EUVIMEDEUVIMED entdecken
Crossref

Differential Biomarker Associations of Anxiety Dimensions: A Psycho-Biological Network Analysis with External Validation

Daria Volontovich, Musawir Abbas Khan, Shi Jiang, Yashuang Zhao

Neuropsychobiology · 09.11.2026

DOI: 10.1159/nps/adyag010 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Neuropsychobiology, online erschienen am 09.11.2026. Verfasst von Daria Volontovich und 3 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Introduction

Anxiety involves dysregulation across peripheral biological systems, yet associations with specific dimensions remain unclear. We constructed a psycho-biological network linking anxiety dimensions with biomarkers spanning inflammation, the hypothalamic-pituitary-adrenal (HPA) axis, the sympathetic-adrenal-medullary (SAM) axis, and metabolic regulation.

Methods

We conducted a cross-sectional network analysis using the Midlife in the United States (MIDUS) 3 Biomarker Project data (n=747). Four network estimation approaches were compared, with EBICglasso selected as the primary model. An a priori psycho-biological mask was applied, and edge stability, bridge centrality, and community structure were assessed. The pipeline was validated in MIDUS 2 (n=1,255) and MIDUS Refresher 1 (n=863).

Results

Anxious arousal (MASQ(AA)) showed the strongest cross-system connectivity. The MASQ(AA)–C-reactive protein (CRP) association was the only significant edge in the primary network (p=0.022, 95% CI [0.019, 0.164]) but was not replicated in validation cohorts. Four candidate cross-system connectors were identified (MASQ(AA), CRP, glucose, and TNFa), yet only MASQ(AA) was replicated across validation cohorts. Inflammatory and metabolic interfaces showed limited reproducibility; HPA-axis and peripheral catecholaminergic biomarkers did not form stable cross-system bridges. Network architecture showed poor reproducibility (Jaccard=0.172 and 0.121; Spearman ρ=0.296 and 0.225 for MIDUS 2 and Refresher 1, respectively), with limited community preservation.

Conclusion

Anxiety dimensions showed differentiated but only partially reproducible psycho-biological connectivity. Anxious arousal represents the most reproducible cross-system dimension, whereas the broader organization of anxiety-related peripheral biology is cohort-dependent. These findings support dimensional network analysis and the need for external validation and longitudinal investigation.

Zitieren

APA

Volontovich, D., Khan, M. A., Jiang, S., & Zhao, Y. (2026). Differential Biomarker Associations of Anxiety Dimensions: A Psycho-Biological Network Analysis with External Validation. Neuropsychobiology. https://doi.org/10.1159/nps/adyag010

Vancouver

Volontovich D, Khan MA, Jiang S, Zhao Y. Differential Biomarker Associations of Anxiety Dimensions: A Psycho-Biological Network Analysis with External Validation. Neuropsychobiology. 2026. doi:10.1159/nps/adyag010

BibTeX
@article{Volontovich2026,
  author = {Volontovich, Daria and Khan, Musawir Abbas and Jiang, Shi and Zhao, Yashuang},
  title = {Differential Biomarker Associations of Anxiety Dimensions: A Psycho-Biological Network Analysis with External Validation},
  journal = {Neuropsychobiology},
  year = {2026},
  doi = {10.1159/nps/adyag010}
}
Kostenlos starten →