Development and Evaluation of Spatial Survival Models

空间生存模型的开发和评估

• 批准号: 7876868
• 负责人: Jiajia Zhang
• 金额: $ 7.39万
• 依托单位: UNIVERSITY OF SOUTH CAROLINA AT COLUMBIA
• 依托单位国家: 美国
• 财政年份: 2009
• 资助国家: 美国
• 起止时间: 2009-09-01 至 2012-08-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/7876868
• 关键词: Bayesian Method; Cancer Etiology; Complex; Data; Data Analyses; Development; Diagnosis; Epidemiologic Studies; Epidemiology; Equilibrium; Evaluation; Failure; Future; Grouping; Judgment; Least-Squares Analysis; Likelihood Functions; Louisiana; Malignant Neoplasms; Malignant neoplasm of prostate; Methodology; Methods; Modeling; National Cancer Institute; Pattern; Performance; Proportional Hazards Models; Research Personnel; Risk; SEER Program; South Carolina; Statistical Methods; Survival Analysis; Survival Rate; Time; United States; Variant; base; interest; male; men; models and simulation; mortality; neoplasm registry; simulation

DESCRIPTION (provided by applicant): Prostate cancer is the most common malignancy in men and a leading cause of cancer mortality among males in the United States. Large geographical variation and racial disparities exist in its survival rate after diagnosis. We will develop new and 0exible statistical methods of spatial survival model to estimate cancer survival. We will apply the proposed method to analyze the prostate cancer data within Louisiana from the Surveillance, Epidemiology, and End Results program and within South Carolina from the South Carolina Central Cancer Registry. We will use this analysis to investigate the spatial patterns and racial disparities of prostate cancer in Louisiana and South Carolina. We will conduct a complete simulation study to compare the performances of existing spatial survival models, and it can help the practitioners or researchers involved in cancer studies to select the right spatial survival model. The proposed method may possibly be extended to include more complex situations in the future, such as groupings and time variations, in epidemiological cancer study.

描述（由申请人提供）：前列腺癌是男性最常见的恶性肿瘤，也是美国男性癌症死亡的主要原因。诊断后生存率存在较大的地理差异和种族差异。我们将发展新的、灵活的空间生存模型的统计方法来估计癌症的生存率。我们将应用所提出的方法来分析路易斯安那州的前列腺癌数据从监测，流行病学和最终结果计划和南卡罗来纳州内的南卡罗来纳州中央癌症登记。我们将使用此分析来调查路易斯安那州和南卡罗来纳州前列腺癌的空间模式和种族差异。我们将进行一个完整的模拟研究，以比较现有的空间生存模型的性能，它可以帮助参与癌症研究的从业者或研究人员选择正确的空间生存模型。该方法在未来的癌症流行病学研究中有可能扩展到更复杂的情况，如分组和时间变化。

项目成果

Bayesian Parametric Accelerated Failure Time Spatial Model and its Application to Prostate Cancer.

• DOI: 10.1080/02664760903521476
• 发表时间: 2011-03
• 期刊: Journal of applied statistics
• 影响因子: 1.5
• 作者: Zhang J;Lawson AB
• 通讯作者: Lawson AB

Semiparametric Accelerated Failure Time Partial Linear Model and Its Application to Breast Cancer.

半参数加速失效时间偏线性模型及其在乳腺癌中的应用

• DOI: 10.1016/j.csda.2010.10.012
• 发表时间: 2011-03-01
• 期刊: COMPUTATIONAL STATISTICS & DATA ANALYSIS
• 影响因子: 1.8
• 作者: Zou, Yubo;Zhang, Jiajia;Qin, Guoyou
• 通讯作者: Qin, Guoyou