Maybe I'm mis-reading, but it looks like all of their subjects were already predisposed to having issues (All patients, drawn from the Coronary Artery Risk Development in Young Adults (CARDIA) study). Not only that, but 3000 people, while plenty for a statistical analysis, doesn't really seem like a large enough population sample.
Say you've got 10% that actually lifted weights similar to what we talk about here. Of those 300, how many did it consistently for 2 decades? And of those, how many of those actual give you a wide genetic representation?
It's very possible exercising had fuck all to do with their issues.
I guess in the end, I'm saying, you shouldn't base any life decisions on a single study. Ever.