Xiaoyan Hu,1,* Yuting Zhan,2,* Jinying Wang3 

1ElCU, Shaoxing Second Hospital, Shaoxing City, Zhejiang Province, People’s Republic of China; 2Department of Psychology, School of Education and Teaching, Ningxia University, Yinchuan City, Ningxia Province, People’s Republic of China; 3Department of Internal Medicine, Shaoxing Second Hospital, Shaoxing City, Zhejiang Province, People’s Republic of China

*These authors contributed equally to this work

Correspondence: Jinying Wang, Email Wjy15715826979@163.com

Background: Sleep quality has emerged as a critical public health concern, yet our understanding of how multiple determinants interact to influence sleep outcomes remains limited. This study employed partial correlation network analysis to examine the hierarchical structure of sleep quality determinants among Chinese adults.
Methods: We investigated the interrelationships among nine key factors: daily activity rhythm, social interaction frequency, work-life balance, light exposure, physical activity level, time control perception, shift work, weekend catch-up sleep, and sleep quality using the extended Bayesian Information Criterion (EBIC) glasso model. The study included 8,127 Chinese adults (51.0% female, mean age = 32.7 years).
Results: Results revealed that 79.9% of sleep quality variance could be explained by surrounding variables in the network. Time control perception emerged as a proximal factor, demonstrating the highest centrality (strength = 1.85, betweenness = 1.92, closeness = 1.88) and strongest connections to sleep quality. Behavioral factors (physical activity level, shift work, work-life balance) functioned as intermediate mechanisms, while environmental and temporal patterns (light exposure, weekend catch-up sleep, social interaction frequency, daily activity rhythm) operated as distal influences. Network stability analysis showed robust estimation precision (CS coefficients > 0.70 for all centrality measures).
Conclusion: These findings advance our theoretical understanding of sleep quality as embedded within a dynamic network of interacting factors and provide empirical support for targeted interventions focusing on time control perception and behavioral mediators to improve sleep outcomes. The network perspective offers novel insights for developing effective, hierarchically structured approaches to sleep quality enhancement in contemporary society.

Keywords: sleep quality, network analysis, time control perception, behavioral mediators, hierarchical structure, partial correlation