• 1. School of Traditional Chinese Medicine and Food Engineering, Shanxi University of Traditional Chinese Medicine, Taiyuan 030619, P. R. China;
  • 2. Department of Biostatistics, School of Public Health, Shanxi Medical University, Taiyuan 030000, P. R. China;
  • 3. Shanxi Provincial Key Laboratory of Major Diseases Risk Assessment, Taiyuan 030000, P. R. China;
  • 4. School of Health Services and Management, Shanxi University of Chinese Medicine, Taiyuan 030619, P. R. China;
ZHANG Yanbo, Email: sxmuzyb@126.com
Export PDF Favorites Scan Get Citation

Objective The re-hospitalization and death events of patients heart failure caused by coronary heart disease are characterized by non-independence, heterogeneity, and censored data. A joint frailty model is established to jointly model the events, explore the risk factors affecting the prognosis of patients, and reduce the re-hospitalization rate and mortality of patients. Methods The sample includes 4 682 patients with heart failure caused by coronary heart disease in two tertiary hospitals from January 2014 and June 2019. The electronic medical record information of patients during hospitalization and their follow-up information were collected. The Cox model, conditional frailty model and joint frailty model were used to analyze patient re-hospitalization and death. Results The joint frailty model identifies patients with a higher risk of both relapse and death (θ=0.209, P<0.001). Risk factors for re-hospitalization were advanced age, Grade 3 hypertension, mental work, no medical insurance, high cystatin C, and low ejection fraction, plus, within certain limits, low free thyroxine-3 and thyroxine-4. Antiplatelet drugs and statins significantly reduced the risk of re-hospitalization. Risk factors for death were advanced age, New York Heart Association classification Ⅲ to Ⅳ, no medical insurance, mental work, high cystatin C level, high troponin-I level, low free thyroxine-3, and low ejection fraction. Percutaneous coronary intervention, and taking antiplatelet drugs and statins significantly reduced the risk of death. Conclusion The joint frailty model can simultaneously model recurring and terminal events, and accurately predict them. Our results suggest that thyroid hormone levels and cystatin C levels of patients should be considered more carefully. People with mental jobs should change bad working habits to reduce adverse outcomes.

Copyright © the editorial department of Chinese Journal of Evidence-Based Medicine of West China Medical Publisher. All rights reserved