ONBOARD: A Feasibility Study of a Telehealth-Based Continuous Glucose Monitoring Adoption Intervention for Adults with Type 1 Diabetes

Abstract

Background:

Continuous glucose monitoring (CGM) can improve glycemic control for adults with type 1 diabetes (T1D) but certain barriers interfere with consistent use including cost, data overload, alarm fatigue, physical discomfort, and unwanted social attention. This pilot study aimed to examine feasibility and acceptability of a behavioral intervention, ONBOARD (Overcoming Barriers and Obstacles to Adopting Diabetes Devices) to support adults with T1D in optimizing CGM use.

Methods:

Adults (18–50 years) with T1D in their first year of CGM use were invited to participate in a tailored, multicomponent telehealth-based intervention delivered over four 60-min sessions every 2–3 weeks. Participants completed surveys (demographics; diabetes distress, Diabetes Distress Scale for adults with type 1 diabetes; satisfaction with program) and provided CGM data at baseline and postintervention (3 months). Data were analyzed using paired t-tests and Wilcoxon signed-rank tests.

Results:

Twenty-two participants (age = 30.95 ± 8.32 years; 59% women; 91% non-Hispanic; 86% White, 5% Black, 9% other; 73% pump users) completed the study. ONBOARD demonstrated acceptability and a high rate of retention. Moderate effect sizes were found for reductions in diabetes distress (P = 0.01, r = −0.37) and increases in daytime spent in target range (70–180 mg/dL: P = 0.03, r = −0.35). There were no significant increases in hypoglycemia.

Conclusions:

Findings show preliminary evidence of feasibility, acceptability, and efficacy of ONBOARD for supporting adults with T1D in optimizing CGM use while alleviating diabetes distress. Further research is needed to examine ONBOARD in a larger sample over a longer period.

Introduction

Type 1 diabetes is a burdensome chronic disease that requires constant attention to glucose levels, food intake and physical activity, and insulin dosing decisions. At present, <25% of adults with type 1 diabetes (T1D) meet recommended glycemic targets of hemoglobin A1c (HbA1c) <7%.^1,2 Major advances in continuous glucose monitoring (CGM) technology can reduce self-management burden and improve health outcomes and quality of life for individuals with T1D.^3,4 Using CGM reduces self-management burden of fingersticks, and there is evidence that CGM use leads to reductions in diabetes distress and worry about hypoglycemia.⁴ CGM is also a key component for closed loop systems that have been shown to reduce hypoglycemia and glucose variability, increase time in range,^5,6 and decrease the mental burden of living with T1D.^7,8 However, despite the many benefits of CGM, adoption rates continue to be low. According to the most recent data available, although greater than half (∼63%) of individuals with T1D use insulin pumps in the United States, only 22% of adults 18–25 years of age and 37% of adults 26–49 years of age use CGM.¹ Cost and insurance barriers are commonly reported⁹ and contribute to lower rates of consistent use and reduced benefit.^10,11 Once adopted, CGM must be used “as close to daily as possible for maximal benefit” according to the American Diabetes Association's Standards of Care.¹² However, a survey found that 27% of CGM users discontinue use in the first year and up to 30% do not use their CGM often enough (<70% of the time) to be considered beneficial.¹³

Adults with T1D have endorsed multiple barriers and hassles associated with using CGM including (1) physical: hassle/burden of wearing devices on the body; (2) managing data: navigating the amount of alerts/alarms and information from devices; (3) social: devices drawing unwanted attention from others; and (4) difficulty trusting the device.⁹ Beyond navigating these hassles and burdens associated with using a diabetes device for daily self-management, CGM users must also adapt to receiving a continuous stream of glucose information. Although some studies have demonstrated decreases in diabetes distress from CGM use,⁴ other studies of CGM users have reported increased diabetes distress, anxiety, and negative affect.^14
–16 The availability of constant data from CGM makes salient the unrelenting nature of T1D, which, in turn, may exacerbate distress and burnout.^17
–19 Therefore, CGM users may benefit from enhanced support and training to manage and navigate common hassles and potentially associated experiences of diabetes distress to experience the benefits of CGM.

In fact, the American Diabetes Association (ADA) recommends “robust education, training, and support” for those learning to use CGM.¹² To date, a limited number of European studies have examined and demonstrated the benefit of an enhanced CGM training and support program^20
–22; two of these studies focused on intermittent flash glucose monitoring. The training program for real-time CGM emphasized knowledge and skills to use CGM and observed a small decrease in HbA1c at the end of training.²² There is an opportunity to develop and test high-quality, evidence-based CGM onboarding support that enhances knowledge and problem-solving skills while concurrently attending to the emotional side of adjusting to interacting with and responding to a high volume of glucose data.²³

We developed a four-session CGM-focused behavioral intervention called ONBOARD (Overcoming Barriers and Obstacles to Adopting Diabetes Devices) to address the need for increased support for CGM initiation. The goals of ONBOARD are to provide information, skills, and psychosocial support for managing the above-mentioned key barriers to device use (physical, data, social, and trust) and attending to the distress that may arise during CGM adoption. The current mixed methods study had two main goals: (1) to examine the feasibility and acceptability of ONBOARD for adults with T1D; and (2) to examine preliminary evidence of efficacy of ONBOARD in its ability to improve glycemic outcomes (i.e., time in glucose target range) and diabetes distress. We hypothesized that ONBOARD would be feasible and acceptable for adults with T1D and lead to increases in time in range and decreases in diabetes distress.

