Abstract
BACKGROUND: Depressive symptom, a pervasive mental health disorder, has garnered increasing attention due to its intricate interconnections with various physiological processes. One emerging avenue of investigation delves into the potential association between depressive symptom and Homeostasis Model Assessment of Insulin Resistance (HOMA-IR), a parameter reflecting insulin resistance. The intricate interplay between these two domains holds promising implications for understanding the multifaceted nature of depressive symptom and its impact on metabolic health. METHODS: We used weighted multivariable logistic regression models with subgroup analysis to explore the relationship between depressive symptom and homeostasis model assessment of insulin resistance. Non-linear correlations were explored using fitted smoothing curves. Then, we constructed a two-piece linear regression model and performed a recursive algorithm to calculate the inflection point. RESULTS: The study included 20,282 participants in the United States. In the regression model adjusted for all confounding variables, the odds ratio (OR) for the correlation between depressive symptom and the Homeostasis Model Assessment of Insulin Resistance (HOMA-IR) was 1.01 (95% CI: 1.00, 1.01). However, a significant discrepancy between trend tests and regression analyses suggests a potential non-linear relationship between depressive symptom and the assessment of insulin resistance using the Homeostasis Model. Constrained cubic spline analysis confirmed this non-linear relationship, identifying an inflection point at 10.47. Before the inflection point, depressive symptom exhibited a significantly positive correlation with the assessment of insulin resistance using the Homeostasis Model. However, after the inflection point, a negative correlation was observed, though it did not reach statistical significance. CONCLUSION: We found a curve-like relationship between depressive symptom and homeostasis model assessment of insulin resistance.