How do I compare 2 sets of measurements taken at different times?

[SiBorg]

Dear All

I'm not sure how to compare the following data.

I am doing a retrospective study to see whether treating surface eye disease can improve a patient's intraocular pressure control when they are using anti-glaucoma drops.

I only have 3 patients so the results may not end up being statistically significant but I would like to know how to do the correct test so that I can at least plan a bigger study in the future.

I want to look back at the notes and take 3 pressure measurements before the surface disease was treated and 3 measurements afterwards.

The results might look like this (I don't have the actual data to hand):

Patient 1

6 months before treatment: 32
4 months before treatment: 30
1 month before treatment: 34

1 month after treatment: 23
3 months after treatment:22
6 months after treatment: 24

And so on for patients 2 and 3.

Now, the first thing to point out is that the timings of the 3 measurements may vary between patients.

What I would like to know is:

Can I just average the 3 before measurements and the 3 after measurements and just do a paired t-test, or do I need something more sophisticated/non-parametric?

Thanks for your help!

With best wishes

Simon