Background and Aims

While real-time continuous glucose monitoring (RT-CGM) has been associated with greater glycemic improvement than traditional blood glucose monitoring (BGM) in T2D, little is known about the differential impact on glucose monitoring satisfaction over time in this population.

Methods

The MOBILE randomized controlled trial documented significantly greater glycemic benefits in RT-CGM (Dexcom™ G6 CGM System®) T2D participants (n=108) than BGM T2D participants (n=57). To compare change in satisfaction over time between RT-CGM and BGM users, participants completed three subscales of the Glucose Monitoring Satisfaction Survey (T2-GMSS): Openness, Emotional Burden and Behavioral Burden, at baseline and 8 months.

Results

Openness rose over time in both groups (ps<.01), though improvement was significantly greater in the RT-CGM (3.06 ± 0.78 to 3.83 ± 0.71) vs. BGM group (3.01 ± 0.78 to 3.37 ± 0.71), resulting in a moderate-to-large between-group effect size at 8 months (d=.67, p<.001). Emotional and Behavioral Burden fell in both groups (ps<.01), with no significant between-group differences. Change over time in all three subscales was associated in the expected direction with HbA1c improvements (rs=0.39-0.47, ps<.001).

Conclusions

Both study arms evidenced significant gains in all three aspects of glucose monitoring satisfaction. However, RT-CGM users reported significantly greater improvement than BGM users along one of those dimensions, Openness, which represents how the monitoring system may unblock rather than constrain one’s lifestyle. This is consistent with previous research illustrating how RT-CGM can enhance one’s comfort and confidence with trying out new foods and activities, thereby allowing individuals to feel less restricted.

Background and Aims

Despite the centennial of insulin’s discovery by Frederick Banting, Charles Best, and colleagues at the University of Toronto in 1921, half of the people living with diabetes worldwide cannot access or afford it. The aim of this study is to present contemporary data concerning out-of-pocket expenses (OoPEs), the extent of insulin and supply underuse, and the degree of financial coverage people with type 1 diabetes (T1D) are experiencing across the world.

Methods

A web-based, cross-sectional survey was conducted from August to December 2020. The analysis included comparisons between responses from countries with no, partial, and full healthcare coverage.

Results

1,066 participants from 64 countries took part in the study. ~25% of respondents reported having underused insulin at least once within the last year due to perceived cost. A significant correlation was observed between OoPEs and reported household income for respondents with partial healthcare coverage. 63.2% of participants reported disruption of insulin supplies and 25.3% reported an increase of prices related to the COVID-19 pandemic.

Conclusions

This study confirms previous reports of ~25% of people in the United States with T1D using less insulin and/or fewer supplies at least once in the last year due to cost, a trend associated with the extent of healthcare coverage. Similar trends were observed in some middle/low income countries. Moreover, patients reported an increase in insulin prices and disruption of supplies during the COVID-19 pandemic. This study highlights the importance of self-reported OoPEs and its association with underuse/rationing of insulin.

Background and Aims

The EU project “FORGETDIABETES” (supported by H2020-FETPROACT Project FORGETDIABETES, n.951933) aims at developing a fully implantable system for automated insulin delivery (AID). It will consist of an intraperitoneal pump and an implantable CGM sensor. Insulin will be refilled via an insulin pill and the pump will be charged wirelessly. It aims at providing fully automated diabetes management with minimal user interaction. To explore human factors of the bionic invisible pancreas (BIP), we performed a qualitative study.

Methods

Semi-structured interviews were conducted with people with type 1 diabetes. All participants provided written informed consent and were shown a video illustrating the components and functionality of the BIP. Participants were asked about aspects of life that could benefit from the BIP and aspects of life that could become more complicated.

Results

Interviews from 38 people with type 1 diabetes were analysed. The most frequently mentioned aspect that could benefit from using the BIP was psychological with expected positive effects on quality of life, reduced diabetes distress, and less thinking about diabetes. Also, better glycaemic control and improvements regarding long-term complications (termed “Diabetes” in Figure 1) were also frequently mentioned as aspects that could benefit. Aspects that might become more complicated related mostly to technical aspects such as the implantation procedure, taking an insulin pill, and being dependent on the technology. Loss of control was also named as a big issue (Figure 2).

