Hello,
We are trying to analyze data from different patients. Here's the summary for the entire cohort:
1) Each patient will be followed over time.
2) Each patient will be given a starting dose of a compound, and a certain lab readout will be generated for that time point at that starting dose.
3) As time progresses, the dose will be increased/maintained, and a different readout will be generated by then. The dose adjustment will be done on a clinical basis.
Our question is:
We wanted to know whether increasing the dose would significantly change the readout at that certain time point. Two-way ANOVA (or a repeated two-way ANOVA) will erroneously provide us with a result. Which tool/statistical test would be appropriate for this scenario?
Thanks!
Since your goal is to understand the relationship between increasing the dose of a compound (measured in numerics) and the corresponding changes in the readout (assuming the readout is also numerical), you could opt for Pearson's correlation.
If you want to understand the level of impact of the increase in the dosage on the readout, you could also opt for a regression test.
All the best.
You have a confounding, as you said that the dose is increased "on a clinical basis". So whatever the clinical basis is will interfere and bias the relationship between the dose and read-out.
Hello Manuel Vidal Consider using sigmoid curves and nonlinear dose-response regression models.
You can do this well by using Prism software and then compare the curves with each other.
In this field, search in the software help. You will get good and easy results.
Statistical tool or test can be used for dose-response curve with time component is the nonlinear regression. You can use nonlinear regression to quantify the fold shift in the dose and its confidence interval, and to compute a P value that tests the null hypothesis that there was no shift.
That method is only applicable when the number of means is greater than 5, otherwise you must use multiple means comparison methods such as Dunca, Tukey, Newman-Keuls, etc. However, the statistical design issues related to dose-ranging studies are not always keenly understood and a poorly designed study can be costly for later development.