Introduction:
Evidence regarding the association of insulin resistance (IR) with chronic diarrhea (CD) remains scarce, especially in the absence of robust epidemiological support from large-scale population cohorts. We aimed to explore the association of IR with CD using a cross-sectional analysis of data from the National Health and Nutrition Examination Survey (NHANES) from 2005 to 2010.

Material and methods:
Multiple logistic regression models combined with subgroup analyses were leveraged to explore the association of IR (triglyceride-glucose [TyG] index, metabolic score for insulin resistance [METS-IR], and homeostasis model assessment of insulin resistance [HOMA-IR]) with the prevalence of CD. Nonlinear associations were assessed using restricted cubic spline (RCS) curves. Finally, the discriminative performance of these indicators was assessed using receiver operating characteristic (ROC) curves and the area under the curve (AUC).

Results:
After full adjustment for covariates (e.g., sex, age, race, education, poverty-to-income ratio (PIR), body mass index (BMI), smoking status), the TyG index (OR = 1.35, 95% CI: 1.12–1.63), METS-IR (OR = 2.00, 95% CI: 1.31–3.06), and HOMA-IR (OR = 1.02, 95% CI: 1.00–1.03) showed positive associations with the prevalence of CD. The RCS curve analysis showed a non-linear association between HOMA-IR and CD (p-non-linear = 0.003), with an inflection point at 2.423. METS-IR and TyG showed a significant positive linear association with CD (both p-non-linear > 0.05). The ROC curve analysis identified METS-IR as the best-discriminating indicator for CD (AUC = 0.645, 95% CI: 0.619–0.672).

Conclusions:
The present research identified a significant positive association between IR and CD among a demographically representative cohort of US adults. Elevated IR scores are significantly associated with the prevalence of CD.