0

I am performing an experiment using caffeine and testing the resting blood pressure before and after caffeine consumption as well as cognitive performance using the Stroop test. I've performed dependent t-tests between the pre- and post-test for the two variables to determine their significance.

Now I know that there has been a significant change in blood pressure and cognitive performance, I want to know if there is a correlation between the increase in blood pressure and the increase in cognitive performance (reaction time in ms) disregarding the use of caffeine?

I've done a scatter plot between the difference (post_caffeine_BP - Pre_caffeine_BP) vs (post_caffeine_Reactiontime - Pre_caffeine_Reaction time) but I'm not sure if this is correct.

T.E.G.
  • 1,676
  • 6
  • 16
  • 29
EMNM26
  • 1

1 Answers1

1

I recommend you read this post first. Your situation seems a little more complicated since your exposures also changed over time. I think linear mixed model will be a very good choice for your study design. Your outcome variable will be the cognitive performance, a continuous variable. You measured it twice at baseline and after caffeine consumption. Your predictor variables will include your baseline cognition performance, blood pressure levels, and caffeine levels all measured twice(probably with some missing)

Your data struct will look like the following:

 PatientID Cognition_score   SBP    caffeine_level  measured_at  base_cog
 1          50               120             0.2      baseline    50
 1          60               130             0.3      post        50
 2          60               100             0.0      baseline    60
 2          70               90              0.1      post        60
...
Deep North
  • 4,527
  • 2
  • 18
  • 38