation status and inflammation indicators.
• Conducting mediation analysis to quantify the role of systemic inflammation in health outcomes.
• Estimating population-attributable fractions for social isolation as a risk factor.
• Training survival analysis models using hazard ratios and confidence intervals from cohort data.

Strengths

• Includes specific, published results: hazard ratios of 1.57 for all-cause and 2.49 for CVD mortality.
• Clear temporal coverage: data drawn from the 1999–2018 National Health and Nutrition Examination Survey.
• Mediation analysis provides quantified contributions: SIRI explained 10.40% and 5.37% of associations.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment for large-scale modeling.
• Data may reflect geographic and temporal bias inherent to the NHANES source survey.

Provenance

Source

National Health and Nutrition Examination Survey (NHANES)

Collection Method

Retrospective cohort study using survey data and mortality follow-up.

Time Range

1999-2018

Freshness

Last updated 2026-04-27 01:00:03; freshness should be verified.

Geography

United States