Department of Neurological Sciences, Rush University Medical Center, 1725 W. Harrison St., Ste. 755, Chicago, IL 60612, USA.
Stat Med. 2013 Jul 10;32(15):2629-42. doi: 10.1002/sim.5717. Epub 2012 Dec 19.
For a heart transplant patient, the risk of graft rejection and risk of death are likely to be associated. Two fully specified Bayesian models for recurrent events with dependent termination are applied to investigate the potential relationships between these two types of risk as well as association with risk factors. We particularly focus on the choice of priors, selection of the appropriate prediction model, and prediction methods for these two types of risk for an individual patient. Our prediction tools can be easily implemented and helpful to physicians for setting heart transplant patients' biopsy schedule.
对于心脏移植患者，移植物排斥风险和死亡风险可能相关。应用两个完全指定的贝叶斯模型来研究这两种风险之间的潜在关系以及与风险因素的关联，这两个模型用于具有相关终止的复发性事件。我们特别关注先验的选择、适当预测模型的选择，以及针对个体患者这两种风险的预测方法。我们的预测工具易于实施，并有助于医生为心脏移植患者制定活检计划。
