摘要
目的:给出基于混合样本的最小检测次数法和序贯检验法,以检验子孢子阳性率是否超过导致疟疾流行的临界水平。方法:在控制两类错误概率的前提下,推导两种方法的期望检测次数和检验功效的数学表达式。对最大期望检测次数进行数学优化,求解最优混合样本的大小,并通过计算机模拟对有关结果进行验证。结果:给出了最小检测次数法和序贯检验法的最优混合样本计算方法、MATLAB程序和具体操作步骤。前者可用于日常监测,后者可用于早期预警。结论:利用本文给出的最优混合样本和检测步骤,在获得比较满意的检验功效(两类错误概率均低干5%)的同时,可以有效降低检测次数。
Objective：To provide a minimum sample size approach and a sequential sampling approach for testing whether the sporozoite rate has exceeded the critical level of malaria epidemics using the pool sampling method. Methods： Formulas of the expected pooled sample size and the power of tests were deduced while controlling the probability of type Ⅰ and type Ⅱ errors. The optimal pool sizes of the 2 approaches were given by minimizing the expected pooled sample size; computer simulation was used to verify the outcomes. Results： The optimal pool size, programming of MATLAB,and the steps of trials of the 2 approaches were given. The minimum sample size approach could be used for routine surveillance and sequential sampling approach could be used for early warning. Conclusion： The optimal pool size in the present study can obtain satisfactory testing power （type Ⅰ and type Ⅱerrors are both lower than 5%） and can effectively decrease the pooled sample size.
出处
《第二军医大学学报》
CAS
CSCD
北大核心
2007年第5期465-469,共5页
Academic Journal of Second Military Medical University
基金
"十五"国家科技攻关课题(2004BA718813).~~
