This is something I've looked into quite a bit actually. The statistical clustering is what gets me - when you look at the raw numbers of people who die "by suicide" in the window between being named as a potential witness and their scheduled testimony date, the frequency is genuinely hard to explain away as coincidence.
Jeffrey Epstein is the obvious example everyone goes to but there are dozens of lesser-known cases that barely made the news. I've been compiling a list for a while now and the pattern holds across different countries, different decades, different power structures.
What I want to know is whether anyone has looked at this through a data lens rather than just case by case. Has anyone done actual statistical analysis comparing the suicide rates of people in this specific legal position vs the general population? Because thats the kind of evidence that would be harder to dismiss than individual cases which can always be explained away one at a time.