各位高手，我近期要设计一个新药的群体药代动力学实验，在设计取点时遇到一些困惑，我想对一部分受试者取大概6个时间点，再找一些受试者，取零散的1-2个点，但是不知道所有点的取点是取在第一次给药后，还是取在达稳态后，若有的病人采第一次，有的取稳态，可以混在一起计算吗？若取6个点，至少要保证多少病人量？
Hi Liangjia, NONMEM could calculate mixture data from both single dose and steady state.
Regarding the sampling points and sample size, I would propose to do a sensitive analysis (eg. simulation approach) to estimate the accuracy of pk parameters and the power. You might meet with some difficulty of accuraly select the priori of your model and parameters. Those will be born from your expertise (extrapolating from one population to the target population).
Hope it helps.
ro gg816
am a little famliar to NONMEM.The advice suggested by gg816 is avaible.However ,could you tell me more about the sensitive ananlysis.Do you mean the $simulation in the nm-tran control stream when you talking about the simulation approach?
至最最上面的楼主：采样点对于使用NONMEM的群体分析可以稀疏一点，但是还要是均匀覆盖到药物体内过程的各相。此外，峰浓度附近可以多点。另外，如果带到稳态不稳态对于数据处理没有影响，只要真实的记录下给药的剂量和时间间隔即可。当然主要感兴趣的协变量的性质也对采样点有影响。
谢谢楼上
如果有PK参数可以用ADAPTⅡ软件，采用FISHER信息矩阵D型优化设计
Hi Liangjia, NONMEM could calculate mixture data from both single dose and steady state.
Regarding the sampling points and sample size, I would propose to do a sensitive analysis (eg. simulation approach) to estimate the accuracy of pk parameters and the power. You might meet with some difficulty of accuraly select the priori of your model and parameters. Those will be born from your expertise (extrapolating from one population to the target population).
Hope it helps.
ro gg816
am a little famliar to NONMEM.The advice suggested by gg816 is avaible.However ,could you tell me more about the sensitive ananlysis.Do you mean the $simulation in the nm-tran control stream when you talking about the simulation approach?至最最上面的楼主：采样点对于使用NONMEM的群体分析可以稀疏一点，但是还要是均匀覆盖到药物体内过程的各相。此外，峰浓度附近可以多点。另外，如果带到稳态不稳态对于数据处理没有影响，只要真实的记录下给药的剂量和时间间隔即可。当然主要感兴趣的协变量的性质也对采样点有影响。
谢谢楼上
如果有PK参数可以用ADAPTⅡ软件，采用FISHER信息矩阵D型优化设计