Relationship of the metabolic score for insulin resistance and the new-onset hypertension: Evidence from CHARLS.

Abstract

Hypertension (HTN) progression is linked to insulin resistance (IR), yet the association between Metabolic Score for Insulin Resistance (METS-IR) and HTN remains underexplored. This study included 4,051 individuals without a history of HTN from the China Health and Retirement Longitudinal Study (CHARLS). Participants were stratified into four groups according to their baseline METS-IR values. It was the development of incident HTN that was the primary outcome. We used Cox regression to assess this association, conducted subgroup and sensitivity analyses, and evaluated METS-IR's incremental predictive value over conventional risk factors (age, sex, systolic blood pressure) using C-statistic, net reclassification improvement (NRI), integrated discrimination improvement (IDI), and decision curve analysis (DCA). Over the 9-year follow-up, 1,572 participants (38.81%) experienced their first incident of HTN. Participants were categorized into quartiles (Q1-Q4) based on their METS-IR levels. After full adjustment for confounders, the hazard ratio (HR) with a 95% confidence interval (CI) for incident HTN demonstrated a progressive increase across ascending METS-IR quartiles, with Q1 as reference: Q2, 0.99 (0.85-1.15); Q3, 1.17 (1.01-1.36); Q4, 1.31 (1.13-1.52). The restricted cubic spline (RCS) model revealed a linear dose-response relationship between METS-IR and the incidence of HTN (P for overall trend < 0.001; P for nonlinear = 0.310). Adding METS-IR to a base model (age/sex/systolic blood pressure) improved HTN prediction (C-statistic Δ= + 0.004; NRI = 16.58%, IDI = 0.75%; all P < 0.001). DCA confirmed greater net benefit across risk thresholds. Subgroup and sensitivity analyses consistently supported the primary findings. Elevated METS-IR independently predicts HTN risk in Chinese adults, suggesting METS-IR as a potential indicator.