Psychosocial Impact of the Insulin-Only iLet Bionic Pancreas for Adults,Youth,and Caregivers of Youth with Type 1 Diabetes

Abstract

Objective:

To evaluate the psychosocial impact and user experience for the insulin-only configuration of iLet bionic pancreas (BP) in persons 6–83 years years of age with type 1 diabetes.

Research Design and Methods:

In this multicenter, randomized controlled, 13-week trial, 275 adults (221 randomly assigned to the BP group and 54 to the standard of care [SC] group) and 165 youth and their caregivers (112 randomly assigned to the BP group and 53 to the SC group) completed psychosocial questionnaires at baseline, mid-study, and the end of the trial.

Results:

In all age groups, most participants would recommend using the BP, including those with previous experience using automated insulin delivery devices. Similarly, the vast majority of participants reported a high level of perceived benefits and a low number of perceived burdens. Adult participants reported significant decreases in the fear of hypoglycemia and in diabetes-specific emotional distress, as well as improvements in their perceived well-being.

Conclusion:

Findings demonstrate acceptability, reduced burden, and positive psychosocial outcomes for adults. Children and teenagers also report high acceptability and reduced burden, but less clear improvements in psychosocial outcomes.

Clinical Trial Registration Number: NCT04200313.

Introduction

Automated insulin delivery systems (AID systems) offer the possibility of improving glycemic outcomes, including reduced A1c, increased time in range, and reduced hypoglycemia.¹ These systems also offer the possibility of reducing the emotional burden of daily management demands of diabetes. In clinical trials of hybrid closed-loop (HCL) systems, patient-reported outcomes show a mixed experience for the user, with some showing improvements in key psychosocial constructs, while others show no difference.

Recent studies assessing the psychosocial impact of AID systems (the Medtronic Minimed 670G, the Tandem Control IQ and the Bionic Pancreas [BP]) in children, adolescents, caregivers, and adults have yielded mixed results. With respect to diabetes-specific emotional distress, improvements in distress were reported in some trials,^2

–6 while no change in diabetes distress was noted in others.^7

–13 Reductions in fear of hypoglycemia and improved confidence in managing hypoglycemia were reported in some trials,^3,9,13 while no change was reported in others.^{2

–5,7,8,11,14}

Similarly, some found improvements in quality of life and general well-being,^7,12,15 yet others did not.^6,11,14 The impact of AID systems on sleep quality was assessed in some trials, with mixed findings as well, with some finding improvements^3,14,15 and others finding no improvement.^4,10
–12 For studies assessing changes in treatment satisfaction or attitudes toward diabetes technology, some found improvements,^{2,6
–8,14,15} while others did not.^12,16 Perceived benefits of AID technology were found to outweigh perceived burdens.^5,13 One study reported an increase in positive expectancies for AID systems among caregivers and teenagers, but not among adults,⁵ while another found increased positive expectancies for children and teenagers.¹¹

Assessing the psychosocial impact of AID systems is necessary as the success of this important technology is predicated upon whether the user finds more benefits than burdens and is therefore willing to not only try the device but also continue using it. These analyses describe the psychosocial outcomes from a multicenter randomized trial of people with type 1 diabetes (T1D), 6–83 years of age, designed to assess the efficacy and safety of the insulin-only iLet BP System (BP; Beta Bionics).¹⁷

The BP is a unique insulin delivery system, as it is only initialized using the person's body weight. The system automates insulin delivery without a warm-up period. Individuals do not count carbohydrates, but “announce” meals by determining if the meal they are about to have is “usual,” “more” or “less” than what they would typically eat for that meal type. The BP continually adapts to the person's insulin needs and autonomously determines basal, mealtime, and correction doses. Neither the user nor the medical team can modify the algorithm, which continuously adapts to an individual's insulin needs.¹⁷ All the insulin doses are determined entirely by the BP.¹⁷ These characteristics differentiate the BP from all other commercially available AID systems, which are known as HCL systems because the systems automate some insulin delivery, while the user determines the remainder.

Given the uniqueness of the BP, it is vital to assess the psychosocial impact and user experience when using the system to inform clinical recommendations related to technology uptake and continued use. Moreover, in the multicenter randomized BP Pivotal trial,¹⁷ a key finding was the significant impact on hemoglobin A1c, where the A1c decreased from 7.9% to 7.3% in the BP group and remained at 7.7% for the standard of care (SC) group. It is possible that group differences in A1c vary based on baseline psychosocial characteristics, such as diabetes distress, fear of hypoglycemia, and other psychosocial factors that are associated with A1c. Therefore, the purpose of this article is to assess (1) psychosocial impact of BP versus SC, (2) user experiences (e.g., benefits, burdens, satisfaction) with the BP, and (3) psychosocial moderators of BP effects on A1c.

