The article made the mistake of analyzing only a tiny piece of the system and attempting to draw conclusions for the whole system. One of the mistakes the arti

The article made the mistake of analyzing only a tiny piece of the system and attempting to draw conclusions for the whole system. One of the mistakes the article makes is in the statement that, if "conservative" girls went out twice as often, the chances of an HIV transfer are decreased because the guy has twice as many potential partners who are HIV-negative. What's missing is the recognition that, if the guy is already HIV positive, he now has twice as many potential partners who he can pass the infection on to. The changed network could either slow down or speed up the spread of HIV, depending on various specific details (how many men have HIV, how many women have HIV, how many connections to non-carriers do the men have vs the women, and so on.) As you correctly note, even if this particular change slows down the rate of infection in the present, it may ultimately increase the total number of infections. It may even slow the infection rate for a few years and then suddenly speed it up as some particular critical threshold is reached.