Greek God of Sugar: Diabetes

I really get bugged when I hear/see conclusions that are taken out of context, or that are useless to begin with.

For instance, the report that came out today or yesterday that links soft drinks to diabetes. Now, the easy headline is/will be "SODA KILLS" or some such thing. What has been failed to be noted, however, is the rather crazy basis on which these results are based:

The study shows that women who drank one or more sugary drinks a day had an 83 percent greater chance of developing type 2 diabetes than women who drank less than one a month. The study shows that over a four-year period, weight gain was highest among women who increased their soda consumption from one or fewer drinks a week to one or more a day.

So, here is the comparison: those who drink one or more sugary drinks a day, compared to those who drink less than one a month. Is this a fair comparison? Doesn't that seem like a foregone-conclusion-waste-of-time comparison? Doesn't that sound like parameters that would pretty much guarantee a result like the one found?

It's like saying: Those who travel to New York City 36 times a year have a greater likelihood of being murdered in New York City than those who only travel to New York City once a year. Also, being victim to New York City muggings was highest amongst those who dramatically increased the number of New York City trips per year that they took.

Yet all we get is the pronouncement: Travelling to New York City linked to being murdered.

Conclusions are so easy.