Does it Work? Installment #2

Posted , posted by Erica Levine | Posted in Does it Work?, Featured

Welcome to our second installment of Does it Work?

Each installment, we review studies testing digital health approaches and discuss whether they provide evidence that technology can make an impact on health outcomes. 

For our second installment, we review a recently published paper by Laura Svetkey, MD, Bryan Batch, MD, Pao-Hwa Lin, PhD, Gary Bennett, PhD (our Director) and colleagues at Duke University in the journal, Obesity. The paper describes the outcomes of a study called Cell Phone Intervention for You (CITY). In it, researchers randomized 365 adults ages 18-35 who were overweight or obese to either: use a weight loss app (App Only), use the app and get a some weight loss coaching (App + Coaching), or control.The app was designed by researchers at Duke. You can read about the study design here.

TL;DR- Obesity treatment in young adults using digital health: an app designed to help people lose weight sees modest results; a health coach and group meetings help, but only a little. 


Young adults have high rates of obesity (around 35%) and the most rapid weight gain rates. Weight gains in young adulthood increase the risk of developing metabolic syndrome later in life. 

Who Participated?

Young adults, aged 18-35 (average age 29). Most (70%) were women, and almost half the sample was non-white (36% Black, 8% other, 6% Hispanic).  Most reported a personal income of less than $50,000/year, and the average BMI was 35 kg/m2.

Cell Phone Intervention for You

Both groups received the same lifestyle guidelines: weigh yourself, eat a moderate amount of calories, exercise, keep a food journal, limit alcohol, and stay connected with the study. 

App Only group – got a connected scale and a smartphone with the CITY app, which had many features:

  • skills training via video, live wallpaper, and quizzes
  • food & physical activity logs
  • self-weighing prompts, progress graphs, and feedback
  • a “buddy” feature allowing participants to connect with each other
  • physical activity challenges 
  • push notifications and reminders to use the app features

App + Coaching group – got the scale and the CITY app, but no push notifications. They also received 6 group sessions during the first 6 months of the intervention, then monthly phone calls with a dietitian. The app was “entirely passive”  — there were no prompts or notifications from the app to the users’ phones.



First, engagement numbers:

In the App Only group, in the first 6 months participants used the app about 4.5 times per day. In months 7-12, about 1.5 times per day. In months 13-24, less than once per day. 

In the App + Coaching group, participants completed about 90% of group sessions and coaching calls over the 2 year study period. They interacted with the app, however, a lot less frequently than the App Only group – starting out with almost twice per day in the first 6 months, then down to less than once a day, then only about every other day in the second year of the study.

After 2 years, the control group had lost 1.44 kilograms, the App Only group lost .99 kg, and the App + Coaching group lost 2.45kg. The App + Coaching group lost the most weight at each time point, but the weight losses between groups were not statistically significant at 24-months.

What do we make of these results?

Two main takeaways: First, getting results that are “not statistically significant” is like science’s way of saying, “it’s not you, it’s me.” You tried, but nah. However! Look at the weight losses in the App + Coaching group – on average, participants in this group lost and kept 5.4 pounds off, using an app and getting a monthly phone call from an interventionist. Fad diet? No. Highly disseminable, cost effective approach that can help people slow their rate of weight gain? We think it shows promise. 

Second, we know there is a good combination of tech + human intervention that can result in weight loss, but it may not be this exact combination…

Let’s break it down a bit more, just for fun:

  1. The control group lost a lot more weight than they were supposed to the research team hypothesized they would. Darn you, placebo effect! (But also we’re very happy that people are losing weight. Yay public health!)
  2. Current medical guidelines recommend a weight loss of >3% to confer health benefits. In CITY, between 22-28% of participants lost at least 5% of their baseline weight over 2 years, depending on the group.
  3. The engagement results track closely with the weight loss results. We’ve noticed a trend with these sorts of technology-based weight loss interventions, and it can be boiled down to: you use it, you lose it. 
  4. Coaching from a human is really helpful when it comes to weight loss. 
  5. Weight loss is HARD. It’s hard to get the pounds off, and even harder to keep them off.
    1. Weight loss is even harder among young adults. This is a group who is young enough not to have health issues looming on the horizon, and old enough to make autonomous food and physical activity choices, but also have well established eating patterns and preferences. What would make them want to change their behavior? ¯\_(ツ)_/¯ 
  6. Data! With the paucity of data from commercial apps, this study and its lessons will be a great resource for anyone designing a weight loss app using evidence-based principles.