Conclusions

Psychological aspects such as improvements in quality of life but also loss of control are important human factors for future AID systems.

Background and Aims

The utilization of CSII in type 1 diabetes (T1D) is associated with increased risk of diabetic ketoacidosis (DKA). The rationale behind using hybrid modality, long-acting insulin for basal coverage and CSII for boluses, is the prevention of insulin delivery failure and subsequent hyperglycemia and DKA.

To explore the hybrid treatment modality in clinical practice in youth with T1D.

Methods

Multicenter, observational study of youth with T1D who initiated hybrid modality between 2013 and 2020. Extracted from the medical records were data on sociodemographic characteristics, reason for hybrid treatment initiation, glycemic metrics, HbA1c and frequency of DKA episodes, collected at initiation, after 6 months, and at last visit.

Results

Fifty-five patients (52.7% males) were treated with hybrid therapy, median [IQR] age at initiation 14.5years [12.4, 17.3], HbA1c 9.2% [8.2, 10.2], mean glucose levels 221mg/dL [181, 226] and treatment duration 18 months [12, 47]. Hybrid treatment was initiated due to fear of sustained hyperglycemia in 41.8%, DKA episodes in 30.8%, refusal to use CSII continuously in 14.6%. HbA1c did not significantly change throughout follow-up (P=0.262 and P=0.195). Mean glucose levels decreased after 6 months (P=0.034), and remained stable thereafter (P=0.274). Frequency of DKA decreased after 6 months and at last visit as compared with 6 months preceding initiation of hybrid therapy (number of events/number of patients: 4/4 and 10/10 vs. 24/14, P=0.002 and P=0.031, respectively).

Conclusions

Our findings suggest that this hybrid therapy is a feasible option in the management of youth with T1D, which may reduce the risk of DKA episodes.

Background and Aims

There is variability in the experience of symptoms of hypoglycaemia, within- and between- individuals with diabetes. We report preliminary data on the prevalence of hypoglycaemic symptoms across glucose ranges by hypoglycaemia awareness status using our novel smartphone HypoMETRICS app.

Methods

Participants with type 1 diabetes (T1D) used the app to report all episodes of hypoglycaemia experienced over the 10-week study, recording glucose level and grading symptoms. The symptoms included sweating, heart palpitations, shaking, hunger, confusion, headache, difficulties in speaking and difficulties in movement and coordination. Heat maps of symptoms were generated based on their prevalence at each glucose level by awareness status, assessed by Gold Score, using R.

Results

2804 hypoglycaemic episodes were reported by 97 participants, mean (SD) age was 46.0 (14.6) years and diabetes duration was 22.9 (15.0) years. 72.2% used Flash glucose monitoring. 75 participants had normal awareness of hypoglycaemia and 22 had impaired awareness of hypoglycaemia (IAH). The most common symptoms were hunger (49.9%), shaking (45.1%), sweating (34.2%) and heart palpitations (28.1%). As shown in Figure 1, episodes with lower glucose were associated with more symptoms and a lower percentage of episodes in IAH were symptomatic in each glucose category.

Conclusions

Heat maps generated from the HypoMETRICS app data provide a visual, quantitative representation of subjective experiences of hypoglycaemia. As the heat maps can be generated for individuals as well as groups, the app offers a novel tool for objective evaluation of hypoglycaemia awareness in T1D, clinically as well as in research.

Background and Aims

New technologies aim to aid diabetes management with personalized insights. However patients bring existing beliefs, which may conflict with new information, and which they may not readily share with their doctors. To understand patient perceptions, we conducted large-scale online studies examining lay beliefs about patient information sharing.

