I read a Chicago study conducted over a four year period. The single largest predictor of being shot was to have been arrested with another person. If you’d been arrested with another person your chances of being shot or killed were 70% as opposed to almost nil for those who had never been arrested. One would think that with stats like these crime would eventually die out.
This looks like it refers to the study you are talking about. http://www.usnews.com/news/articles/2013/11/14/gun-violence-significantly-increased-by-social-interactions
It indicates that only demagogues write about how the US has a “gun problem” when, in fact, what it has is a criminal subculture associated with third-world murder rates, nested within a much larger society with murder rates not in any way exceptional (notwithstanding strong firearms ownership rights).