Applying Artificial Intelligence to Chronic Condition Management
Every member is different, and their coaching should be, too.
But that’s only possible with artificial intelligence and machine learning. So what is AI and why should you care about it for your employees and members?
AI is not about aggregating information and poring over interesting insights (analytics) that make themselves available in analyst reports and research papers - AI is about getting real-time feedback from members about what impacts their behaviors and feelings have on their chronic condition, and iterating on that feedback with real-time, 24/7 coaching. The end result is better outcomes and a member experience that feels intimate and rewarding.
Applying AI and training it on over 1.9 million patients has led to the only data-driven, personalized interventions to drive behavior change in chronic disease management and prevention.
Lark has applied AI and Machine Learning technologies to build a complete chronic condition management solution that provides:
1-on-1 Personal Health Coaching, 24/7
Health Data Monitoring from Connected Devices, Activity, Eating Habits Claims Data, and More
Behavior Change With Clinically Validated AI & Coaching by Integrating Cognitive Behavioral Therapy
Personalized Care Management Scalable to Millions with Seamless Deployment and Enrollment
Not Device-Centric, and Not Just a Single Point Solution
Artificial intelligence and machine learning surround us, and they’re the natural procession of the advent of big data. With all of the data being collected, disseminated, and stored digitally, AI and ML are the ways of using it by layering models on top. But models are for analysts - which is why Lark’s insights are used to coach, guide, and reinforce embers in their data to day life.
"Chronic conditions are the biggest challenge that AI has come up against. Lark's AI coaches have been trained on over 1.9 million members… Lark's mission to provide everyone with a chronic condition with care and compassion means that immediate, compassionate, and impactful feedback on members' blood glucose and blood pressure measurements must be both compassionate and adaptive. I'm thrilled to be part of this journey," Dr. Rapoport, Lark’s Vice President of Data Science, said.
– Dr. Sharon Rapoport, Vice President of Data Science
"Sharon has headed Data Science teams at immense scale across industries like travel and financial services, and her leadership will help us scale from currently serving 1.5 million members to serving tens of millions of patients. We have an opportunity to provide infinitely scalable chronic condition management and prevention services by combining cutting edge AI and behavior change technology and easy-to-use connected health monitors," said Lark Chief Executive Officer Julia Hu. "She's a perfect fit to join our management team."
– Julia Hu, Cofounder and Chief Executive Officer
To be effective, AI-based condition counseling must be based on proven behavioral change methods. Lark’s AI was first put to the test in this study in JMIR showcasing weight loss and diabetes prevention with AI.
“This study demonstrates AI’s potential to provide compassionate care that is associated with weight loss, increased healthy lifestyle behaviors, and user trust that can reduce diabetes risk,” the study reads.
AI applied to chronic disease management s can use evidence-based solutions in additional ways. For example, Lark’s Diabetes Prevention Program is a CDC-Recognized Diabetes Prevention Program which combines human and AI coaches. But peer-reviewed findings published in JMIR Diabetes revealed 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. Diabetes Prevention Program (DPP) cirriculum has been shown to lower type 2 diabetes risk by over 50%.  The health coach, which is delivered via smartphone app, also incorporates CBT and evidence-based coaching on blood sugar-lowering behaviors including weight loss, physical activity, and sleep.
Careful design of disease management programs can pay off, as this health coach has been shown to help users lose weight,  which is an important component of diabetes prevention. Each kilogram of weight loss in an obese individual with prediabetes is linked to a 16% reduction in type 2 diabetes risk. 
”Lark’s AI Health Coach supports weight management and diabetes prevention by monitoring a patient’s health through phone sensors and integrated health devices. The technology then mimics nurses’ empathetic counseling through text-based interventions.”
– Point Clear Solutions
The Potential for AI in Disease Management: Patient Behavior Change
What if there were evidence-based, proven interventions to cause patients to adopt these and similar healthy behaviors? Consider the potential of automated reminders to take medications, replace costly medications with generics, make medical appointments, get instant feedback on meal choices, and suggestions for when to get in a workout or get ready for bed.