Methods

Participants and procedure

Eligible participants were adults 18–50 years of age with T1D who were in their first year of using CGM (or had their supplies and were willing to start using the device during study participation). Participants were recruited beginning in January 2019 and data collection was completed in October 2019. Participants were recruited through multiple avenues: in-clinic at an academic medical center's adult endocrine clinic; through online and social media postings to T1D community groups; through referrals from diabetes educators outside of the academic medical center; and through an independent research database of individuals with T1D who expressed interest in learning about research studies. Potential participants were screened for eligibility and if eligible, informed consent was conducted and they were enrolled into the study. They completed an online survey through REDCap, a secure electronic data capture program and then were scheduled to attend their first ONBOARD intervention session. Sessions with interventionist (M.L.T.) were held every 2–3 weeks over an HIPAA-compliant videoconferencing software (Zoom). After completion of the full 4-session ONBOARD intervention, participants completed a follow-up assessment (electronic survey, CGM data download, HbA1c if available). Participants then attended a focus group to provide feedback on ONBOARD. Focus groups were audio recorded and transcribed; transcripts were then checked for accurateness and deidentified. Glucose data were downloaded from their CGM at baseline and follow-up. After completion of the study, participants received a total of $100 as a gift card. Study procedures were approved by the Stanford University Institutional Review Board. This study is registered on ClinicalTrials.gov (NCT 04161131) see supplementary data for study protocol. Figure 1 presents the flow of recruitment and participation for this pilot.

FIG. 1.

ONBOARD recruitment and participation flow. CGM, continuous glucose monitoring; ONBOARD, Overcoming Barriers and Obstacles to Adopting Diabetes Devices.

ONBOARD intervention

ONBOARD is a multicomponent intervention developed to support adults with T1D in optimizing their CGM use. The development of this intervention was informed by our prior research on common barriers and burdens that come up when using diabetes devices (physical, data, social, and trust),⁹ as well as by the Technology Acceptance Model,²⁴ which states that technology's perceived usefulness and ease of use influence likelihood of continued use and benefit from the technology. Overarching goals of ONBOARD were to clarify the personal benefits (perceived usefulness) of CGM while providing support and building skills for navigating common, previously identified issues, barriers, and burdens that may arise when using CGM to increase comfort with the technology (ease of use).

ONBOARD consists of four 60-min sessions delivered individually by a doctoral level psychologist with diabetes expertise over a 3-month period (every 2–3 weeks). ONBOARD was developed as a telehealth-based intervention to increase access for adults with competing demands on their schedules. Each one-on-one session targets a key theme relevant to CGM use: physical, data, social, and trust (see Table 1 for session content). Sessions were designed to be delivered one-on-one to be able to better tailor session content to each participant's individual experience of CGM and associated benefits and burdens. Each session included an introductory 5–10 min “digital storytelling” video that presented adults with T1D sharing their real-life experiences with their CGM relevant to the theme for that session. After the video, session content included reflection on video as a segue into discussing the session theme. Session content included psychoeducation, cognitive and behavioral strategies, problem-solving, motivational interviewing, and discussion of the participant's individual experience of their own benefits and barriers when using CGM. The final session included a reflection and summary of skills learned and insights developed during their participation in the program.

Table 1.

Overcoming Barriers and Obstacles to Adopting Diabetes Devices Intervention Session Content^a

No.	Session name	Strategies used	Session goals
1	Overview; overcoming physical barriers	Experiential wear	Rationale for ONBOARD; introduce CGM benefits vs. barriers and hassles; discuss common physical issues with wearing a device (appearance, insertion, skin reactions or other issues with adhesive, discomfort, placement on the body), interference with daily activities, and strategies for managing these issues.
2	Managing CGM data	Cognitive restructuring	Individualized alert thresholds; reviewing patterns and trends; target fear of hypoglycemia
3	Social barriers	Problem solving	Relevant scenarios for social concerns (e.g., being at work and an alarm goes off) to prompt problem solving
4	Trust in CGM accuracy; reflection	Motivational interviewing, review of prior sessions	Perceived benefits of the technology as they relate to, or are weighed against, perceived issues. Anticipated future problems that may lead to CGM discontinuation to brainstorm possible alternative actions

Sessions are delivered over videoconference every 2–3 weeks and are 60 min in length.

CGM, continuous glucose monitoring; ONBOARD, Overcoming Barriers and Obstacles to Adopting Diabetes Devices.

Multicomponent interventions are advantageous because they capitalize on effective pieces of existing, evidence-based interventions. In particular, multicomponent interventions that include motivational interviewing, problem-solving and/or cognitive behavioral therapy have demonstrated ability to improve HbA1c in T1D.^25
–27 Both problem-solving interventions and motivational interviewing have proven to be effective for diabetes management.^28,29 Furthermore, interventions that include first-person stories modeling how individuals worked through relevant problems, thus incorporating a social learning component, have been shown to help develop problem-solving skills and self-efficacy through “vicarious learning.”^27,30

Measures

Feasibility and acceptability

To assess feasibility, we tracked recruitment and retention rates (i.e., session attendance). We also tracked session length to calculate number of hours spent delivering the program. To assess acceptability, we administered a brief satisfaction survey for participants to rate the helpfulness of ONBOARD. We also collected feedback on acceptability through focus groups after participants completed session participation. Focus groups used a semi-structured interview guide and asked open-ended questions eliciting session-specific and overall feedback on the intervention content, length, frequency of sessions, as well as asking who participants felt ONBOARD would be most helpful for.

Demographic and medical history

We collected demographic data (age, gender, and race/ethnicity) and medical data (insulin pump use, diabetes duration, and brand of CGM).

