The Challenge of Tracking What People Eat
By Rita Rubin
Reviewed by Michael W. Smith, MD
Dec. 12, 2013 (Washington, DC) -- Anyone who's ever belonged to a weight loss program knows the value of keeping a food diary -- a written record of every morsel or sip that enters the mouth.
The problem is, what people say they eat isn't often the whole truth. Sometimes they knowingly omit an embarrassing pig-out, like an entire sleeve of cookies or a whole carton of Rocky Road. Other times they simply underestimate a serving size.
Even inaccurate food diaries can help people lose weight -- although perhaps not as quickly as a careful record-keeper might -- but such journals can drive nutrition researchers crazy. The need for more accurate accounting of the what, when, and how much people eat has spurred the development of a variety of new gadgets. They include a microphone you wear around your neck that counts swallows, and the "HAPIfork," which alerts dieters when they shovel food into their mouths too quickly.
"If you want to study nutrition, you have to know what people are eating," Dale Schoeller, PhD, a professor of nutritional sciences at the University of Wisconsin, told an audience at the recent American Society for Nutrition meeting in Washington, DC.
In a perfect world, Schoeller says, scientists would be able to hook research subjects up to a GPS device to track their movement, monitor what they eat and how much, listen to their digestive tract, and upload all the information to their computer. While that might work for dairy cattle -- Schoeller showed a slide of "well-connected" Holly the Holstein, whose owner was aiming to produce the most milk with the least food -- humans would likely rebel over being tracked that closely in their daily lives.
But scientists have been aware for decades that food diaries are fraught with problems, given that people don't lose as much weight as their records indicate they should, he says.
The Good, the Bad and the Fattening
"People don't always tell you the bad, or what they perceive as bad," like that half-gallon of ice cream they ate after a tough day at work, Schoeller says. Plus, "the average patient figures a serving is what I put on my plate." But plates have gotten bigger over the years, he says, from 12 inches to 14 inches wide, resulting in about a 40% larger surface area.
In the United States, the higher their body mass index (BMI), the more people underestimate how much they ate, he says. And for the most part, the longer people keep food diaries, he says, the more likely they are to trim the fat, so to speak, from their entries.
So what's a scientist or a dieter supposed to do?
Schoeller has tried asking people to provide before-and-after photos of their plates, bowls, and glasses. After all, food photos are nearly as popular as selfies in social media these days.
"But you have to analyze it, which is a lot of work," he says. Then you run into the same problem you do with food diaries: People might not want to photograph that second or third helping of mashed potatoes or wine.
A microphone attached to your neck can provide information on how frequently you chew and swallow. "It tells you a little bit about what's being eaten," Schoeller says. "You can tell if it's a crunchy food or a soft food. Of course, the participant does have to be willing to wear it."
Then there's the HAPIfork, which will set dieters back $100, and the related wrist actigraph devices. The actigraph devices are commonly used in sleep studies to study sleep based on movement. But they can also be set to track wrist movements from plate to mouth.
While these less traditional devices provide objective information about the timing and frequency of meals and snacks, "they are not good at identifying what the food is at all," Schoeller says.
Toward that goal, scientists are working to identify biomarkers, or footprints, of specific foods, in blood, saliva, and urine. Some studies are now starting to use biomarkers in addition to participants' eating records.
These findings were presented at a medical conference. They should be considered preliminary as they have not yet undergone the "peer review" process, in which outside experts scrutinize the data prior to publication in a medical journal.
SOURCE: Dale Schoeller, PhD, a professor of nutritional sciences at the University of Wisconsin. American Society for Nutrition meeting, Dec. 5-7, 2013, Washington, D.C.