A real, robust knowledge of the relative danger of steroids is tough to develop.

Because you can’t plausibly do a long-term blinded trial where you take a healthy population and dope them up. The in vitro data is enough to make any institutional review board blanch.

So you have to rely on self-reported usage. Which is incredibly unreliable, and full of confounds. (My bias, my fairly well-informed bias, is that nobody is just doing steroids. There is always another drug or high-risk activity.)

So yeah. Steroids are dangerous. Do we know exactly how bad, especially in comparison to other drugs of abuse? Not as far as I can tell. But bad enough that you should absolutely be wary of steroids. (Also there’s gradations of dangerous. You stay away from trenbolone.)

If I based my risk of death on people I personally knew, I'd believe that I was immune to dying from, say, being hit by a car, and would almost certainly die of either heart disease or suicide. That's a ridiculous way to do things.

I don't know that many people in the steroid-abusing demographics (it was less of a thing in my age demographic than it is with the youth of today), but I am aware that it's dangerous.

Dangerous isn't always fatal: https://www.drugabuse.gov/publications/drugfacts/anabolic-st...

I would look at the many early deaths of professional wrestlers as a good example of deaths due to steroid usage.

Steroid use is linked among other things with embolism, stroke, heart attack and sudden death.