Diabetes distress for T1D

We assessed diabetes distress at baseline and postintervention follow-up with the 28-item Diabetes Distress Scale for adults with T1D (T1-DDS).³¹ Items are on a 6-point Likert scale (sample items included the following: “feeling like I have to hide my diabetes from other people”; “feeling that I am not as skilled at managing diabetes as I should be”). The T1-DDS contains seven subscales: Powerlessness, Management Distress, Hypoglycemia Distress, Negative Social Perception Distress, Eating Distress, Physician Distress, and Friend/Family Distress. Subscale scores were created by averaging item scores within each subscale. A total T1-DDS score was created by averaging all item scores. Established cut points were as follows: mild (1–1.9), moderate (2.0–2.9), and high distress (3.0).³¹ The internal consistency in this sample was 0.89.

Glycemic metrics from CGM

Glycemic metrics were calculated using CGM data from 2 weeks before session participation (baseline) and 2 weeks after participation in the last intervention session (postintervention). These metrics included the following: % data available; glycemic variability (% CV); time spent between 70–140 mg/dL and 70–180 mg/dL overall and during daytime; time above target range (>250 mg/dL) and below target range (<70 and <54 mg/dL).

Data analysis

Descriptive statistics were used to calculate frequencies and means for demographic and medical information. To assess preliminary efficacy and examine changes from baseline to postintervention in glycemic and distress outcomes, we used paired t-tests (for normally distributed variables) and Wilcoxon signed-rank tests (for non-normally distributed variables). Traditional significance tests were complemented by tests of magnitude (effect size: Cohen's d for parametric t-tests, r for nonparametric Wilcoxon signed-rank tests) given the sample size. Focus group qualitative data were analyzed by three coders using content analysis.³²

Results

Feasibility

To recruit participants, we initially approached or were approached by 73 participants from all recruitment sources (Fig. 1). Of those, 16 (22%) did not respond after initial contact. We assessed 57 participants for eligibility. Of those, 22 (39%) did not meet eligibility criteria either because of already using CGM for more than a year (n = 13) or insurance barriers to using CGM (n = 9). An additional 7 (12%) declined to participate. Twenty-eight (49%) met eligibility criteria and provided informed consent. One participant was lost to follow-up before completing baseline survey and three were lost to follow-up after survey completion but before intervention session attendance (14% dropout rate). An additional two participants were withdrawn from the study after baseline survey completion because of not having met initial eligibility criteria. The remaining 22 eligible participants went on to complete all 4 ONBOARD sessions. Sessions lasted 57.7 ± 4.9 min on average (range: 45–70 min). The longest session on average was Session 3 (problem solving: 59.3 ± 4 min). The shortest were Sessions 1 and 2 (physical: 57 ± 5.1 min; data: 56.6 ± 5.4 min). An average of 3.9 ± 0.2 h were spent in total with each participant over the course of the intervention. Characteristics of ONBOARD participants are given in Table 2.

Table 2.

Overcoming Barriers and Obstacles to Adopting Diabetes Devices Participant Characteristics (N = 22)

Characteristic	n (%)	Mean (SD)	Range
Age (years)		30.95 (8.32)	21–50
Gender (female)	13 (59.1)
Ethnicity
Non-Hispanic	20 (90.9)
Hispanic	2 (9.1)
Race
White	19 (86.4)
Black/African American	1 (4.5)
Other	2 (9.1)
Marital status
Single	9 (40.9)
Married	9 (40.9)
Living with someone	4 (18.2)
Education
Some college	4 (18.2)
College graduate	9 (40.9)
Graduate degree	9 (40.9)
Annual household income
<$25,000	1 (4.5)
$25,000–75,000	6 (27.3)
$75,000–125,000	8 (36.4)
$126,000–175,000	4 (18.2)
$ >175,000	2 (9.1)
Unknown/declined to state	1 (4.5)
HbA1c (mg/dL) n = 15		7.63 (1.91)	5.3–11.5
Diabetes duration (years)		13.5 (8.42)	1–28
Currently using insulin pump (%Yes)	16 (72.7)
Length of insulin pump use
<0.5 year	3 (18.8)
2–5 years	1 (6.3)
>5 years	12 (75)
Pump brand
Medtronic	7 (43.8)
Tandem	6 (37.5)
Insulet	3 (18.8)
Other	0 (0)
CGM use in past month (% time)
<50	2 (9.1)
50 to <75%	4 (18.2)
75 to <100%	2 (9.1)
100%	11 (57.9)
CGM brand
Dexcom	17 (84.2)
Medtronic	2 (5.3)
Abbott	3 (10.5)
Length of CGM use
<0.5 year	13 (59.1)
0.5–1 year	9 (40.9)

HbA1c, hemoglobin A1c; SD, standard deviation.

Acceptability

Satisfaction survey results are given in Table 3. All 22 participants said they would recommend the ONBOARD program. The majority (81%) of participants rated ONBOARD as either helpful or “very helpful.” Most were satisfied with the online format and the number and length of sessions.

Table 3.

