August 31st, 2018

Lark Diabetes Prevention Program (DPP) Outcomes Update

  • Average Body Weight Loss at Year 1: 6.53%

  • Year-over-Year DPP Patient Growth: 13x

  • Average Enrollment Completion: 92%;

  • Average Number of Sessions per Month: 7.6

November 7, 2017


For more information on Lark, contact:

Abigail Fredenburg,


JMIR Diabetes Study Finds AI Health Coaches as Effective as In-Person Health Care Professionals

Fully automated, conversational AI program promotes weight loss to prevent type 2 diabetes 

MOUNTAIN VIEW, CALIF., November 7, 2017 – New peer-reviewed findings published in JMIR Diabetes demonstrate that an artificial intelligence (AI) health coach is just as effective as in-person weight loss programs, such as the Centers for Disease Control and Prevention-led National Diabetes Prevention Program (DPP), a clinically-proven weight loss program focused on type 2 diabetes prevention.

The jmir publications study found that Lark's personalized AI health coach was not only widely accepted and used at high rates, but also had comparable results to programs led by in-person health care professionals when it comes to weight loss, a key measure to preventing those with pre-diabetes from getting type 2 diabetes.

“With diabetes or pre-diabetes affecting half of the American population, it is critical that we use every technology at our disposal to treat and prevent this disease. With AI, we can provide the care and compassion people deserve in a way that is cost-effective and scalable to all who need it,” said Lark CEO and co-founder Julia Hu. 

Participants in the study could have unlimited coaching sessions with Lark. They averaged 103 individual coaching sessions over the course of the study. Users lost an average of 2.4 kg (or 2.4 percent body weight), compared with an average of 2.32 kg reported in a DPP meta-analysis of 22 lifestyle intervention studies with in-person components. Additionally, the percentage of healthy meals eaten increased by 31 percent. Patients were engaged with their health coach and reported a satisfaction score of 87 out of 100.

“These findings reveal that people are not only reacting positively to AI coaching, but also following the health guidance they are given. The clinically proven results show the great potential for bringing cost-effective chronic disease prevention and management interventions to millions of people,” said Natalie Stein, MPH, MSc, lead author and an instructor in Michigan State University’s College of Human Medicine.

The longitudinal study observed weight loss, changes in meal quality, and app acceptability among 239 overweight (body mass index ≥ 25) patients who used the Lark Weight Loss Health Coach, which encourages weight loss through elements of cognitive behavioral therapy. The Lark AI Health Coach monitors a patient’s health through phone sensors and integrated health devices, and then mimics nurses’ empathetic counseling through casual conversations that use text-based communication and other interactive elements.

“Lark uses a holistic approach to not only treat chronic disease, but also strengthen a person’s mental and physical health. Rooted in cognitive behavior therapy, it provides care and compassion to help a person prevent or maintain their chronic disease for life,” said Jo Solet, PhD, assistant clinical professor of medicine at Harvard Medical School and a Lark Health Committee member.  

Designed by leading behavior change experts, medical informatics, health faculty from Harvard and Stanford, and AI technologists, Lark’s one-on-one AI chronic disease solution was developed over the past six years through 355 million interactions with a million patients. Every Personal Lark Coach uses open access scientifically-proven methods to help members manage or prevent chronic conditions with lifestyle wellness coaching and behavioral health counseling, type 2 diabetes prevention, type 2 diabetes management, and hypertension management solutions. Because it uses AI and is accessible via smartphone, Lark is available to patients anytime, anywhere, and is infinitely scalable. With this ease of use, Lark has engaged 83 percent of DPP-eligible patients within 72 hours of being referred to its AI platform. This year, Lark became the first AI-led DPP to be fully reimbursable with a CPT code.

“Lark is tackling some of society’s largest health problems – rising health care costs, chronic disease management, and a shortage of healthcare professionals needed to meet patient demand. This study proves that Lark’s AI Personal Health Coach is as effective as an in-person nurse. This is a game-changer that provides a turnkey solution for payors and enterprises wanting to improve people’s health and bring down costs,” said Cora Tellez, former president of Health Net health plans, senior executive at Blue Shield of California and Kaiser Permanente, and a Lark Health Committee member.  

The full study, “A Fully Automated Conversational Artificial Intelligence for Weight Loss: Longitudinal Observational Study Among Overweight and Obese Adults” is available at

About Lark

Lark is an AI chronic disease platform that brings unlimited, one-on-one chronic disease prevention and management to all those who need it. Relying on cutting-edge AI proven in more than 355 million interactions with a million patients, Lark offers payors and employers an affordable, infinitely scalable solution that can be implemented quickly and seamlessly. Lark is the only fully-AI driven Diabetes Prevention Program recognized by the Centers for Disease Control and Prevention, and has engagement rates exponentially greater than in-person programs as well as superior health outcomes. Fully reimbursable and available to 40 million Americans to manage pre-diabetes, Lark is able to bring the care and compassion that people with chronic diseases deserve, but too often cannot find. Backed by Fenox Capital and the Stanford StartX Fund, Lark is headquartered in Mountain View, Calif., and has been recognized by Business Insider as one of the “10 Most Innovative Companies in the World,” alongside Airbnb, Snapchat, and Uber.