Supplementary file 1_Early-pregnancy BMI-derived adiposity change rate and gestational diabetes mellitus risk: modification by a fatty liver index–defined hepatic-metabolic phenotype.docx

Multivariate logistic regression, restricted cubic splines, threshold analysis, subgroup interaction testing, exploratory mediation analysis, and propensity score–matched sensitivity analyses were also performed. After adjustment for maternal age, parity, pre-pregnancy BMI, a higher eBFP change rate was associated with increased odds of subsequent GDM (odds ratio [OR], 1.58; 95% confidence interval [CI], 1.14–2.21; P = 0.006), with an approximately linear dose–response pattern.

The FLI_Ln showed a nonlinear association with GDM, with a data-derived threshold of 2.90. The eBFP change rate–GDM association was evident among women with FLI_Ln ≥2.90 (OR, 1.82; 95% CI, 1.22–2.71; P = 0.004), but not among those with FLI_Ln <2.90 (OR, 0.51; 95% CI, 0.22–1.20; P = 0.123; P for interaction = 0.018). Exploratory mediation analyses suggested that FLI_Ln partially accounted for the observed association, although the magnitude of this mediation was imprecisely estimated.

In this secondary cohort analysis, a faster early-pregnancy BMI-derived adiposity change was associated with a higher risk of GDM, particularly among women with an elevated composite hepatic–metabolic phenotype. These findings support an early adiposity–hepatic metabolic susceptibility framework for GDM risk enrichment but require external validation before clinical application.</p>