Clinical Focus ›› 2026, Vol. 41 ›› Issue (2): 148-154.doi: 10.3969/j.issn.1004-583X.2026.02.008

Previous Articles     Next Articles

Construction of a sleep disorder risk prediction model for maintenance hemodialysis patients based on LASSO-logistic regression

Wang Jingjing1, Jiang Qianqian2(), Gao Ju3, Zhou Yeping1   

  1. 1. Department of Nephrology, Changzhou Cancer Hospital, Changzhou 213000, China
    2. Second Department of Clinical Medicine, Shanghai Changning Mental Health Center, Shanghai 200335, China
    3. Department of Psychiatry, Suzhou Guangji Hospital, Suzhou 215131, China
  • Received:2025-12-26 Online:2026-02-20 Published:2026-03-05
  • Contact: Jiang Qianqian, Email: 226334295@qq.com

Abstract:

Objective To investigate the risk factors for sleep disorders in patients undergoing maintenance hemodialysis (MHD), and to develop a risk prediction model, thus providing references for the early clinical identification of high-risk individuals. Methods A total of 222 patients who received MHD between June 2024 to June 2025 were enrolled. Demographic characteristics, clinical data, and laboratory indicators were collected. Sleep quality was assessed using the Pittsburgh Sleep Quality Index (PSQI), with PSQI ≥7 indicating the presence of sleep disorders. Candidate predictors were first screened using univariate analysis, followed by variable shrinkage and selection via the least absolute shrinkage and selection operator (LASSO) regression. Variables retained in the LASSO model were subsequently entered into a multivariable logistic regression analysis to construct a nomogram. Prediction performance of the nomogram, including discrimination, calibration, and clinical utility, was comprehensively evaluated using the receiver operating characteristic (ROC) curve, C-index, bootstrap internal validation, calibration curve, and decision curve analysis (DCA). Results The incidence of sleep disorders in this study was 56.31% (125/222). LASSO-logistic regression showed that age, dialysis duration, anxiety, depression, uremic pruritus, and restless legs syndrome were independent risk factors for sleep disorders in patients treated with MHD, whereas higher serum calcium levels played a protective role (all P<0.05). The nomogram constructed based on these factors yielded an area under the curve (AUC) of 0.928(95%CI: 0.894-0.962). The calibration curve showed good agreement between predicted and observed values, and the Hosmer-Lemeshow test (χ2=4.14, P=0.844) indicated good calibration performance. DCA demonstrated that the nomogram provided a considerable net clinical benefit across a threshold probability range of 0.05-0.75. Conclusion The nomogram constructed based on age, dialysis duration, anxiety, depression, uremic pruritus, restless legs syndrome, and serum calcium level can effectively predict the risk of sleep disorders in MHD patients, facilitating early identification and intervention for high-risk groups.

Key words: sleep disorders, maintenance hemodialysis, LASSO regression, logistic regression, nomogram, risk prediction model

CLC Number: