About a year after starting treatment, I noticed that during the final week of that eight-week cycle and continuing for a few days after treatment, my time on Twitter was much higher than during the other 6.5-ish weeks. I figured it was something psychological, and was able to get my doctor to increase the frequency of doses, which curbed my Twitter visits even more.
Recently, I had a bout where due to insurance nonsense, I was delayed in getting my injection. I was heavily on Twitter, not as a result of conscious choice, but just because I felt like it. I took my shot a couple weeks ago, and have barely logged into Twitter over the last week and a half.
Correlation does not imply causation, but using my own anecdata, I can't rule out this study and in fact have some personal evidence to confirm it. I find it interesting that someone attempted to test a hypothesis that I had informally stated a few years ago.