You might want to keep the following story away from your kids: A new study finds that not eating candy may make them fat. So saith the scientists:
For the study, published in Food & Nutrition Research, researchers at Louisiana State University tracked the health of more than 11,000 youngsters between the ages of two and 18 from 1999 to 2004. They found that children who ate sweets were 22 percent less likely to be overweight or obese than kids who shunned sweets. Adolescents? Those who ate candy were 26 percent less likely to be overweight or obese than their non-candy-eating counterparts.
And that wasn’t the only surprising finding. Researchers also found that the blood of candy-eating kids had lower levels of C-reactive protein. That’s a marker of inflammation in the body and a risk factor for cardiovascular disease and other chronic illnesses.
There’s a “but . . .” after that part but I stopped reading because I didn’t want to ruin it for my inner six-year-old. I wish I had been armed with this study when I was growing up. It would have made the debates with my mother more interesting:
Mom: Put down the candy bar, we’re going to eat supper in an hour. You’ll ruin your appetite.
Me: Exactly. That’s the plan.
Mom: What? What are you talking about? What’s the plan?
Me: By eating this Snickers I will ruin my appetite. By ruining my appetite I will be less hungry during dinner and consume fewer calories. Ergo, by eating candy I will avoid getting fat.
Mom (slapping the candy bar out of my hand): You’re an idiot.
Okay, so that probably wouldn’t have worked out too well. Had my own kid tried it, though, I would have been totally convinced. She would be singing a song of praise, “Dad is great. Gives us the chocolate cake. . . ”
June 30th, 2011 | 3:57 pm
Post hoc ergo propter hoc. Kids who don’t eat candy are (in part) those whose parents realize that they have a weight problem. Kids who eat candy are (in part) those whose parents feel they can afford treats now and then, as they don’t have a weight problem.
Population studies like this are USELESS. Study the physiological effects of candy on the body through actual clinical research. That’s how you figure out how candy intersects with weight issues.
June 30th, 2011 | 4:46 pm
pentamom –
For setting policy, sure.
For setting a research agenda by identifying correlations in need of causal explanation, they can be quite useful.
June 30th, 2011 | 5:56 pm
Don’t we already know that it’s worth studying what effects sugar has on kids, though? What added value is there to this study –
except the value of the resulting published paper to the person in quest of either a PhD or his publishing quota for tenure? THAT’s why studies like this get done.
June 30th, 2011 | 10:00 pm
pentamom – Even common sense needs to be checked sometimes. There have been plenty of things that pretty much everyone believed that turned out to be wrong. That the brain was basically just a cooling mechanism for the heart, for example. Or that educating women would impair the development of their reproductive organs.
In this case, the conclusion isn’t just a total slam-dunk. This study offers good evidence that sugar, even in concentrated doses, isn’t an automatic ticket to obesity. That’s actually contrary to the expectations of a lot of people these days.
July 1st, 2011 | 8:22 am
Ray — I didn’t say we already knew what effect sugar has on kids.
I said we already knew it was *worth studying* what effect sugar has on kids.
Are are you seriously proposing that there was no scientifically established reason for studying the effect of sugar on kids until the study was published in the latest issue of Food and Nutrition Research????
July 1st, 2011 | 8:54 am
Pentamom: precisely. Distinguishing correlation and causation seems to be beyond the average Ph.D.
July 1st, 2011 | 11:22 am
pentamom –
Nope.
I was suggesting that this study points out an interesting correlation that cries out for an investigation of causation.
You, for example, have put forth a hypothesis – that (at least a significant portion of) parents who don’t think their children are obese allow their kids to eat candy, and parents who think their kids are already obese don’t allow them to eat candy. A properly-designed followup study could check this hypothesis.
There are other potential explanations – perhaps a natural appetite for concentrated doses of sugar is associated with less desire for fatty foods (like, say, french fries). Or perhaps it’s associated with a tendency to eat more vegetables. In any case, studying just the “effect of sugar on kids” wouldn’t tease out such a ‘tuning of the tastebuds’.
July 1st, 2011 | 12:19 pm
I don’t know that I need my hypothesis “checked” — I wasn’t actually proposing it at a hypothesis that needed folllowing upm I was merely giving an example of the bazillions of variables that could explain such a correlation.