Satisfaction Survey Feedback (N = 22)

Question	Median (Range)	Mean (SD)
(1) Overall, how helpful was the ONBOARD program for you?	4 (2–5)	3.95 (0.84)
(2) Overall, how satisfied were you with the ONBOARD program?	5 (2–5)	4.5 (0.86)
(3) Overall, how satisfied were you with receiving the ONBOARD program online?	5 (2–5)	4.56 (0.86)
(4) I would recommend the ONBOARD program	4.5 (4–5)	4.5 (0.51)
(5) Overall, how satisfied were you with the number of sessions?	4 (3–5)	4.32 (0.71)
(6) Overall, how satisfied were you with the length of the sessions?	5 (2–5)	4.32 (0.89)

1 = very unsatisfied/unhelpful/strongly disagree; 5 = very satisfied/helpful/strongly agree

Glycemic outcomes

CGM data were analyzed for 19 participants who had complete baseline and follow-up CGM data (Table 4). One participant did not provide baseline or follow-up data; one was not wearing CGM postintervention and one participant experienced technical difficulties uploading baseline CGM data. There was a significant increase from baseline % daytime 70–140 mg/dL (50.9% ± 20.6%) to postintervention (57.7% ± 22%); t(18) = −2.77, P = 0.013, Cohen's d = −0.64 (moderate effect), with participants on average spending an additional 67 min in range of 70–140 mg/dL per day. In addition, there was a significant increase from baseline % daytime 70–180 mg/dL (72.6% ± 19.6%) to postintervention (77.5% ± 19.8%), Z = −2.173, P = 0.02, r = −0.35 (moderate effect), indicating on average an additional 48 min in range of 70–180 mg/dL per day. In addition, there was a small decrease in time spent >180 mg/dL from baseline (25% ± 20.8%) to postintervention (21.5% ± 21.2%), Z = −1.69, P = 0.09, r = −0.27 (small effect). In addition, there was a small decrease in % time >250 mg/dL from baseline (7.2% ± 9.6%) to postintervention (6.7% ± 10.6%), Z = −1.15, P = 0.25, r = −0.19 (small effect). In addition, there was a small decrease in time spent >180 mg/dL from baseline (25% ± 20.8%) to postintervention (21.5% ± 21.2%), Z = −1.69, P = 0.09, r = −0.27 (small effect). No significant changes were found between baseline to postintervention for hypoglycemia metrics (<54 or <70 mg/dL).

Table 4.

Paired t-Tests and Wilcoxon Signed-Rank Tests for Continuous Glucose Monitoring Metrics

Variable	Baseline	Postintervention	P	Effect size^a
% Data available	76.4 ± 29.6	81.3 ± 23.4	0.16	r = −0.23
Glycemic variability (%CV)	33.1 ± 6.7	31.6 ± 6.9	0.37	d = 0.21
Mean glucose	8.3 ± 1.9	8 ± 1.8	0.23	r = −0.2
GMI	51.7 ± 8.9	50.3 ± 8.6	0.23	r = −0.2
TAR: % time >250 mg/dL	7.2 ± 9.6	6.7 ± 10.6	0.25	r = −0.19
TAR: % time >180 mg/dL	25 ± 20.8	21.5 ± 21.2	0.09	r = −0.27
TIR: % daytime 70–140 mg/dL	50.9 ± 20.6	57.7 ± 22	0.01	d = −0.64
TIR: % time 70–140 mg/dL	49.7 ± 20.6	55.8 ± 21.1	0.04	d = −0.50
TIR: % daytime 70–180 mg/dL	72.6 ± 19.6	77.5 ± 19.8	0.03	r = −0.35
TIR: % time 70–180 mg/dL	71.4 ± 19.4	75.1 ± 20	0.10	r = 0.28
TBR: % time <70 mg/dL	3.6 ± 3.7	3.4 ± 4.6	0.60	r = −0.09
TBR: % time <54 mg/dL	0.6 ± 0.9	0.7 ± 1.4	0.93	r = −0.02

Bold text indicates statistical significance (P < .05) and moderate effect sizes (d ≥ .50; r ≥ .30).

Cohen's d used for paired t-tests (for normally distributed variables); r used for Wilcoxon signed-rank tests (for non-normally distributed variables)

GMI, glucose management index; TAR, time above range; TBR, time below range; TIR, time in range.

Diabetes distress

Changes in diabetes distress from baseline to postintervention are given in Table 5. At baseline, 40.9% of participants (n = 9) reported mild or no diabetes distress and 59.1% (n = 13) reported moderate or high diabetes distress. Postintervention, 54.5% (n = 12) reported mild or no diabetes distress and 45.5 (n = 10) reported moderate or high diabetes distress. There was a significant decrease in the total T1-DDS score from baseline (2.27 ± 0.66) to postintervention (2.02 ± 0.60); Z = −2.483, P = 0.013, r = −0.37 (moderate effect). This statistically significant decrease of −0.25 represented a larger magnitude than the minimal clinically important difference (MCID) of ±0.019 established for the T1-DDS,¹⁹ indicating a clinically significant decrease in diabetes distress. Significant decreases in distress were found in the Powerlessness, Management Distress, and Eating Distress subscales, all with moderate effects (r = −0.48, −0.45, and −0.40, respectively).

Table 5.

Wilcoxon Signed-Rank Tests for Diabetes Distress Scale for Adults with Type 1 Diabetes and Its Subscales

	Baseline	Postintervention	P	Effect size
T1-DDS total score	2.27 ± 0.66	2.02 ± 0.60	0.01	r = −0.37
Powerlessness	3.33 ± 1.18	2.78 ± 1.01	0.001	r = −0.48
Management distress	2.21 ± 1.26	1.72 ± 0.80	0.003	r = −0.45
Hypoglycemia distress	2.01 ± 1.04	1.96 ± 0.79	0.84	r = −0.03
Negative social perception distress	1.73 ± 0.72	1.77 ± 1.11	0.46	r = −0.11
Eating distress	2.77 ± 1.39	2.23 ± 0.99	0.008	r = −0.40
Physician distress	1.90 ± 1.18	1.68 ± 0.79	0.29	r = −0.16
Friend/family distress	1.84 ± 0.66	1.86 ± 0.72	0.58	r = −0.08

Bold text indicates P < .05 and moderate effect sizes (r = .30 − <.50).

