If Your Phone Could Warn You When You're About To Overeat

Here's my dream of a dieting app: It picks up your brain waves, and when you start to crave chocolate or chips, it instantly hyper-activates your neural receptors for feelings of fullness, so you have that "couldn't eat another bite" feeling until the cravings pass.

Sigh. Someday. Right now, what we mainly have is app upon app that lets you record what you eat and add up those calories, then record your exercise and subtract those calories. This is not a bad thing. Keeping a food diary is a proven tool for successful dieting. But, as obesity researcher Sherry Pagoto writes on Psychology Today, these apps are somewhat underwhelming, not the game-changers that mobile tech seems to have the promise to deliver.

So Pagoto, a clinical psychologist and an associate professor at the University of Massachusetts Medical School, is aiming to develop a more sophisticated app, one that incorporates "sensing" technology — from GPS for location to heart rate for possible stress. She and University of Massachusetts colleague Deepak Ganesan have just won a $185,000 grant from the university President’s Science & Technology Initiatives Fund to launch that effort.

[module align="right" width="half" type="pull-quote"]You drive in to McDonald's and your phone pings you: 'Are you sure you want to go in? Last week you ate a 1,200-calorie meal here.'[/module]

Imagine: You drive into McDonald's and your smartphone, picking up your location via GPS, pings you: "Are you sure you want to go in? Last week you ate a 1,200-calorie meal here." Or it's 3 p.m., just an hour away from your usual mid-afternoon junk-food foray, and the app suggests: "You often grab candy at 4 p.m. Perhaps you'd like a snack of protein and whole grains now to head that off?" Or it's a brutal workday, and the app says, "Your body is showing signs of stress. Perhaps you'd like to do some deep breathing rather than self-soothing with food?"

When Pagoto works with weight management clients, she tends to see them weekly, which means they come in and describe the moments when they fell down on their weight-loss efforts only after the fact. "I'm not there; I can't help them," she said. "I'd love a mobile phone to be me, but me with you all the time" — even if "I'm putting myself out of business by doing this."

Similar work is under way on apps for drug and smoking addicts, aiming to detect when a relapse is especially likely based on place or stress levels or other cues.

(Some of that work is also under way at the University of Massachusetts, ExtremeTech reports.)

Pagoto's weight-loss app work is just getting under way and doesn't have a catchy name yet, she said. (Readers? Suggestions?) But she can already sketch out how it will work. For an initial period, a user will enter the food they eat and other information into the phone. The app's sensors will also evaluate external data, such as location and physiological functions. The researchers will then see which factors most predict when a person is going to overeat, and begin to generate warnings and support.

Getting a bit more concrete: She often hears something like this from patients: "I did great all week, and then on the weekend I blew it. I went out to eat and there was a party and I ate a ton more than I planned."

[module align="left" width="half" type="pull-quote"]'Friday night is a tough night for you. It's only noon. Let's talk about what we're going to do tonight.'[/module]

The app, she said, would be smart enough to know that Friday nights out are a crux point for you. "So maybe earlier in the day on Friday, it could pop up and say, 'Friday night is a tough night for you. It's only noon. Let's talk about what we're going to do tonight.' Maybe you could pull up the menu for the restaurant and decide what to order. So you're not sitting at the restaurant, starving since 3:00, confronted with the bread basket and the pina coladas."

The app might also suggest, "Let’s try to manage your hunger between now and when you go to the restaurant. What if you have a high-protein snack at around 4:00, so when you get to the restaurant at 6 or 7, you're not ready to chew your arm off?"

Confession: Every afternoon at around 3 or 4, I'm hit by massive carbohydrate cravings that I can only satisfy with copious quantities of crackers or bread. How, I asked, might the app help me?

The app could perform "mini-interventions," Pagoto said, making suggestions as that carb-craving period approaches. It could pop up at lunch and ask, "Have you had enough protein?" Or, "You just ate lunch and your whole day you've only had 600 calories. You'll be hungry by 3, you need to have a snack with protein."

Hmmm. Until the phone app work comes to fruition, I think I'll try a lower-tech version: a daily routine of high-protein snacking at 2:30.

And meanwhile, I commented to Pagoto that it sounds like the app's interventions will be very gentle. Mightn't they get a little harsher for those of us who need sticks added to our carrots?

It's a thought, she said. If you wanted to make McDonald's off-limits to yourself, perhaps you could program your app to say: "What are you doing?! You're going into McDonald's! I'm going to power your phone down for an hour if you go inside!"

Postscript: It doesn't do the physiological sensing — at least, not yet — but the app PhotoCalorie has long since embarked in the direction the UMass researchers are headed. We hear from Dr. Mark Boguski, an MD-PhD genomics and informatics expert whose entrepreneurial enterprises include development of PhotoCalorie, that it already has the capability to predict when dieters are at risk of falling off the wagon. It analyzes their eating patterns, predicts when they're slipping and sends them an email to help fight off temptation. "Some people like the reminders, others want us to turn them off," he says.