Not pointed out in the article, but perhaps in the study(?) - the United States has been engaged in an active war for the past 15 years that likely contributes to an increased number of those taking painkillers for injuries or having some sort of psychological disorder. In my mind I liken it to shellshock and after-effects felt in societies post WWI / WWII. I don't have specifics on the number of wounded / in treatme…
1. The number of veterans is very small
2. Veterans are overwhelmingly from the richer half of the population
About 0.5% of the cohort currently between the ages of 20 and 40 had any military service (compare to 12% of the WWII cohort and 6% of the Vietnam cohort), and only about half of that population saw service in Iraq or Afghanistan (e.g. I was active duty for almost all of the Iraq war and never went to theater, nor did almost anybody in my battalion).
Furthermore (and this has been troubling the Pentagon for a while), the enlistment/commissioning standards the military has make it very difficult for poor Americans to serve. The poorest 20% of Americans are the least likely to serve, followed by the next-poorest. In fact it's nearly a perfect correlation except that the richest 20% are slightly less likely to serve than the second quintile. This yields a recruit and officer candidate pool that is both richer and whiter than the nation as a whole. (On the other hand, richer whiter servicemembers have abysmal retention rates, so that the standing force is browner and poorer than the nation as a whole because most of the middle class white people leave after one tour).