T1-DDS, Diabetes Distress Scale for adults with type 1 diabetes.

Qualitative results

General reflections and changes made because of ONBOARD participation

Many participants appreciated individual sessions as a safe, welcoming space to share their diabetes challenges and receive validation. Participants found the program to be helpful and they appreciated incorporation of emotional aspects of diabetes. For example, one participant felt that talking about challenges with CGM and watching the videos helped her feel less “alone, because it is a very lonely disease” (39-year-old woman). Some noted that sessions helped clarify their own pros and cons of using CGM. In terms of changes participants made through their participation, they mentioned improved diabetes self-management, applying problem-solving strategies in their lives, customizing their alert settings, trying a new location for device placement, and increased willingness and openness to talking about diabetes with others in their lives. Some expressed leaving the program with increased interest in learning about diabetes technology. Some also described increased intentionality when viewing their CGM data and decreased frustration with diabetes management. Participants felt ONBOARD would benefit someone who is new to CGM, or who are “struggling more with their diabetes as a way to discover more tools and talk through it with someone” (23-year-old woman).

Session-specific feedback

For the most part, participants appreciated that each session had a different focus. Some session topics were more personally relevant than others for each participant. For the Physical session, those who do not experience issues with adhesive or physical placement of CGM did not find this session particularly helpful. For example, one participant shared, “The physical part of it has never been a barrier at all [for me]” (28-year-old man). However, others found it helpful to learn about different adhesive strategies and to discuss how to make devices more comfortable on their body. One participant noted, “I at least gained another infusion site, so that's cool” (22-year-old man).

For the Data session, all participants found it helpful. Some reported benefiting from viewing data (through Clarity or LibreView) in session and some appreciated enhanced understanding of trend arrows and “how to control a bunch of different alerts” and “turning off what's not useful and keeping on what is useful” (22-year-old man). Participants also found it helpful to talk about the ways they engaged with their CGM data to reduce data overload. For example, one participant shared, “It was helpful to me in thinking, don't overreact to the information [from my CGM]. Absorb it, take it how you need it, and then get on with your day” (32-year-old man). This session helped reinforce thinking of CGM data “more as a tool versus that I'm doing a bad job or something” (39-year-old woman).

For the Social session that incorporated a problem-solving activity, some participants found this activity to be extremely helpful. One participant stated, “We did actually sit down and solve an actual diabetes problem that I had been having” (28-year-old man). Some felt these strategies were not applicable or useful to them. As one participant shared, “I don't know how much I would use it in my day-to-day life” (21-year-old woman). The Social topic was more relevant for participants who expressed some discomfort in devices attracting unwanted attention from others than for participants who had a high level of comfort with talking about diabetes with others.

Finally, feedback on the Trust session depended on participants' existing level of trust of their CGM. Those with high trust in their device found the topic less relevant, “I've never had any trust issues with the devices so the [Trust session] was the least applicable” (28-year-old man); however, participants with low trust found this session helpful in clarifying the process of building trust and having back-up plans when the CGM is not accurate. This session also helped some participants reflect upon the benefits that come from trusting CGM data. One participant shared, “10% of the time it's wrong or not reading correctly. I'm clearly going to keep wearing it and trusting it for the other 90% of the time” (24-year-old woman).

Video feedback

Most participants found the videos to be “a helpful launching point for our conversations” (30-year-old woman). Many appreciated the video content because, as one participant stated, “Pretty much everything everyone said, I can relate to in some way” (33-year-old woman). Some participants gained new insights about their own relationship with diabetes devices from watching the videos, and others reported that the videos helped them feel less alone in living with T1D. Many participants liked the diversity of perspectives and gender ratio of speakers in the video. However, some suggested the need for representation of younger CGM users as well as CGM users more diverse racial, ethnic, and socioeconomic backgrounds. In addition, one participant felt she would have wanted in-person interaction with others living with T1D instead of watching videos.

Suggestions for improvement

Participants suggested additional content areas including advocating for oneself with providers to gain access to diabetes technology and resources to help with navigating cost and insurance coverage. Some suggested providing additional education on common pitfalls associated with starting to use CGM (e.g., compression lows, bleeding with insertion, and signal loss). Some participants expressed that they would have liked more formal, active “challenges” to work on between sessions. For example, one participant suggested a challenge of “Do you want to try wearing [your CGM] on your arm for a week?” (24-year-old woman). Finally, some asked for the emotional and psychological aspects of living with T1D to be more explicit components of ONBOARD.

Discussion

Consistent CGM use is associated with glycemic benefits for people living with T1D, but little is known about what guidance and support adults with T1D need, beyond standard device initiation education, to optimize their device use. This study was a pilot of an individual program delivered virtually and aimed to support adults with T1D in their early use of CGM by providing them with necessary tools, education, and problem-solving skills to sustain their use of the device and experience benefits of the technology for their overall diabetes management.

