Shortly after the op-ed was published, a television producer wanted to film us doing a before-and-after Smarter Lunchroom Makeover of a middle school. Why a middle school? Apparently elementary students act too random in front of TV cameras (remember that picnic for squirrels?), and most high schoolers aren’t photogenic enough for television--too many strange clothes, weird hair colors, piercings, and uninterested looks. The TV people wanted us to find a middle school that would do a total lunchroom makeover for less than $50--and film it all MTV-style.[ii]
            After finding the perfect middle school and watching students eat lunches for a week, we isolated ten changes we could easily make for less than $50 total that would probably help them eat better without even realizing it--things like changing the location of the fruit, giving fun names to healthy foods, moving the cookies behind the counter, putting the vegetables first, and so on. The food service director and producer were cool with the changes, so we got to work.[iii]Twenty-five kids were hand-picked to be secretly filmed by three hidden cameras. We hid cameras in a ceiling tile, a hat, and even in our fake water bottle. Everything was set--and then came the catch. We were asked, with the cameras rolling, to predict the sales for each food item. ​

        After lunch was over, the smoke cleared, and the dishes washed, we were able to calculate just what had happened. The makeover was a nutritional victory--kids took a lot more salads, fruit sales doubled, white milk sales went up 38 percent, sugary drinks sales dropped by 17 percent, and they ran out of the healthy bean burritos--renamed Big Bad Bean Burritos--for the first time ever. These kids ate an average of 18 percent fewer calories, and they ate better than they typically did.[iv]
            What didn’t work was putting the cookies behind the counter. We thought this would decrease sales by 30 percent, but it did nothing. Even worse, we predicted that moving vegetables to the front of the line would increase sales by 11 percent, but it instead dropped by 30 percent.[v]What happened?
            A little bit of sleuthing showed that cookies were the cafeteria’s big “destination food.” They were five inches of hot, freshly baked gooey goodness--the main reason some kids ate school lunch. Wild horses couldn’t have pulled these kids away from the cookies without pulling them away from eating lunch there altogether.
            The vegetables were a different story. As I mentioned, our lab studies showed that lunchgoers were 11 percent more likely to take whatever vegetable they saw first compared to whatever they saw third. Well, that’s true when three vegetables are in the middleof the serving line, but here we put them in the frontof the line. Nobody scoops up a plate of green beans and then looks for the entrée that goes with it. People pick out the entrée and thenthe vegetable. They didn’t want to take a veggie until they knew what they were having for a main course. 
            When the interview got to this point, the producer asked, “You’ve been doing eating research for twenty-five years. Sales didn’t increase by 11 percent, they dropped by 30. Why were you so far off?” I said, “Well, if we always knew what we were doing, we wouldn’t call it research.” (He seemed amused enough by this answer to not report these missed predictions in his story.)
            Still, nailing five out of seven predictions was pretty decent. Our prediction report card wasn’t straight As, but it was better than the report cards I got in high school. Most important, we were able to show in real-TV-time how only $38 and two hours of tweaking made a bigger difference than hefty expert commission reports. ​

Halloween is like Thanksgiving for candy bars.

We were talking about Halloween Trick or Treating last night, and people were talking about what they were going to do with all of the left over candy that they bought, or which their kids bring home.  Nobody planned on throwing it out.  They would all eventually eat it.  Probably pretty quickly. 

This reminds me of a study we did for the New England Journal of Medicine that showed that every year Americans start gaining more weight from today and for the next two months.  The key take-away is that we shouldn't wait until January 1st to make a resolution to lose weight.  We should make a Halloween resolution to not gain weight. 

Below are some nice details related about the study.  The Blue line is the Americans, and you can see that just before November starts, the average weight of Americans grows higher and higher until just after January, when it drops again.

Labor Day, Halloween, Thanksgiving . . . the summer is almost over but the indulgent holiday season is near. This study we conducted found that many of us spend months getting rid of that excess weight gained during the holiday season. The study showed that according to yearly national weight patterns, it takes American’s nearly  between Thanksgiving and Easter. 

We also found  in a similar study that the weight of Americans begin rising around Thanksgiving, and peak around Christmas and the New Year. It isn’t until after Easter, about a 5 month period, that weight patterns even out.  

We also analyzed yearly weight patterns in Germany and Japan. Similar to the US, those in Germany weighted the most around Christmas/New-Year period and those in Japan weight the most during Golden Week in April – a major Japanese holiday. Each country also showed a peak in weight for New Years. 

Everyone gains weight over the holiday -- starting on Halloween  Instead of making a New Year’s Resolution, the best time to make a resolution to keep the pounds off this holiday season might be now.