Earlier thread (50 comments), https://news.ycombinator.com/item?id=44527484
Correlation.
I take gabapentin for pain, so this study absolutely caught my eye. And I'm going to ask my doc about it. But the study is clear that this is a correlation, not a causation.
True, but tendency to dementia causes a fondness for a particular drug? Some 3rd effect causes both?
Those questions are exactly why I'm going to ask my doc. But because gabapentin is specifically for neuropathic pain, I could certainly envision a scenario where a deeper neurological root cause is giving someone both pain and dementia, so they are both symptoms.
I'm not banking on that, of course. It is worth digging into to find out more.
Definitely, best of luck ...
Another plausible interpretation is that the profile of a patient who is commonly prescribed the drug substantially overlaps with the profile of a patient who one day develops dementia.