I have some info culled from international registries of non-US rehab medicine clinical trials
one of the questions of interest is whether a trial included subjects over a certain age (e.g., 65 or 85); note that I do NOT have info on the inclusion/exclusion criteria
for most trials I have the total N, but not for all trials
for some trials I have the mean and SD (though for some I only have the mean) or I have info that can be used to estimate these values (I am using the formulae suggested in Wan, X, et al. (2014), "Estimating the sample mean and standard deviation from the sample size, median, range and/or interquartile range", BMC Medical Research Methodology, 14: 135)
If I have the N, the mean and the SD AND if I can assume that the ages are approximately normally distributed, then answering the over65 or over85 question is easy. However, I am uncomfortable making this assumption and I am looking for citations that provide guidance for either (1) different distributions (esp skewed ones) and/or (2) truncated distributions but where the truncation point is unknown. If I could find such cites, I could do a sensitivity analysis re: the assumed normal distribution answer
So, does any one know of any such citations or have other suggestions?
By the way, I know that I could do a bunch of simulations to get there but I think this would be more expensive then my clients want to go.
one of the questions of interest is whether a trial included subjects over a certain age (e.g., 65 or 85); note that I do NOT have info on the inclusion/exclusion criteria
for most trials I have the total N, but not for all trials
for some trials I have the mean and SD (though for some I only have the mean) or I have info that can be used to estimate these values (I am using the formulae suggested in Wan, X, et al. (2014), "Estimating the sample mean and standard deviation from the sample size, median, range and/or interquartile range", BMC Medical Research Methodology, 14: 135)
If I have the N, the mean and the SD AND if I can assume that the ages are approximately normally distributed, then answering the over65 or over85 question is easy. However, I am uncomfortable making this assumption and I am looking for citations that provide guidance for either (1) different distributions (esp skewed ones) and/or (2) truncated distributions but where the truncation point is unknown. If I could find such cites, I could do a sensitivity analysis re: the assumed normal distribution answer
So, does any one know of any such citations or have other suggestions?
By the way, I know that I could do a bunch of simulations to get there but I think this would be more expensive then my clients want to go.
Comment