Digital coaching systems for type 2 diabetes “reduce the need for more expensive, in-person visits,” said Anne L. Peters, MD, arguing for the pro side during a debate at the recent virtual American Diabetes Association (ADA) 81st Scientific Sessions about whether such coaching systems are worth the cost.
They offer a different way — among other options — for patients to manage chronic health issues such as diabetes, hypertension, weight management, and mental health, she said, adding, “My only caveat is it only works if people use it.”
The topic is timely given the proliferation of these systems over the past 5 years.
But Kasia J. Lipska, MD, arguing for the con side, wonders whether such systems are a good investment. Although digital coaching has potential, we don’t yet have a robust body of evidence that it is effective, she added.
“Many patients still lack access to healthcare and medications (including insulin) required to sustain life, so is digital technology the right use of our [limited] resources?” wondered Lipska, associate professor of medicine, Yale School of Medicine, New Haven, Connecticut.
“The marketing is way ahead of the science. Let’s get some better evidence before we start cashing our patients’ checks.”
What Is a Digital Coaching System for Type 2 Diabetes?
Following the debate, Medscape Medical News invited the speakers to dig deeper on the topic.
A digital coaching system “involves a smart phone/computer/tablet that connects a patient to a program that involves glucose monitoring, advice [from a health coach] on how to manage diabetes, weight, blood pressure, and even mental health,” said Peters, professor of clinical medicine, Keck School of Medicine, University of Southern California, noting that generally patients subscribe and pay a monthly fee.
“We were talking about commercial systems that support patients between clinic visits with their diabetes care provider,” not telemedicine, Lipska explained.
The systems can consist of a glucometer that triggers algorithmic responses based on blood glucose readings in the form of text messages, or a “real” (human) coach may connect with a patient via text or video.
Diabetes Coaching Systems Do Cut Costs
Most of the diabetes coaching programs started as diabetes prevention programs, said Peters.
According to a report by Close Concerns, published in Jan 2021, Noom has the largest membership (more than 50 million members), but this is also a weight management program, Peters stressed.
The next 10 largest programs, which are specifically for diabetes or prediabetes management, are One Drop, mySugr, Lark Health, Livongo (which is now Teladoc), Omada Health, Cecelia Health, Canary Health, Dario Health, Onduo, and Virta Health.
According to Peters, the programs are generally “more similar than dissimilar.”
Diabetes care is expensive, especially if patients develop complications, so employers and other insurance groups are looking for ways to cut costs.
A recent study reported cost savings and reduced healthcare use by employees in Michigan who used a digital diabetes prevention program (J Health Econ Outcomes Res. 2020;7:139-147).
Compared with the prior year, in the first year the program was used, healthcare spending dropped by more than $1000 per person, mostly due to fewer hospital admissions and shorter stays.
In another study, 10,000 employees with diabetes were invited to opt in to a Livongo diabetes coaching program with costs covered by the employer (J Med Econ. 2019;22:869-868). About 2000 individuals joined and 8000 did not. Use of the program was associated with a 22% lower rate of medical spending (P < .01).
Another study evaluated the Better Choices, Better Health Diabetes (BCBHD) self-management program in 558 patients versus controls (J Med Internet Res. 2018;20:e207). Emergency room and out-patient visits decreased among program users, with direct and indirect cost savings of $815/year and $1504/year, respectively.
Other researchers evaluated a digital program for gestational diabetes in 161 women. The women who were randomized to the digital program had 56% fewer medical visits and 16% lower direct costs, but there were no between-group differences in A1c, or maternal or baby outcomes (Diabetes Technol Ther. 2020;22:195-202).
And another study of a user utility score that measured patient engagement with a diabetes coaching system over 12 months found that higher scores (more interaction) were associated with lower A1c levels (JMIR Mhealth Uhealth 2021;9:e17573).
It is important to note that patients need to have intact cognitive ability, digital access, and ability to use email, as well as insurance coverage or ability to self-pay, Peters stressed, which will exclude certain patients.
However, “if somebody is interested and connected to this digital world [and is insured or can cover the cost], this can help. These programs can help reduce cost; they can help improve quality.”
Diabetes Coaching Systems Do Not Cut Costs
Lipska noted that the digital diabetes care market will be worth US $742 million in 2022, according to one report (Research2Guidance report).