Materials and Methods

The BP Pivotal Trial was a 13-week multicenter randomized trial of individuals 6 to 83 years of age with T1D, which evaluated the efficacy and safety of the insulin-only configuration of the BP system. The protocol was approved by a central institutional review board, and written informed consent was obtained from each participant. An investigational device exemption for the conduct of the trial was approved by the U.S. Food and Drug Administration. The main trial findings have been published.^17,18 The full protocol is available on NEJM.org.

Participants

Participants included 275 adults (107 randomly assigned to the BP group with either insulin aspart or insulin lispro, 114 to the BP group with fast-acting insulin aspart, and 54 to the SC group), age ranged from 18 to 83 years (mean 43 ± 16 years). For this article, the two BP groups using different insulins were combined and compared against the SC group. Of the 221 participants in the BP group, 217 completed the trial and all 54 of the SC group participants completed the trial. Participants also included 71 adolescents 13–17 years old (48 randomly assigned to the BP group with insulin aspart/lispro and 23 to the SC group), and 94 children 6–12 years old (64 in BP group with insulin aspart/lispro and 30 in SC group). All enrolled adolescents and children completed the trial.

Study participants self-identified ethnic and racial background and included the following: Non-Hispanic White (77%), Non-Hispanic Black (10%), Hispanic or Latino (9%), and Other or more than one race (4%; Table 1). At the time of randomization, 31% of participants reported using an HCL system, 4% reported using a system with a predictive low-glucose suspend (PLGS) feature, 30% reported using an insulin pump without automation, and 34% reported using multiple daily injections.

Table 1.

Baseline Participant Characteristics

	Adults (N = 275)	Teenagers (N = 71)	Children (N = 94)
Age at randomization, years, mean ± SD	43 ± 16	15 ± 1	10 ± 2
Diabetes duration, years, mean ± SD	26 ± 14	8 ± 4	5 ± 2
A1c at randomization, %, mean ± SD^a	7.7 ± 1.2	8.2 ± 1.4	7.9 ± 0.9
Sex: male, n (%)	135 (49)	41 (58)	54 (57)
Female, n (%)	140 (51)	30 (42)	40 (43)
Race: White non-Hispanic, n (%)	230 (84)	53 (75)	55 (59)
Black non-Hispanic, n (%)	26 (9)	5 (7)	11 (12)
Hispanic or Latino, n (%)	16 (6)	7 (10)	17 (18)
Asian, n (%)	1 (<1)	3 (4)	1 (1)
American Indian/Alaskan Native, n (%)	1 (<1)	0	1 (1)
Multiracial, n (%)	1 (<1)	2 (3)	8 (9)
Unknown, n (%)	0	1 (1)	1 (1)
Technology use: MDI+No CGM, n (%)	26 (9)	4 (6)	3 (3)
MDI+CGM, n (%)	65 (24)	20 (28)	31 (33)
Pump+No CGM, n (%)	18 (7)	1 (1)	0
Pump+CGM, n (%)	57 (21)	17 (24)	41 (44)
Pump+CGM+PLGS, n (%)	13 (5)	2 (3)	4 (4)
Pump+CGM+HCL, n (%)	96 (35)	27 (38)	15 (16)
Family income: <$25,000, n (%)	5 (2)	1 (1)	3 (3)
$25,000 to <$35,000, n (%)	15 (5)	1 (1)	4 (4)
$35,000 to <$50,000, n (%)	10 (4)	2 (3)	7 (7)
$50,000 to <$75,000, n (%)	36 (13)	6 (8)	7 (7)
$75,000 to <$100,000, n (%)	36 (13)	12 (17)	14 (15)
$100,000 to <$200,000, n (%)	91 (33)	23 (32)	29 (31)
≥$200,000, n (%)	44 (16)	20 (28)	22 (23)
Unknown, n (%)	38 (14)	6 (8)	8 (9)
Insurance: Private, n (%)	237 (86)	59 (83)	71 (76)
Medicare/Medicaid, n (%)	25 (9)	6 (8)	18 (19)
Other government insurance, n (%)	6 (2)	5 (7)	5 (5)
None, n (%)	3 (1)	0	0
Unknown, n (%)	4 (1)	1 (1)	0

Two adult participants missing baseline A1c.

CGM, continuous glucose monitoring; HCL, hybrid closed loop; MDI, multiple daily injections; PLGS, predictive low-glucose suspend; SD, standard deviation.

Methods

Participants in the BP groups were provided with the BP system, which included the Dexcom G6 continuous glucose monitor (CGM), as well as a blood glucose meter and test strips, and a ketone meter and test strips. The SC group participants continued to use their own personal insulin delivery method, which could be multiple daily injections of insulin (36%), an insulin pump without automation (29%), a system with a PLGS feature (5%), or an HCL system (30%). All participants in the SC group were provided with a real-time CGM (Dexcom G6) for daily use; if they used another CGM at baseline, they could use it in addition to the provided Dexcom G6.

