Tracking symptom dynamics in people living with HIV on different treatment: a longitudinal cross-lagged panel network analysis

Objective To characterize dynamic interrelationships among physical, cognitive, and psychological symptoms in people living with HIV and identify bridge symptoms between subgroups receiving distinct antiretroviral therapy regimens.Methods A longitudinal design was employed in this study. A total of...

Full description

Saved in:
Bibliographic Details
Main Authors: Meilian Xie, Xiaoyu Liu, Yanping Yu, Zhiyun Zhang, Li Zhang, Jieli Zhang, Dongxia Wu
Format: Article
Language:English
Published: Taylor & Francis Group 2025-12-01
Series:Annals of Medicine
Subjects:
Online Access:https://www.tandfonline.com/doi/10.1080/07853890.2025.2541090
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Objective To characterize dynamic interrelationships among physical, cognitive, and psychological symptoms in people living with HIV and identify bridge symptoms between subgroups receiving distinct antiretroviral therapy regimens.Methods A longitudinal design was employed in this study. A total of 676 individuals diagnosed with HIV filled out the Self-Report Symptom Checklist (HIV-SRSC) at baseline, 3-month, and 6-month follow-ups in the clinic. We evaluated bridge symptoms—those that connect various communities—within longitudinal networks utilizing cross-lagged panel network analyses (CLPN).Results The longitudinal networks presented differences and similarities between the Traditional Medication Regimen Group (TMR group) and the Novel Medication Regimen Group (NMR group). The expected influence of bridging symptoms in the CLPN predominantly centers on the cognitive symptom clusters, both in the TMR andNMR groups. However, the TMR group has stronger directional connections compared to the NMR group. ‘Memory loss’ (COGS3) demonstrates a bridging influence 1.62 (T1→T2) versus 1.51 (T2→T3) in the NMR group, and ‘Having difficulty in concentrating’ (COGS1) demonstrates a bridging influence 1.65 (T1→T2) versus 1.47 (T2→T3) in the TMR group. Nevertheless, physical symptoms, such as clusters related to gastrointestinal issues (PHYS12, PHYS13 and PHYS14), showed stable connections across both T1→T2 and T2→T3.Conclusion While TMR amplified cognitive-to-psychological symptom cascades, NMR fragmented cross-domain connectivity. These findings necessitate regimen-personalized interventions that preemptively target cognitive symptoms in TMR recipients while stabilizing bridge symptoms in NMR patients. Furthermore, cognitive dysfunction is a high-priority therapeutic target across all ART classes, which provides mechanistic evidence for ART-specific symptom management frameworks.
ISSN:0785-3890
1365-2060