N = 44 and the authors are concluding two separate trends? (Positive correlation in patients with GAD and negative in patients without). Since each correlation

N = 44 and the authors are concluding two separate trends? (Positive correlation in patients with GAD and negative in patients without). Since each correlation only applies to the subset of people this is at best an N = 26 study. Seems like an interesting topic, but with a sample size so small I struggle to believe (or disbelieve) any of the conclusions.