Lao C S
United States Food and Drug Administration, Center for Devices and Radiological Health (HFZ-542), Rockville, Maryland 20850, USA.
J Biopharm Stat. 1998 Nov;8(4):619-33. doi: 10.1080/10543409808835264.
The Cox proportional hazard model is used to identify important covariates related to survival for this beagle dog study. The Weibull and the Gompertz parametric probability models are used to fit the survival curves for 1680 beagle dogs given whole body Co-60 gamma radiation or sham irradiation and held for life-span observation. Deaths from cancer and noncancer are the primary diagnoses. Of dogs dying from cancer, female dogs showed significantly greater cancer mortality (p < 0.0001) than did males, and irradiated dogs had significantly greater cancer mortality (p = 0.022) than did the unexposed dogs, by either the Weibull or the Cox model. However, there were no sex and exposure differences in mortality for dogs dying from noncancer causes. The fitted Weibull and Gompertz models have been successful in projecting the actual mortality experience in this experiment and could be used for similar life-span experiences.
在这项比格犬研究中，Cox比例风险模型用于识别与生存相关的重要协变量。威布尔（Weibull）和冈珀茨（Gompertz）参数概率模型用于拟合1680只接受全身钴-60γ射线照射或假照射并进行寿命观察的比格犬的生存曲线。癌症和非癌症死亡是主要诊断结果。在死于癌症的犬只中，无论是根据威布尔模型还是Cox模型，雌性犬的癌症死亡率均显著高于雄性犬（p < 0.0001），且受照射犬的癌症死亡率显著高于未受照射犬（p = 0.022）。然而，死于非癌症原因的犬只在死亡率方面不存在性别和照射差异。拟合的威布尔模型和冈珀茨模型成功预测了本实验中的实际死亡率情况，可用于类似的寿命研究。