But again, the idea that this study was needed to prompt some kind of investigation into the either the causes of childhood obesity, or the effects of sugar on the body, seems ludicrous. And if this study wasn’t needed to prompt such investigations, then I don’t see how my comment that such things are “useless” is effectively disputed by your pointing out that sometimes such population studies can be used to raise possibilities. Of course they can sometimes, but this is not an example of one of those times; if it’s utterly superfluous for the purpose, then “useless” seems as good a word as any.
July 1st, 2011 | 5:26 pm
pentamom – So, you agree there is an explanation for the correlation, you just have no curiosity whatsoever as to what the explanation might be, and assume that – whatever the explanation might turn out to be – it will have no practical applications?
Indeed, I went and looked at least at the abstract of the study in question. It argues against your hypothesis already: Diet quality was determined using the Healthy Eating Index-2005 (HEI-2005)… Mean HEI-2005 score was not different in total candy and sugar candy consumers as compared to non-consumers, but was significantly lower in chocolate candy consumers (46.7±0.8 vs 48.3±0.4, p=0.0337).
So, there didn’t seem to be a significant difference in diet otherwise between kids eating candy and those that didn’t. If it’s parental control that’s restricting candy consumption in obese kids, then candy is apparently the only area they’re doing that. Indeed, the abstract states that “Total candy consumers had higher intakes of total energy (2248.9 kcals±26.8 vs 1993.1 kcals±15.1, p<0.0001) and added sugars (27.7 g±0.44 vs 23.4 g±0.38, p<0.0001) than non-consumers." – i.e. the kids eating candy actually did consume more calories than the kids who weren’t. Yet the kids eating candy were noticeably less likely to be obese.
Right now we have a datapoint that indicates that the relationship between candy consumption and obesity is regressive. Quite possibly consumption of candy really does have a previously-unsuspected beneficial effect. Consider the positive effects that have been found arising from (moderate) consumption of alcohol – it’s hardly impossible.
Maybe it’s not affecting metabolism so much as it’s the way candy is served – in bite-size pieces – which leads to kids ‘feeling full’ faster than if they eat a bowl of ice cream. If so, that would offer a way to retrain obese kids with a sweet tooth to satisfy their hankering in a more desirable way.
As Isaac Asimov put it, “The most exciting phrase to hear in science, the one that heralds new discoveries, is not ‘Eureka!’ (I’ve found it!), but ‘That’s funny…’”
The point is, this study actually does point out something contrary to what most people would expect. I guess you can assume that has no practical relevance… but it didn’t take me much effort to come up with what I think are interesting followup questions that should be addressed by further study.
(Recall the study Joe linked to before (http://www.firstthings.com/blogs/firstthoughts/2011/05/06/how-contraceptives-are-like-polygamy/) where even Dr. Barr chided him for apparently expecting a study to answer all questions at once?)
July 2nd, 2011 | 11:27 am
I may be new here, but couldnt this article have been written with the child eating anything before dinner? He/She could have eaten a bowl of grapes and been full enough to not eat all of his dinner. Just a “common sense” observation. :) Think this might be over-analyzing a bit.
July 2nd, 2011 | 6:19 pm
Ray, the problem is that if you get any detailed set of data, you can find correlations for different things. You can replace “eating candy” with “reading comic books” or “keeping a banana in your pocket.” Because humans are complex, you still will find differences and correlations, but they wont be linked to a single behavior.
Heck, you can even just tell someone something via the placebo effect and find variances. I think when a system gets complex enough, interactions can’t be reduced to single effects.
July 5th, 2011 | 8:38 am
amomswish, according to the study, they were looking at kids from the “1999-2004 National Health and Nutrition Examination Survey”. So, unless it was a really long wait for dinner, it doesn’t appear so. :)
Dblade –
Sometimes those correlations are statistical flukes. And sometimes they’re real. That’s why I suggested, y’know, followup research.
Are you prepared to say that this correlation is illusory? How much are you willing to bet?
At what point did I say that there was only one single factor in obesity, or anything?
Links
Blogs
Find Us
Contact