Abstract
Background: Despite the agreement on the significance of the relationship between the C-reactive protein (CRP) and depression, research results have been discrepant by gender.
Objective: We attempted to address this uncertainty via a generalized additive model and more carefully analyzed the shape of the CRP-depression relationship in terms of sex.
Methods: This is a secondary data analysis using the National Health and Nutrition Examination Survey (2017-2020) data targeting 1,581 obese middle-aged adults (40-70 years, 51.4% women) with CRP <= 10 mg/L. The study outcome was depression severity, assessed by the nine-item Patient Health Questionnaire. Several models were built, adjusting for relevant sociodemographic, lifestyle, and clinical covariates. The value of the effective degree of freedom (EDF) quantifies curvature of the relationship. The threshold effect was investigated using a two-piecewise linear regression model, when needed.
Results: Among men, an increasing linear pattern was found (EDF [almost equal to] 1). Contrastingly among women, the EDF value was > 2 in all unadjusted and adjusted models, indicating the smooth (curved) association. The threshold level affected the association pattern particularly for women, among whom the depression severity related to CRP significantly increased as the CRP level increased to an inflection point of 3.6 mg/L but decreased thereafter.
Discussion: Assuming linearity for the CRP association with depression may not be appropriate for middle-aged obese women. Although we do not claim to provide a definite method of assessing the CRP-depression relationship, we hope to offer a different perspective when exploring this relationship. Thus, the results should be interpreted cautiously, and future studies on this topic should replicate this approach with generalized additive models.