We published our first efficacy analysis back in 2015, when MyNetDiary had 5 million users. That study measured weight outcomes for users who tracked actively for at least four consecutive weeks. It served us well for years, but the app, the user base, and the science have all changed since then. We retired it.
This is the replacement -- and it is a fundamentally different kind of study. We submitted a peer-reviewed research letter to the Journal of Medical Internet Research (JMIR), one of the highest-ranked academic journals in digital health. The pre-print is available for open review at JMIR Preprints.
Here is what the study covers, without giving away the results:
We took every user who signed up for MyNetDiary in a single month -- over 400,000 people -- and applied strict filters to find the ones who genuinely committed to food tracking over six months. The final cohort of 3,878 users had to set a weight-loss goal, log food on at least 21 days, and weigh in every month for six consecutive months. Users who gained weight were included. Nobody was excluded for having bad results.
We then examined whether how often someone tracks their food -- daily versus a few times a week versus intermittently -- predicts how much weight they lose, after adjusting for age, sex, and starting body weight. We also looked at something that has not been studied at this scale before: whether building a consecutive-day tracking streak predicts outcomes independently of average tracking frequency.
The differences from the 2015 study go beyond the numbers. That analysis required only four weeks of active tracking, excluded anyone who lost less than one pound, and was published on our website without peer review. The new study follows users for six months, includes everyone regardless of outcome, applies a multivariable logistic regression, and is going through formal peer review at an academic journal.
The study was co-authored by Sergey Oreshko, CEO and Co-Founder of MyNetDiary, and Susan Heikkinen, MS, RDN, CDCES, BC-ADM, ACE-PT, our registered dietitian nutritionist. The analysis used MyNetDiary's staff-verified food database, which draws from the same USDA and NCC research-grade sources used in clinical nutrition research and tracks up to 108 nutrients per food entry.
We will publish the full results and a detailed breakdown on this blog once the peer review process is complete. In the meantime, the pre-print is open for public review at JMIR.