Results of this pilot demonstrated initial feasibility, acceptability, and efficacy of the ONBOARD program that demonstrated initial evidence of clinically significant benefits to diabetes distress and daytime spent in target glucose range. At the time patients entered this study they were already in very good glycemic control with a mean sensor glucose of 149 mg/dL equivalent to a glucose management index of 6.9%.³³ At the end of the study, although there was no significant change in mean glucose on average, there was a significant increase in time in range during daytime (from 72.6% to 77.5% time between 70 and 180 mg/dL). In addition, participants experienced a clinically significant reduction in diabetes distress from baseline to the end of the intervention. The number of participants endorsing moderate/high distress decreased from n = 13 to n = 10 from baseline to follow-up, and the decrease in total T1-DDS score was larger than the MCID established for the scale. The largest decreases in distress were in the Powerlessness, Management Distress, and Eating Distress subscales. More research is needed to understand the factors contributing to this decrease as well as longitudinal research to understand whether these decreases endure over time. In addition, participants who completed ONBOARD on average reported a clinically meaningful decrease in diabetes distress as well as an additional hour of time spent in their target glucose range per day. This preliminary evidence points to the potential benefits of tailored behavioral support focused on diabetes device adoption for diabetes management and the emotional side of living with diabetes.

Furthermore, qualitative data from focus group feedback suggest that participants accepted this type of device-focused intervention. Although they noted that not all the topic areas applied to them directly—particularly if they had no issues wearing a device on their body or having other people notice their device—participants appreciated these topics and found personal relevance in at least some of the content. Overall, participants shared that they appreciated having a safe and welcoming space to talk about diabetes-related issues they may not normally have an opportunity to talk about in their daily life. The first-person videos elicited mostly positive feedback from participants who described relating to the content and finding validation in listening to people who have experienced similar things with diabetes devices and CGM specifically.

Of promise is the finding that 100% of participants who attended their first ONBOARD session completed all four sessions. The online format and flexibility with scheduling sessions because of the one-on-one nature of sessions were likely factors contributing to attendance. Participants appreciated being able to attend ONBOARD using virtual visits. ONBOARD was initially as a telehealth-based program for greater convenience and accessibility for adults who have competing demands of work, school, and/or family duties. The pilot sessions were conducted between January and July 2020 and were able to continue during the pandemic; participants accepted this format both before and during the context of COVID-19. Indeed, use of virtual medical visits has increased dramatically during the pandemic³⁴ and satisfaction with virtual services is high.³⁵ In diabetes care specifically, the availability of CGM data and ease of sharing with a health care provider has enabled the delivery of remote care and may signal a major shift in the future of T1D care,³⁶ provided that increasing numbers of people with T1D have affordable access to this technology.³⁷ Finally, that participants attended four sessions with the same study interventionist enabled rapport-building that may have further promoted session attendance.

Participants shared suggestions for improving upon ONBOARD that have been used to refine this program. Some participants expressed a desire to see a greater level of racial, ethnic, and age diversity represented in the videos, as the adults who appeared in the video were White and in their 30s. Video narratives will be expanded to include a greater representation of voices of adults with T1D. In addition, some participants advocated for including “challenges” or other experiential assignments to try between sessions. In refining the intervention, possible between-session “homework” could include trying a different physical device placement on the body; experimenting with different alert thresholds; sharing data or other possibilities to focus on developing problem-solving skills and/or practical skills with using and navigating the technology. Given that not all topics were personally relevant for every participant, it may be beneficial for ONBOARD sessions to allow for greater flexibility and tailoring of content to the particular barriers a participant is experiencing. One method for tailoring ONBOARD content may be to administer an initial questionnaire (i.e., a needs assessment) at the beginning to prioritize concerns and barriers. This approach may also allow for shortening the intervention duration to increase feasibility, acceptability, and cost-effectiveness.

In addition, the pilot of ONBOARD was delivered by a doctoral-level clinical psychologist. Some participants expressed interest in receiving CGM support directly from peers (e.g., a “buddy”), possibly with group support as an option to complement individual support. When thinking about future development, refinement, and translation of ONBOARD to clinical settings, it will be important to consider whether a master's level clinician, an expert peer, or a diabetes educator could deliver this intervention. Finally, ∼40% of participants in this study had used their CGM for >6 months, whereas the remainder had used it for <6 months. Given the sample size, we did not explicitly assess whether length of time using CGM had an influence on likelihood of benefitting from ONBOARD. However, in focus groups some participants mentioned that they felt the program would be best for those first starting out using CGM and others suggested that those experiencing burnout could benefit. Future research should explore ideal timing of this support in addition to further tailoring material to individual needs and barriers.

One limitation of this pilot study is that eligible participants were those who had already obtained CGM supplies independent of the research study that limits the generalizability of this intervention to those who either cannot obtain CGM through their insurance or have not yet decided to try to use CGM. Study participants were those willing and motivated to take part in helping to provide feedback and shape a CGM support pilot program, so we cannot draw conclusions from these results about how ONBOARD would be received by a general clinic patient population. Additional education and resources may be beneficial to participants who are seeking to obtain CGM and/or are just starting to use this technology. Furthermore, participants on average entered this study already meeting ADA-recommended glycemic targets, which may limit generalizability to those with elevated HbA1c. Future examination of ONBOARD in a population of adults with elevated HbA1c is needed that may have the potential for greater benefit.

Additional limitations of this pilot study are that it involved a small sample for assessing initial acceptability and preliminary efficacy and that follow-up data were collected soon after completion of the four-session intervention. Thus, it is unknown if preliminary evidence of decreases in diabetes distress and increases spent in target glucose range can be sustained over time, and for how long, following the end of the program. Further research should examine the impact of ONBOARD with a larger, fully powered sample followed over a longer period and should track rates of CGM usage in addition to glycemic outcomes. In addition, many participants in this study identified as non-Hispanic (90.9%) and White (86.4%). Follow-up studies of ONBOARD should make efforts to recruit a more diverse participant sample. Finally, the design of this study involved assessing a single group pre- and postintervention. A follow-up study that compares participants receiving ONBOARD when starting CGM with a control group of participants who initiate CGM without the ONBOARD intervention would address potential threats to internal validity. This approach would also enable the ability to evaluate the potential added role of ONBOARD beyond standard CGM initiation.

