> "Yet even though many health statistics have been improving over the past few decades, a few illnesses are rising mysteriously. From the early 1980s through t

> "Yet even though many health statistics have been improving over the past few decades, a few illnesses are rising mysteriously. From the early 1980s through the late 1990s, autism increased tenfold; from the early 1970s through the mid-1990s, one type of leukemia was up 62 percent, male birth defects doubled, and childhood brain cancer was up 40 percent. That language is sloppy and imprecise. What's increased is the (I presume the rate of) diagnoses. Not cases. Phrasing it like that allows the reader to consider many plausible confounders instead of leading them on to the predetermined conclusion of the article, or at least appearing to do so. Has detection of that particular type of leukemia increased dramatically with new technology? Are there cultural reasons why diagnoses with few if any objective measures, like autism, are more readily handed out by doctors? There are plenty of controversies about how the contents of DSM, the book that spells out the gory details of psych diagnoses, are changing over time (i.e. DSM-4 vs. DSM-5). The right conclusion might very well be the one authors are trying to suggest, but "helping it along" so transparently is unhelpful.