Association between depression and parkinson based on the PHQ-9 score and MESD-10 score: Evidence from NHANES and CHARLS

Abstract Background Depression, a prevalent non-motor symptom in Parkinson’s disease (PD), demonstrates complex associations with PD risk that remain incompletely characterized. This cross-sectional study investigates the consistency and non-linear dynamics of depression score-PD associations across Western (NHANES) and Asian (CHARLS) populations using standardized metrics (PHQ-9 and CESD-10). Methods We analyzed data from 34,308 U.S. (364 PD cases) and 19,395 Chinese participants (197 PD cases). Weighted multivariable logistic regression models assessed depression score-PD relationships adjusted for sociodemographic and clinical confounders. Non-linear relationships and threshold effects were examined via smooth curve fitting and segmented regression.ROC analysis compared the diagnostic accuracy of depression scales. Results Elevated depressive symptom scores were observed in PD cases versus controls (PHQ-9: 6.07 vs. 3.00; CESD-10: 12.63 vs. 8.56; both P < 0.001). Adjusted models revealed strong, dose-dependent associations (NHANES: OR = 1.12 per PHQ-9 unit, 95% CI 1.09–1.14; CHARLS: OR = 1.08 per CESD-10 unit, 95% CI 1.05–1.10; both P < 0.0001). Participants in the highest depression quartiles exhibited 5.80-fold (NHANES) and 5.12-fold (CHARLS) PD risk increments versus Q1. Non-linear analysis identified threshold effects at PHQ-9/CESD-10 scores of 15 (P < 0.001). Subgroup analyses revealed stronger associations in younger individuals and Mexican Americans (p for interaction < 0.05). PHQ-9 and CESD-10 demonstrated moderate discriminative accuracy (AUC 0.672 vs. 0.670, respectively). Conclusions Depression severity exhibits robust, cross-population associations with PD risk, characterized by dose-response relationships and threshold effects. Standardized depression scales (PHQ-9/CESD-10) may serve as pragmatic tools for PD risk stratification, particularly below score thresholds of clinical relevance.