I am measuring the drug release of a few drug-eluting implants in dissolution medium in different time intervals. All the measurements are in triplicates. Would ANOVA or TTEST is an appropriate statistical analysis for determining p-values for these kinds of studies? To more clarify it, the number of measurements, in total, would be as follows: 3 (triplictes) * 4 (time intervals) * 5 (implants).
Ameer Muhammad is guessing.
You need to study the literature on pharmacokinetics. There are several ways in which you can assess the behaviour of drugs over time, such as the area under the curve (giving you a measure of total exposure) or the time to reach peak, or the peak value. Which measure is most important depends on the nature and purpose of the drug and its desired characteristics.
Helpfully, Stata has extensive documentation for their suite of pharmacokinetic commands, and you can follow the worked examples even if you don't have Stata.
https://www.stata.com/manuals13/rpk.pdf
See page 1380 of attached document. Lots of references and talks of what the FDA approves of, etc.
Ronán Michael Conroy thank you for adding this aspect of the data. Then he has to used pksummobtains summary measures based on the first four moments from the empirical distribution of each pharmacokinetic measurement and tests the null hypothesis that the distribution of that measurement is normally distributed.
Ameer Muhammad The help page for pksumm states very clearly that you need to read the introductory help section for the pk commands first. A test for normality is not necessarily appropriate for many situations, in fact.
My point was that pharmacokinetics is an established field – and Luisiana Cundin has kindly supplied some important information about FDA practice – and you need to know what you are doing.
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis