Data and code for albuminuria–TyG discordance and cardiovascular mortality in NHANES 1999–2018

This record contains the processed participant-level analysis dataset, data dictionary, source manifest, analysis scripts, figure-supporting datasets, and publication-quality Figures 1–5 for a secondary analysis of the National Health and Nutrition Examination Survey (NHANES), 1999–2018, with public-use linked mortality follow-up through December 31, 2019. The locked exposure cohort comprises 21,694 adults and 997 cardiovascular deaths. Albuminuria was defined as urinary albumin-to-creatinine ratio (UACR) ≥30 mg/g, and the triglyceride-glucose (TyG) index was split at its survey-weighted median of 8.575. Files support reproduction and audit of survey-weighted cause-specific Cox models, cumulative-incidence analyses, interaction, time-varying and sensitivity analyses, competing-risk sensitivity analyses, and manuscript figures. Upstream source records are public-use CDC/NCHS files; this record contains author-processed derivatives and a cycle-by-cycle provenance manifest. The 2019 public-use linked mortality files include NCHS disclosure-protection procedures for selected follow-up/cause-of-death fields, as documented by NCHS.