These seemingly simple choices are not easy for patients to make on their own (if they were, weight loss would not be challenging!). Nor are they adequately facilitated by healthcare professionals. Lark Health’s AI-driven coaching is designed to increase the healthy behaviors that can improve health so dramatically.
While we can’t all carry an MD in our pockets—that would be expensive and, frankly, awkward—we can and already do bring our phones along. And maybe a sophisticated AI coach in your pocket is what could make you rethink the next time you reach for the bacon.
Benefits of AI for Disease Management: Better Health (And Lower Costs)
Any amount of disease prevention or improved management lower costs, and Lark’s AI is designed to improve some of the most important behaviors that affect patients’ health:
Self-monitoring of values such as blood sugar in diabetes or blood pressure in hypertension.
Medication adherence in conditions ranging from depression and schizophrenia to hypertension and high cholesterol.
Weight loss and physical activity in prediabetes, diabetes, hypertension, and obesity.
One source estimates that, “Over 80% of U.S. healthcare spending is strongly rooted in unhealthy behaviors and poor drug compliance!” 
The intelligence of AI and the compassion of Cognitive Behavioral Therapy have led to the first integrated feedback loop for chronic disease management in Lark.
Benefits of Digital Disease Management Programs and AI: Scalability
It is one thing to offer an expensive device or health coach to the most at-risk employees or plan members. It is quite another to offer scalable, compassionate AI to tens of thousands with a few clicks.
Each participant can still receive:
The same high quality of coaching that incorporates the expertise of multiple health authorities;
A customized program with personalized coaching due to innovative AI;
Instant responses 24/7 without you needing to pay for more nurses or hire thousands of health coaches to operate the phone lines.
Scaling up can happen quickly and without headaches if you work with Lark. Consider these additional benefits:
An enrollment team to get you started by identifying members/employees, verifying eligibility, and enrolling new patients.
Seamless delivery to each patient’s door and easy setup of a connected device, such as a body scale, glucometer, or blood pressure meter.
A dedicated customer support team to keep things running smoothly.
Benefits of Artificial Intelligence: Widespread Access
If we know how to prevent and better manage these conditions, why are we not doing better for the millions of Americans with them? Lack of access to healthcare is a major barrier to proper – adequate, high-quality – care. Only a fraction of those who could benefit get the care they need. Access is limited because of reasons such as:
A shortage of doctors and other healthcare professionals, making appointments expensive and hard to get.
Only a small reach of preventive programs compared to treatment for more advance conditions.
Trouble accessing care because of distance (such as in rural areas) or time (such as needing to take time off work).
Tendency for poorer-quality care to be available in less privileged areas.
Disease management programs can narrow, and eventually close, these gaps. Using AI or other automated response mechanisms helps overcome shortfalls in healthcare provider availability. Remote treatment, such as health coaching via smartphone apps, reporting mood via a website, or automatically sending blood sugar measures to doctors’ computers let patients get care regardless of where they live or what their schedules are.
Benefits of Lark’s AI in Disease Management: Big Data
Are medications effective? Which lifestyle choices have the most impact on which conditions? Are certain demographics more responsive than others to certain treatments, behaviors, or strategies?
Research is traditionally expensive and limited. A single clinical trial can easily cost millions of dollars and leave many questions unanswered as it investigates only one specific question. Disease management programs give free reign to imagine what continuously collected, low-cost data could do to revolutionize healthcare.
Drug companies could find out minute-to-minute information about their medications’ metabolism, effects on target markers, and potential interactions with other medications, foods, or substances. Providers could learn more about specific effects on mood, blood pressure, heart rate, or even hormones, of patient behaviors ranging from deep breathing to lifting weights. This data can all be made available for analysis and distribution for only a tiny bit of the cost of traditional data collection.
The Future Is Here
AI may drive the healthcare of the future, but it’s also happening right now. You are losing ground if you are not already using AI digital therapeutics for your members or employees to improve health, prevent chronic diseases, and/or mitigate the ones they already have. You can get started today and learn more.