The numbers in this article are very frustrating, and the graph is misleading. The rise in absolute popularity of drugs with names with z and x has a lot less meaning if it's not presented along with the rise in total drugs too. That number (80% rise) is also mentioned, but you have to do mental arithmetic to figure out how that affects the other numbers, and it doesn't help you normalize the graph.<p>In fact, judging by the graph, if the total number of drugs increased by 80%, it looks like the relative popularity of "x" is actually static / declining over time.<p>All numbers for increases in popularity of x and z should have been normalized to the increase in drug count; in other words, described as an increase in the share of the total set of drug names (or share of new market entrants, or whatever).