NHANES 2005–2018 data reveal high albumin-bilirubin scores are associated with depression
Abstract Background Depression has been recognized as a leading global contributor to disease burden and mortality. While the etiology of depression involves a complex interplay of biological, psychological, and social factors, its precise pathophysiological mechanisms remain incompletely understood...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2025-07-01
|
| Series: | BMC Psychiatry |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12888-025-07082-x |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Abstract Background Depression has been recognized as a leading global contributor to disease burden and mortality. While the etiology of depression involves a complex interplay of biological, psychological, and social factors, its precise pathophysiological mechanisms remain incompletely understood. The albumin-bilirubin (ALBI) score, originally developed as an objective measure of hepatic function, has demonstrated prognostic value in diverse clinical contexts including liver diseases, brain tumors and chronic heart failure. However, the relationship between ALBI scores and depression has not been systematically investigated. This study aims to examine the association between ALBI scores and depression in a nationally representative cohort. Methods This population-based cross-sectional study utilized data from the National Health and Nutrition Examination Survey (NHANES) spanning seven survey cycles (2005–2018). The ALBI score was calculated using the established formula: ALBI = (log10 bilirubin [μmol/L] × 0.66) + (albumin [g/L] × -0.085). Depression was assessed using the Patient Health Questionnaire-9 (PHQ-9). Weighted multivariable logistic regression models were constructed to analyze the association between ALBI and depression, adjusting for potential covariates. Nonlinear relationships were explored using restricted cubic splines (RCS). The robustness of the results was validated through subgroup analyses. The relationship between ALBI and depression symptoms severity was evaluated via ANOVA. For prognostic analysis, Cox proportional hazards regression models assessed the link between ALBI and clinical outcomes, complemented by Kaplan–Meier (K-M) survival curves and RCS plots to illustrate trends. Results Weighted multivariable logistic regression models revealed a significant association between elevated ALBI scores and depression risk. Participants in the higher ALBI (ALBI≧ -2.87) demonstrated 16% increased odds of depression compared to the reference (ALBI < -2.87) (adjusted OR = 1.16, 95% CI 1.03–1.31, P = 0.01). Restricted cubic spline analysis confirmed a linear relationship between continuous ALBI scores and depression risk. Subgroup analyses stratified by age, gender, race, education, smoking, drinking and diabetes consistently identified high ALBI as an independent risk factor for depression. The analysis of variance revealed a significant dose–response relationship between ALBI and depression symptom severity (P < 0.05). Cox proportional hazards models revealed significant associations between ALBI increments and mortality risks in the depression cohort. For all-cause mortality, each 1-unit ALBI elevation was associated with a 284% increased risk (adjusted HR = 3.84, 95% CI 2.77–5.33, P < 0.01). Cancer-related mortality showed a 235% risk escalation per ALBI unit (HR = 3.35, 95% CI 1.54–7.26, P < 0.01), while non-cancer deaths exhibited an stronger association at 296% increased risk (HR = 3.96, 95% CI 2.76–5.67, P < 0.01). Conclusion By analyzing data from seven cycles (2005–2018) of the NHANES database, this study identified a positive association between the ALBI score and depression among U.S. adults. The ALBI score was also linked to the severity of depressive symptoms, and was associated with all-cause mortality, cancer-related mortality in individuals with depression. Trial registration Clinical trial number: not applicable. |
|---|---|
| ISSN: | 1471-244X |