节点文献
现时生存分析方法的应用实践与评价
The Application and Evaluation of Period Survival Analysis
【摘要】 目的利用林州市肿瘤登记处的资料,应用现时生存分析方法计算食管癌、胃癌的5年相对生存率,并与传统的队列方法比较,评价现时生存分析方法在预测癌症患者长期生存率时的应用价值。方法分别利用队列、完全生存分析和现时生存分析方法计算食管癌、胃癌的5年相对生存率,与随访5年后得到的实际生存率进行比较,以三种方法计算的数据与实际值的绝对差值平方评估差异的大小。结果应用队列、完全生存分析和现时生存分析方法计算林州市食管癌的5年相对生存率分别为28.00%、32.08%、36.67%,胃癌23.19%、29.03%、33.98%,食管癌和胃癌的实际5年相对生存率分别为36.82%和34.04%。三种方法计算的结果与实际生存率差异的平方分别是196、48、0。结论现时生存分析方法得到的5年相对生存率与实际值最为接近,能及时地反映现时癌症诊治的水平与效果,可以作为评估癌症长期生存指标的一种统计方法。
【Abstract】 Objective To calculate 5-year relative survival rate of esophageal cancer and stomach cancer using period survival analysis method based on the data from Linzhou Cancer Registry.Compared with traditional methods and the value of period analysis in predicting long-term survival rate of cancer patients was evaluated.Methods The relative survival rate of esophageal cancer and stomach cancer were calculated by cohort,complete and period method respectively and the results were compared with the actual survival rate achieved by following up for 5 years.The difference was evaluated by the square of absolute difference of the results calculated by the three methods and the actual value.Results The 5-year relative survival rate of esophageal cancer calculated by the three methods were 28.00%,32.08% and 36.67% respectively.And that of stomach cancer were 23.19%,29.03% and 33.98% respectively.The actual survival rate of esophageal cancer and stomach cancer was 36.82% and 34.04%.The square of the difference between the results of three methods and actual survival rate were 196,48 and 0 respectively.Conclusion The 5-year relative survival rate derived by period survival analysis is most close to the actual survival rate.The period survival analysis can reflect the present level of diagnosis and therapeutic efficacy for cancer timely and it could be used to evaluate the long-term survival rate of cancer as a statistical method.
【Key words】 Period survival analysis; Relative survival rate; Esophageal cancer; Stomach cancer;
- 【文献出处】 中国卫生统计 ,Chinese Journal of Health Statistics , 编辑部邮箱 ,2011年01期
- 【分类号】R73-31
- 【被引频次】13
- 【下载频次】616