Some companies sell diabetes digital coaching system memberships to consumers (such as Noom at $59/month and mySugr at $49/month), whereas others like Lark Health, Livongo (now part of Teladoc), Omada Health, and Onduo sell their services to insurance companies, health plans, or employers. Some companies do both.
“I’m going to disappoint you,” Lipska said, “because there isn’t a whole lot of evidence that these coaching systems work. Longer-term evidence about clinical effectiveness is truly lacking.”
A single-arm study with no control group evaluated digital diabetes prevention coaching from Omada Health (J Aging Health. 2018;30:692-710). Of 9498 patients who were invited to join, 796 responded (8%) and 501 were enrolled. At 12 months, they had a 7.5% weight loss, a 0.14% decline in A1c (only available for 14% of participants), and a decrease in total cholesterol (only available for 27%), but this study had inherent weaknesses.
In a stronger study, unpublished findings from the PREDICTS trial presented at Obesity Week 2020, showed that of 599 individuals randomized to a control group or to receive digital diabetes prevention coaching from Omada Health for 12 months, A1c decreased more in those who received coaching, but the difference was less than 1% (Omada Health press release).
And in a study from New Zealand, 429 adults were randomized to a control group or to the BetaMe/Melon mobile-device and web-based program with health coaching for 12 months. The study was well done and published, but the intervention didn’t work (Diabetologia. 2020;63:2559-2570), Lipska said.
“I am a little bit skeptical whether digital coaching can replace that human in-person relationship,” she said, emphasizing, “Really what’s required is that feedback between the patient and clinician, based on a robust relationship that we build over time.”
And patients can feel a little overwhelmed with all the feedback from digital systems, she said. “I get free strips, I just ignore all the calls,” one of her patients told her.
“The thing that really bothers me about these digital coaching systems,” Lipska added, “is that they are really not aligned with the healthcare team. They typically operate outside the healthcare system. I don’t have input into what kind of feedback my patient gets, what kind of goals are set for them.”
This may lead to a disjointed and confusing care for some patients.
However, Lipska agreed with Peters that if digital coaching systems are to be used, they are best for patients who have the necessary skills to engage with the technology, are comfortable with pervasive behavioral surveillance, and have insurance or can pay for these programs themselves.
The Debaters‘ Rebuttals
“The topic of the debate was cost, not efficacy,” Peters said. “High quality randomized controlled trials are unlikely to be done with these programs, and the bottom line is to prove noninferiority to ‘in person’ programs.”
Digital coaching systems may be cost effective because they don’t require a physical office building for staff and patients, she stressed. Moreover, “if they are useful for some but not all patients, I consider that a positive because many patients don’t get the care they need currently and engaging in a digital health program might fill the gap.”
“There is no one-size-fits-all solution to diabetes management,” Peters concluded, “but if we work to develop multiple high-quality approaches to care, then it is a win-win for patients and the healthcare system.”
On the flip side, Lipska said: “I agree with many points that Dr Peters made. And I’m of course not philosophically opposed to digital coaching systems. However, we still differ on some points.”
The question is whether digital coaching systems are worth the cost. To answer this question, we have to ask whether they work — ie, whether they improve outcomes, Lipska stressed.
“Once we are on solid ground with that and understand what the magnitude of the benefit is (if indeed it’s there) and who they work for, then we can start looking at whether they are associated with cost savings.”
“I am still unconvinced that we have enough data to say that they work. The studies are small-scale, often uncontrolled, often not randomized, and thus subject to selection bias. The few studies that were randomized and controlled showed no benefit or a benefit of a tiny magnitude,” she reiterated.
“I haven’t seen anything in Dr Peters’ presentation that changed my position about the state of the evidence,” Lipska concluded.
Peters has reported serving on advisory boards for Abbott, Eli Lilly, Medscape, Novo Nordisk, Vertex, and Zealand; receiving research funding from Dexcom and Insulet and devices from Abbott; and receiving stock options from Omada and Teladoc (formerly Livongo). Lipska has reported no relevant financial relationships.
ADA 2021 Scientific Sessions. Presented on June 26, 2021.
Need Your Help Today. Your $1 can change life.