Rui Luo, Fanghui Shi, Xiaoming Li, Yuejiao Zhou, Shuaifeng Liu, Xueying Yang
Resilience occupies a central role in the stress-psychophysiological network among PLWH. Key central and bridge nodes may serve as hypothesis-generating candidates for future longitudinal and intervention research.
OBJECTIVES: This study applied network analysis to characterize an integrated psycho-physiological stress network, identify central and bridge nodes, and determine psychological nodes most proximally associated with stress biomarkers among people living with HIV (PLWH).
METHODS: Baseline data were drawn from 800 PLWH in Guangxi, China. Perceived stress and resilience were assessed using validated scales, and cortisol/DHEA ratio was quantified from hair samples. Gaussian graphical models integrating cortisol/DHEA, perceived stress, and resilience were estimated using EBICglasso. Expected influence, bridge centrality, and network stability were evaluated. Sex-based differences were examined using the Network Comparison Test. Two sensitivity analyses examined robustness to biomarker skewness and sociodemographic confounding.
RESULTS: "Achieving despite obstacles" (EI = 1.226) was the most central node, followed by "Coping with uncertainty" (EI = 1.185) and "Growth through stress" (EI = 1.105). "Mastery over life" (Bridge EI = 0.142) emerged as the key bridge node. Among the weak biomarker-related edges, "Upset by unexpected events" showed the largest positive partial correlation with cortisol/DHEA, whereas "Growth through stress" showed the largest negative partial correlation. All cortisol/DHEA-related edges were small in magnitude (|partial r| < 0.04). No significant sex differences were found in global network structure; however, "Growth through stress" showed nominally higher centrality in males than females (1.126 vs. 0.839, uncorrected p = 0.032), a tentative difference not corrected for multiple comparisons.
CONCLUSION: Resilience occupies a central role in the stress-psychophysiological network among PLWH. Key central and bridge nodes may serve as hypothesis-generating candidates for future longitudinal and intervention research.