If a larger scale examination of the individual-based ONBOARD intervention proves effective in sustaining CGM use and its associated glycemic and quality-of-life benefits, future efforts could be directed toward examining more efficient, cost-effective models of delivering similar content. In its current format, ONBOARD consists of 4 h of content delivered by a doctoral-level psychologist across four separate hour-long sessions. Thus, ONBOARD is relatively time and resource intensive. Alternatives to explore in future research may be a group format, peer-led sessions, asynchronous support (e.g., via an app or website), and/or condensing content into fewer sessions. In addition, it may be important to determine who most benefits from this type of intervention (vs. who would successfully adopt CGM without needing additional structured support) and tailoring the content and amount of time spent per patient depending on need and potential to benefit.

To conclude, ONBOARD, a multicomponent behavioral intervention designed to support adults with T1D in their early use of CGM, was shown to be feasible and acceptable to participants. We observed clinically meaningful decreases in diabetes distress as well as increases in time in target glucose range at the end of the program, providing preliminary evidence of ONBOARD's efficacy. ONBOARD addresses a current gap in health services and integrates empirically supported components in a novel manner to improve adoption of CGM. Although innovation is the goal and incorporated in the novel integration of behavior change components, addressing a gap in services is viewed as equally significant. Novel aspects of the ONBOARD program are (1) the incorporation of relatable, real-life stories from CGM users as a teaching tool into the curriculum and (2) providing a supportive space for adults with T1D as they adjust to the incorporation of new technology into their daily lives with diabetes.

Footnotes

Disclaimer

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Acknowledgments

The authors thank the participants who took part in this pilot study and provided valuable input to help shape and refine it.

Author Disclosure Statement

B.A.B. reports receiving grant support and advisory board fees from Medtronic Diabetes and ConvaTec, grant support and presentation fees from Insulet, advisory board fees from NovoNordisk and Profusa, grant support from Eli Lilly, grant support and equipment from Dexcom, and holding patent 61197230 on a hypoglycemia prediction algorithm. D.M.M. has had research support from the NIH, JDRF, NSF, and the Helmsley Charitable Trust and his institution has had research support from Medtronic Diabetes, Dexcom, Insulet, Bigfoot Biomedical, Tandem, and Roche. D.M.M. has also consulted for Abbott, the Helmsley Charitable Trust, Sanofi, Novo Nordisk, Eli Lilly, Medtronic, and Insulet. K.K.H. has received consulting fees from Lifescan Diabetes Institute and MedIQ and an investigator-initiated grant from Dexcom, Inc., M.L.T., J.N., S.J.H., M.B., D.H., and S.M. have no conflicts to disclose.

Funding Information

Research reported in this publication was supported by the National Institute Of Diabetes And Digestive And Kidney Diseases of the National Institutes of Health under Award Number K23DK119470 (MLT) and through a Pilot and Feasibility grant from the Stanford Diabetes Research Center (P30DK116074). This publication was also supported by the Stanford Maternal and Child Health Research Institute through an Instructor K Support Award to M.L.T.

Supplementary Material

Supplementary Data

References

Foster

, Beck

, Miller

, et al.: State of type 1 diabetes management and outcomes from the T1D exchange in 2016–2018. Diabetes Technol Ther, 2019; 21:66–72.

American Diabetes Association: 6. Glycemic targets: standards of medical care in diabetes—2021. Diabetes Care, 2021; 44(Supplement 1):S73–S84.

Tamborlane

, Beck

, Bode

, et al. Continuous glucose monitoring and intensive treatment of type 1 diabetes. N Engl J Med, 2008; 359:1464–1476.

Polonsky

, Hessler

, Ruedy

, Beck

: The impact of continuous glucose monitoring on markers of quality of life in adults with type 1 diabetes: further findings from the DIAMOND randomized clinical trial. Diabetes Care, 2017; 40:736–741.

Hovorka

, Kumareswaran

, Harris

, et al.: Overnight closed loop insulin delivery (artificial pancreas) in adults with type 1 diabetes: crossover randomised controlled studies. BMJ, 2011; 342:d1855.

Luijf

, DeVries

, Zwinderman

, et al.: Day and night closed-loop control in adults with type 1 diabetes. Diabetes Care, 2013; 36:3882–3887.

Iturralde

, Tanenbaum

, Hanes

, et al.: Expectations and attitudes of individuals with type 1 diabetes after using a hybrid closed loop system. Diabetes Educ, 2017; 43:223–232.

Tanenbaum

, Iturralde

, Hanes

, et al.: Trust in closed loop systems: a qualitative study of perspectives of experienced system users. J Health Psychol, 2020; 25:429–438.

Tanenbaum

, Hanes

, Miller

, et al.: Diabetes device use in adults with type 1 diabetes: barriers to uptake and potential intervention targets. Diabetes Care, 2017; 40:181–187.

10.

Addala

, Maahs

, Scheinker

, et al.: Uninterrupted continuous glucose monitoring access is associated with a decrease in HbA1c in youth with type 1 diabetes and public insurance. Pediatr Diabetes, 2020; 21:1301–1309.

11.

Prahalad

, Addala

, Buckingham

, et al.: Sustained continuous glucose monitor use in low-income youth with type 1 diabetes following insurance coverage supports expansion of continuous glucose monitor coverage for all. Diabetes Technol Ther, 2018; 20:632–634.

12.