Methods

Across four experiments, we recruited U.S residents aged 18-64, who do not work in healthcare (N=1125 in analysis). Participants saw vignettes about a person who was recently diagnosed with T2D and shares a health belief with their new doctor. The beliefs varied in reasonableness (true, reasonable, unreasonable, conspiracy theory) and centrality to diabetes management (central, peripheral), and the experiments varied characteristics of the patient (e.g., race, gender). Participants rated three key aspects of a medical encounter: doctor perception of patient, patient efficacy, and patient trust.

Results

Across all experiments, the less reasonable a patient’s beliefs were, the more participants thought they would be negatively perceived by their doctors, less effective at managing their diabetes, and less trusting of doctors. As shown in the figure, patients were also penalized more for incorrect beliefs central to diabetes. Interestingly, there was no effect of race or gender.

Conclusions

Providing information to patients without understanding their existing beliefs may be ineffective. However we find that people perceive significant possible negative impacts of information sharing. Our follow-up studies with primary care doctors suggest these fears may be well-founded. Thus, there is an urgent need for clinicians to understand and address these perceptions.

Background and Aims

This pivotal parent study (NCT03563313) showed an overall increase in time in target range (TIR) of 11% [95%CI 9 to 14; p<0.001] favorable to AID use compared to sensor-augmented pump (SAP) therapy, which contributed to the FDA’s authorization of its commercialization. However, such technology is utilized less among Black, Indigenous, and People of Color (BIPOC) and lower socioeconomic status (SES) individuals. This study investigated improvements in glycemic outcomes for participants of different racial and SES groups while using AID vs SAP in this trial

Methods

Case summaries of participants differing in income, education, and race/ethnicity were analyzed to determine each group’s percentages of TIR (70-180mg/dL), time in hyperglycemia (>180mg/dL), and time in hypoglycemia (<70mg/dL) while using AID vs SAP therapy.

Results

The majority of participants in the trial were White and from higher income and education groups. TIR percentages using SAP compared to AID increased 18.6% for individuals making <$50,000 (7.1% of participants), compared to a 9.7% and 9.6% increase for those making $50,000-$100,000 (29.8%) and >$100,000 (61.7%), respectively. Those without (10.6% of participants) and with a bachelor’s degree (57.6%) spent 26.7% and 13.4% less time in hyperglycemia using AID, respectively. BIPOC individuals (10.6% of participants) spent 51.8% less time in hypoglycemia, whereas White individuals (87.6%) spent 43.9% less time <70 mg/dl. No group differences were significant.

Conclusions

Although BIPOC and lower SES populations were underrepresented in this trial, data indicates that these groups achieved similar glycemic outcomes compared to higher income/education and Caucasian participants showing a potential equalizing benefit from this type of technology.

Background and Aims

Health coaching has shown to reduce HbA1c and blood glucose (BG) excursions, and lifestyle recommendations are more accepted during teachable moments. This study investigated the impact of Diabetes Response Specialists’ (DRS) real-time feedback during BG-triggered alerts on estimated A1c (eA1c) improvement.

Methods

This retrospective cohort study used data from participants enrolled in a remote diabetes program (RDP) for a minimum of 12-months, which offered education and self-management tools through mobile technology. Improvement in eA1c was defined as a reduction ≥0.3%. Multivariable logistic regression was used to analyze the associations of number of successful DRS interactions and scheduled certified diabetes care and education specialists (CDCES) coaching on eA1c improvement at 12-months adjusting for participant demographics, characteristics, and program utilization.

Results

Participants (N=167,095) were 47% women with mean age of 55 years old [SD 11.7]. A majority were individuals with type 2 diabetes mellitus on oral medications only (59%) with 3% utilizing CDCES coaching. DRS interaction was utilized by 33% of participants who received a mean of 9.5 successful DRS contacts. The number of successful DRS contact was significantly associated with eA1c improvement at 12-months with odds ratio of 1.63 (95% CI: 1.51, 1.84). The number of scheduled CDCES coaching sessions was not associated with eA1c improvement 1.03 (95% CI: 0.94, 1.12).

Conclusions

DRS interactions during BG excursions were associated with improved glycemic control in a RDP and may reach members not served by scheduled CDCES coaching.

