Association Analysis of Insulin Resistance Metabolic Score (METS‐IR) and Gestational Diabetes Mellitus: Based on National Health and Nutrition Examination Survey Database From 2007 to 2018

ABSTRACT Objective This study focused on the association of the Insulin resistance metabolic score (METS‐IR) with the risk of gestational diabetes mellitus (GDM) using data from the National Health and Nutrition Examination Survey (NHANES). Methods Data from 6 cycles of NHANES (2007–2018) were analysed. Weighted logistic regression models were constructed to explore the relationship between METS‐IR and GDM. Stratified and subgroup analyses with adjustment for confounding factors were carried out to explore the association between METS‐IR and GDM. Results A total of 5189 samples were analysed. Based on the weighted logistic regression model, Ln(METS‐IR) was positively associated with GDM with full adjustment (OR = 1.94, 95% CI 1.08–3.46, p < 0.005). After transferring Ln(METS‐IR) into a categorical variable by quartiles, the positive connection between Ln(METS‐IR) and GDM was still observed in the higher Ln(METS‐IR) group compared to the lowest Ln(METS‐IR) interval (OR of 1.86, 1.76 for participants in the Q3(3.73, 3.93) and Q4(3.93, 4.83) quartile, respectively, p < 0.05). The threshold effect model showed that when Ln(METS‐IR) ≤ 4, the positive correlation between Ln(METS‐IR) and GDM was more significant (β = 2.69, 95% CI 1.55–4.67, p = 0.0004). The area under the ROC curves of Ln(METS‐IR) for GDM was 0.603, suggesting Ln(METS‐IR) a more systematic predictor for GDM. Specifically, the OR and 95% CIs of GDM for women above high school in the Q2, Q3, and Q4 quartiles were 2.05 (1.04, 4.02), 3.41 (1.72, 6.78) and 2.78 (1.55, 4.99), respectively. Conclusion METS‐IR in women elevates the likelihood of GDM occurrence. METS‐IR serves as a comprehensive alternative to HOMA‐IR rather than HbA1c and non‐based insulin level to predict GDM.