Education is certainly a solution for a lot of things, but I've long thought the obsession with education as a solution for tobacco and alcohol use and obesity is highly overrated. The first evidence was a set of studies that demonstrated that the only measure that prevented youth smoking was higher tobacco taxes, and that controlling for tobacco taxes, education at any age and in any form didn't matter (I don't believe the studies tested the use of things like "Joe Camel", which I could certainly see influencing long-term tobacco consumption, though perhaps not short term). This is very plausible, because most people know that smoking is bad for you, even if they don't know exactly how bad, so insisting that knowledge of the consequences is going to sway the decisionmaking probably ignores everything we know about teenage decisionmaking.
This study, which indicates that informed people make poor choices when there's an immediate gain for a larger long-term cost, lends credence to the idea that people discount hyperbolically, so education alone isn't sufficient to reduce bad decisionmaking - there needs to be some sort of commitment device or additional short term cost. I don't love the study - points seem like a poor proxy for emotional temptation - but the fact that they got results at all is somewhat telling, depending on how the actual questions themselves were designed.
(Note that the title of the article is highly misleading - I can be informed about what kind of car to buy, or what I should look for in health insurance, and not make poor decisions in the way the study says - it's a strictly hyperbolic discounting phenomenon)