Background and Aims

Automated bolus calculators (ABCs) help PwD with calculating insulin doses. Clinical studies have consistently shown the clinical benefits of ABCs. However, ABCs differ in terms of their design and mathematical approaches but the real-world impact of such differences was never evaluated. We evaluate the real-world impact of a significant change introduced to the algorithm of an over-the-counter ABC that is integrated into a mobile health application.

Methods

We use a retrospective pretest-posttest design to evaluate changes in glycemic control and patient reported outcomes (PROs) after 3 months of using an updated ABC. We test for clinical relevant changes in mean glucose, coefficient of variability and hypoglycemic episodes calculated from self-monitored blood glucose data. Diabetes Treatment Satisfaction Questionnaires (DTSQ) are sent at device initiation (baseline) and at 3 month follow-up.

Results

Improvements in glucose control and PROs are expected as a result of the updated algorithm. At the moment of writing, baseline surveys were sent to 6872, received by 4414 (65%) and completed by 720 (17%) users out of which 50% are expected to complete the follow-up survey. Analysis is ongoing and preliminary results are expected by the end of the year.

Conclusions

We evaluate the real-world impact of a significant design change introduced to the algorithm of an over-the-counter ABC on glycemic control and treatment satisfaction. High exclusion rates are accounted for by the analysis of a comparably large population. Preliminary results are expected by the end of the year.

Background and Aims

A large subset of PWD assume that physical activity alone can bring glycemic control irrespective of the stage of diabetes. During Covid, there was an increased emphasis on physical activity in PWD. People with uncontrolled diabetes were avoiding physical visits to hospitals due to the fear of contracting the disease. In this scenario, we analysed the effects of physical activity on glycemic control as measured by CGM in terms of TIR.

Methods

We analysed the EMR of T2D patients who performed CGM atleast once from March 2021 - June 2021 after the lockdown period and extracted data on exercise (type, frequency and duration of exercise), BMI and TIR. Out of 603 T2D, 332 (those who performed at least one kind of physical activity) were categorised into the study group (SG) and 271 (those without any physical activity) into the control group (CG).

Results

TIR was compared between SG and CG using analysis of covariance model with TIR as the dependent variable, treatment as a fixed effect and BMI as a cofactor. There was no significant difference (p=0.7305) in TIR between the groups. When taken together, 16.3% achieved a TIR >70% and 6.5% achieved a TIR >90% irrespective of physical activity. Percentage of patients in various TIR categories is shown(Fig.1)

Conclusions

Physical activity alone as an independent parameter may not have a significant role in improving TIR. The findings emphasize the fact that physical acvity should be combined with medical nutrition therapy and therapeutic interventions for better optimal outcomes in the management of diabetes.

Background and Aims

Real-time CGM systems have configurable alerts for hypoglycemia and hyperglycemia. We examined alert activation frequencies and their associations with default settings for novice users of the Dexcom G6 CGM System.

Methods

A recent G6 software update changed the default (low, high) alert thresholds from (80, 200) to (70, 250) mg/dL, resulting in a wider range of non-alerting glucose levels for later adopters. The Dexcom customer database provided two anonymized cohorts of US-based users who installed the G6 app and began uploading data from iOS devices before or after the index date of 11-MAR-2020. Glycemia and frequency of alert activations were monitored for the first 180 days of use.

Results

The (80, 200) and (70, 250) cohorts (n=10,306 and n=9,154, respectively) were well-matched with respect to baseline attributes. Cohorts had similar TIR, TAR, TBR, and mean glucose levels (p>0.1). Low alerts were less frequent than high alerts, both alert frequencies decreased with time, and the (80, 200) cohort experienced consistently more alerts than the (70, 250) cohort (Figure). Users in the (70, 250) cohort were less likely to adjust their settings than users in the (80, 200) cohort (p<0.001).

Conclusions

Onboarding novice CGM users with the wider range of non-alerting glucose values implemented in the most recent software resulted in fewer alerts for hypoglycemia and hyperglycemia, with no detriment to glycemic control.