American Diabetes Association: 7. Diabetes technology: standards of medical care in diabetes—2021. Diabetes Care, 2021; 44(Supplement 1):S85–S99.

13.

, Varughese

, Li

, Kushner

: Healthcare resource waste associated with patient nonadherence and early discontinuation of traditional continuous glucose monitoring in real-world settings: a multicountry analysis. Diabetes Technol Ther, 2018; 20:420–427.

14.

Smith

, Albanese-O'Neill

, Macieira

, et al.: Human factors associated with continuous glucose monitor use in patients with diabetes: a systematic review. Diabetes Technol Ther, 2019; 21:589–601.

15.

Murata

, Kuroda

, Matsuhisa

, et al.: Predictive factors of the adherence to real-time continuous glucose monitoring sensors: a prospective observational study (PARCS STUDY). J Diabetes Sci Technol, 2020:1932296820939204.

16.

Sørgård

, Iversen

, Mårtensson

: Continuous glucose monitoring in adults with type 1 diabetes: a balance between benefits and barriers: a critical incident study. J Clin Nurs, 2019; 28:3318–3329.

17.

Abdoli

, Hessler

, Vora

, et al.: Descriptions of diabetes burnout from individuals with Type 1 diabetes: an analysis of YouTube videos. Diabetic Med, 2020; 37:1344–1351.

18.

Abdoli

, Jones

, Vora

, Stuckey

: Improving diabetes care: should we reconceptualize diabetes burnout?. Diabetes Educ, 2019; 45:214–224.

19.

Fisher

, Hessler

, Polonsky

, et al.: Diabetes distress in adults with type 1 diabetes: prevalence, incidence and change over time. J Diabetes Complications, 2016; 30:1123–1128.

20.

Hermanns

, Ehrmann

, Schipfer

, et al.: The impact of a structured education and treatment programme (FLASH) for people with diabetes using a flash sensor-based glucose monitoring system: results of a randomized controlled trial. Diabetes Res Clin Pract, 2019; 150:111–121.

21.

Pintus

, Ng

: FreeStyle Libre flash glucose monitoring improves patient quality of life measures in children with type 1 diabetes mellitus (T1DM) with appropriate provision of education and support by healthcare professionals. Diabetes Metab Syndr, 2019; 13:2923–2926.

22.

Schlüter

, Freckmann

, Heinemann

, et al.: Evaluation of the SPECTRUM training programme for real-time continuous glucose monitoring: a real-world multicentre prospective study in 120 adults with type 1 diabetes. Diabetic Med, 2021; 38:e14467.

23.

Heinemann

, Klonoff

: An Opportunity to Increase the Benefit of CGM Usage: the Need to Train the Patients

Adequately

. Los Angeles, CA: SAGE Publications Sage CA, 2020.

24.

Venkatesh

, Davis

: A theoretical extension of the technology acceptance model: four longitudinal field studies. Manage Sci, 2000; 46:186–204.

25.

Lansing

, Stanger

, Budney

, et al.: Pilot study of a web-delivered multicomponent intervention for rural teens with poorly controlled type 1 diabetes. J Diabetes Res, 2016; 2016:1–8.

26.

Stanger

, Ryan

, Delhey

, et al.: A multicomponent motivational intervention to improve adherence among adolescents with poorly controlled type 1 diabetes: a pilot study. J Pediatr Psychol, 2013; 38:629–637.

27.

Mulvaney

, Rothman

, Wallston

, et al.: An internet-based program to improve self-management in adolescents with type 1 diabetes. Diabetes Care, 2010; 33:602–604.

28.

Rees

, O'Hare

, Saeed

, et al.: Problem-solving therapy for adults with diabetic retinopathy and diabetes-specific distress: a pilot randomized controlled trial. BMJ Open Diabetes Res Care, 2017; 5:e000307.

29.

Channon

, Huws-Thomas

, Rollnick

, et al.: A multicenter randomized controlled trial of motivational interviewing in teenagers with diabetes. Diabetes Care, 2007; 30:1390–1395.

30.

Schweier

, Romppel

, Richter

, et al.: A web-based peer-modeling intervention aimed at lifestyle changes in patients with coronary heart disease and chronic back pain: sequential controlled trial. J Med Internet Res, 2014; 16:e177.

31.

Fisher

, Polonsky

, Hessler

, et al.: Understanding the sources of diabetes distress in adults with type 1 diabetes. J Diabetes Complications, 2015; 29:572–577.

32.

Pope

, Mays

: Qualitative Research in Health

Care

, 2nd ed. London: BMJ Books, 2000.

33.

Bergenstal

, Beck

, Close

, et al.: Glucose management indicator (GMI): a new term for estimating A1C from continuous glucose monitoring. Diabetes Care, 2018; 41:2275–2280.

34.

Wosik

, Fudim

, Cameron

, et al.: Telehealth transformation: COVID-19 and the rise of virtual care. J Am Med Inform Assoc, 2020; 27:957–962.

35.

Olayiwola

, Magaña

, Harmon

, et al.: Telehealth as a Bright Spot of the COVID-19 Pandemic: recommendations From the Virtual Frontlines (“ Frontweb”). JMIR Public Health Surveill, 2020; 6:e19045.

36.

Ahn

: The COVID-19 pandemic: a “tech”-tonic shift toward virtual diabetes care. J Diabetes Sci Technol, 2020; 14:708–709.

37.

Kerr

, Warshaw

: Clouds and silver linings. COVID-19 pandemic is an opportune moment to democratize diabetes care through telehealth. J Diabetes Sci Technol, 2020:14:1107–1110.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.21 MB