Home / Resources / Articles / Diabetes Risk Reduction Seen with a Digital Health Program, Human Coaching

Diabetes Risk Reduction Seen with a Digital Health Program, Human Coaching

May 6, 2017

Study measures effects of apps on weight reduction.

As the prevalence of obesity continues to rise in the United States (35% in males, 40% in females in 2016), the risks of type 2 diabetes rises as well, supporting the need for ongoing weight control as a simple means of reducing diabetes risk. This concern is especially important in the aging adult population, where at least 66% of those over 65 meet current Centers for Disease Control (CDC) criteria for obesity or being overweight, while 51% of the same population is at risk for type 2 diabetes (US Census Bureau, 2014). The practice of healthy eating and regular exercise in overweight patients reduces the risk of progression from prediabetes or metabolic syndrome to diabetes. Several studies that support this notion have led to the American Association of Clinical Endocrinologists and the American College of Endocrinology including these lifestyle modifications in their first clinical practice guidelines in 2016.


Over time, lifestyle intervention programs were traditionally conducted in a face-to-face environment, typically with an in-person coach. While highly effective, patient barriers to participation (lack of access or lack of interest in this type of program) have demonstrated lack of response in patients who were unable to partake when compared to outcomes in patients who were able to do face-to-face sessions. Recent advances in technology have afforded patients with tools to remotely achieve high-level interaction with a live coach for behavioral counselling, group support, self-paced education, and development of important skills geared toward success.  However, very few data are available looking at the Diabetes Prevention Program combined with technology use in older adults. With the rise of Internet availability to the aging population, C.M. Castro-Sweet et al. retrospectively observed a group of Medicare prediabetic patients, examining the effects of an Internet-based DDP digital application on weight status and progression to diabetes.

Subjects were selected from the Humana Medicare Advantage insurance program in the United States. High-risk participants were invited through a 3-week marketing campaign of direct mail, phone call and email contacts describing eligibility for the study as a benefit of their insurance plan. A total of 9,498 members were contacted, of which 501 eligible patients were enrolled. Eligibility was defined as presence of metabolic syndrome, or a combination of three of the following:  prediabetes, hypertension, hyperlipidemia, and obesity.  Weights were measured and electronically uploaded with an electronic scale. Using Internet-enabled devices (smart phones, computers, tablets), the participants partook in a 16-week intensive curriculum of weekly interactive lessons. Each module was available to all participants for one week at a time, allowing the patients to learn asynchronously but on the same schedule.  The program application allowed participants to track weight loss, physical activity, and daily food intake. Following the 16-week weight loss program, a 36-week program focused on weight maintenance was started, for a total intervention length of 12 months. Throughout the duration of the study, all patients were monitored for psychological markers and overall progress using three different approved surveys: The World Health Organization-5 (WHO-5) Well-Being Index (measures subjective well-being), the Patient Health Questionnaire for Depression and Anxiety (PHQ-4), and the Summary of Diabetes Self-Care Activities (SDSCA) scale (reports self-management behaviors relating to diabetes).

Of the 501 participants (mean age 68.8 years), 95% completed at least 4 weekly lessons, and 92% completed 9 or more lessons in the initial 16-week phase. Only 2% failed to complete at least one lesson. Weight change from baseline was found to be significant (mean loss: -6.5%, SD 4.0 at 16 weeks, -8.0%, SD 7.7 at 6 months, and -7.5%, SD 7.8 at 12 months; p=0.01 for all). In the subset of patients (n=96) with available HbA1C and cholesterol measurements, the declines from baseline were also statistically significant (12-month improvement: HbA1c -0.14%, p=0.001; cholesterol -7.08 mg/dl, p=0.0001). As the self-reported surveys were optional, 57% (n=285) provided the surveys. The Who-5 Well-Being Index showed improvement in scores (+12%, p=0.0001), while the PHQ-4 Depression and Anxiety showed improvement in depression scores (-0.26, p=0.0001), but not in anxiety scores (-0.10, p=0.1922).  The SDSCA also showed improvement in all 5 markers of healthy lifestyle, with statistical significance, with healthy modifications (better diet, exercise habits) increasing, and unhealthy behaviors (fast food consumption) decreasing.

This study found that willing participants enjoyed a high degree of success, meaningful engagement, and important benefits. Weight loss at for 6 and 12 months (8% and 7.5%) exceeded the standards set by the CDC’s National Diabetes Prevention Program (5% reduction). Population presence in 37 states shows good geographic generalizability, supporting the scale that this approach could achieve in reaching patients. Limitations included a lack of a control group, although on average the subjects were not meeting physical activity standards, but by week 16, were all meeting the recommendations. A control group would have provided a better comparator for all data collection points in the 12-month period.

Practice Pearls:

  • With the rise in prevalence of obesity in the U.S., diabetes risk is also on the increase.
  • Rapidly advancing technology and its increasing availability has provided a useful means of educating and interacting with the aging population.
  • Behavioral modifications through digital applications have shown to have benefits in diabetes risk reduction.


Castro Sweet CM, Chiguluri V, Gumpina R, et al. Outcomes of a Digital Health Program With Human Coaching for Diabetes Risk

Mark T. Lawrence, University of Colorado-Denver, School of Pharmacy NTPD