Freakonomics has a podcast on the evils of sugar.
KAHN: The next one is something called temporality. In other words, is there association in time between sugar consumption and obesity? That held pretty true from about 1985 to the year 2000. Where obesity levels went up, sugar consumption went up. But thereafter starting in the year 2000, even to today, sugar consumption has declined somewhere around 15 to 20 percent, whereas obesity rates continue to rise. We don’t see that at all with smoking, the analogous situation. Cigarette smoking rises, cancer rises. Cigarette smoking declines, cancer declines.