Jing Yang,1,2,* Dongmei Xu,3,* Huanhuan Lu,4 Xuwen Yin,1 Haiyan Song,1 Weiwei Zong,5 Dandan Xu,5 Xiaohui Lu,5 Lan Wei,5 Hong Zhu,5 Shiyin Zhai,5 Zejuan Gu1 

1School of Nursing, Nanjing University of Chinese Medicine, Nanjing, People’s Republic of China; 2Nursing Department, The First Affiliated Hospital of Nanjing Medical University, Nanjing, People’s Republic of China; 3Rheumatology and Immunology Department, The First Affiliated Hospital of Nanjing Medical University, Nanjing, People’s Republic of China; 4School of Nursing, Nanjing Medical University, Nanjing, People’s Republic of China; 5Cardiovascular Medicine, The First Affiliated Hospital of Nanjing Medical University, Nanjing, People’s Republic of China

*These authors contributed equally to this work

Correspondence: Zejuan Gu, School of Nursing, Nanjing University of Chinese Medicine, Nanjing, 210029, People’s Republic of China, Email jassicagu@163.com

Objective: The Huawei Band 9 (HWB 9), a consumer sleep-monitoring device with a high market share and a large user base in China, can provide sleep staging parameters and has broad representativeness in sleep monitoring applications. This study aims to compare the accuracy bias of polysomnography (PSG) and consumer sleep-monitoring devices, specifically the HWB 9, in measuring total sleep time (TST) and explore the factors affecting accuracy bias.
Methods: This study employed a sequential explanatory mixed-methods design, with quantitative research comprising 108 samples and qualitative research comprising 18 samples. Select hospitalized patients who required polysomnographic monitoring due to their condition from November 2024 to March 2025 were chosen as the research subjects, and who used PSG and HWB 9 for synchronous sleep monitoring throughout the night. Quantitative data analysis was conducted using descriptive statistics, the Wilcoxon test, Bland-Altman plots, univariate analysis, and multiple linear regression analysis. Qualitative content data were analyzed using NVivo 14.0 software.
Results: The statistical analysis showed a significant difference (P < 0.05) between the HWB 9 and PSG in measuring TST. The Bland-Altman plot showed that the measured values deviated from the consistency interval, indicating systematic overestimation bias. The multiple linear regression analysis showed that turning frequency and sleep posture were significant factors affecting measurement bias. Two main themes were found in the qualitative research: sleep habits and environmental factors, and individual differences and psychological perceptions.
Conclusion: Considering the significant variations in individuals, data from such devices should be used with caution in clinical practice.

Keywords: PSG, consumer sleep-monitoring devices, TST, accuracy bias, influencing factors, mixed-methods