队列半数生存期分析:非参数方法

在估价不同疗法对恶性疾病或其他慢性疾病疗效时,生存率比较有时会导致错误结论。因而由一个流行病学队列分析资料建立了非参数生存函数,绘制出生存折线及其95%的Alderson置信带。求出队列半数生存期(hlc)及其95%的置信限。以hlc作为评价统计学指标可以获得确切的结论,避免了生存率比较的缺点。在卫生学中,hlc拟用于估计寿命。在流行病学队列分析中,拟用于描述恶性疾病对寿命的影响,评价疗效和预防效果等。

The half-life of a cohort:a non-parametric approach

Companison of survival rates could be misleading when it was used to assess the efficacy of different treatments for malignant or other chronic diseases. Thus non-parametric survival functions were established from the data of an epidemiologie cohort analysis. A survival curve and its 95 per cent confidence band (Alderson) were depicted. The half-life of the cohort (hlc) and its 95 per cent confidence limit were calculated from the survivorship function. The hlc was used in analyzing the data as an evaluative statistical index and a clear result was obtained. Hence some defects in the comparison of survival rates was avoided. The hlc can be used to estimate lifespan in hygiene, to describe the influence of malignancies upon life span, and to evaluate the efficacy of treatment, preventive measures, etc. in epidemiologic cohort analysis.