To assess psychosocial outcomes, participants in both treatment groups completed online surveys through a secure website at baseline, midpoint (6 weeks post-baseline) and trial endpoint (13 weeks). Adults completed surveys assessing diabetes distress, fear of hypoglycemia, confidence in managing hypoglycemic episodes, perceived health status, emotional well-being, and attitudes toward diabetes technologies. Adults also completed measures assessing treatment satisfaction and positive expectations toward AID systems at baseline and endpoint.

Youth 8–17 years of age and caregivers of children 6–17 years of age completed surveys assessing diabetes distress, fear of hypoglycemia, perceived health status, and emotional well-being. Children 6 or 7 years of age did not complete questionnaires, but their caregivers completed the surveys. Like adults, measures of treatment satisfaction and positive expectations of AID systems were completed at baseline and endpoint. Adults and youth enrolled in the BP treatment group completed a measure of perceived benefits and burdens of the BP at endpoint. See Table 2 for a list of measures.

Table 2.

Description of Questionnaires

Questionnaire	Cohort	Description of measure
T1-DDS²²	Adults	Gold standard measure of diabetes distress. 28 items Range of 1 to 6 with higher scores indicating more distress
HFS^21,23,24	Adults Teenagers Children 8–12 years Parents of Teenagers Parents of Children	Assesses both worries and behaviors related to hypoglycemia. For adults: 33 items For teenagers and children: 25 items For parents: 26 items Each item has a range of 0 to 4 and total score calculated by summing all items. Higher scores indicating more hypoglycemia fear
HCS²⁵	Adults	Evaluates level of confidence in managing different situations under which hypoglycemia can occur. 8 items Range of 1 to 4 with higher scores indicating more confidence
PAID-T, P-PAID-T, PAID-C, P-PAID-C^20,26	Teenagers Children 8–12 years Parents of Teenagers Parents of Children	Gold standard measures for understanding distress symptoms related to diabetes. For teenagers: 14 items For children: 11 items For parents of teenagers: 15 items For parents of children: 16 items Range of 16.67 to 100 with higher scores indicating greater burden
EQ5D-5L²⁷	Adults	NICE-approved perceived health status measure. 5 items Each item has a range of 1 to 5. Each item is compared to a value set to determine the summary index score. Higher summary scores indicate better perceived health.
EQ5D-5Y²⁸	Teenagers Children 8–12 years	NICE-approved perceived health status measure. 5 items No summary index score. Higher scores indicate better perceived health.
WHO-5²⁹	Adults Teenagers Children 8–12 years	Measure of perceived mental well-being. 5 items Range of 0 to 100 with higher scores indicating better well-being
DSAT³⁰	Adults	Assesses attitudes related to diabetes technologies and devices. 5 items Range of 5 to 25 with higher scores indicating a more positive attitude toward diabetes technologies and devices.
DTSQ³¹	Adults Teenagers Parents of Teenagers Parents of Children	Assesses current state of satisfaction with treatment regimen (state version) and satisfaction with regimen after an intervention (change version). For adults: 6 items For teenagers: 8 items For parents: 9 items Each item of state version (given at randomization) has a range of 0 to 6, while each item of change version (given at 13 weeks) has a range of −3 to 3. The total score is calculated by summing all items and higher scores indicate greater satisfaction.
INSPIRE³²	BP Group Only Adults Teenagers Children 8–12 years Parents of Teenagers Parents of Children	Assesses positive expectations regarding AID systems. For adults: 22 items For teenagers and children: 17 items For parents: 21 items Range of 0 to 100 with higher scores indicating better experience with the BP device
BPUOS⁶	BP Group Only Adults Teenagers Children 8–12 years Parents of Teenagers Parents of Children	Assesses perceived benefits and burdens of the BP. 35 items—21 benefits, 14 burdens Each item is on a range from 1 to 5 with higher values indicating better experience with BP device for the benefits subscale and worse experience with BP device for the burden subscale.

AID, automated insulin delivery; BPUOS, Bionic Pancreas User Opinion Survey; BP, bionic pancreas; DDS, Diabetes Distress Scale; DSAT, Diabetes Specific Attitudes about Technology; DTSQ, Diabetes Treatment Satisfaction Questionnaire; EQ5D-5L, EuroQOL 5 Dimension 5 Level Adult Version; EQ5D-5Y, EuroQOL 5 Dimension Youth Version; HCS, Hypoglycemia Confidence Scale; HFS, Hypoglycemia Fear Survey; INSPIRE, Insulin Delivery Systems: Perceptions, Ideas, Reflections and Expectations; PAID, Problem Areas in Diabetes; P-PAID, Parent of Teen PAID; PAID-T, Teen PAID; WHO-5, World Health Organization - 5 Well Being Index.

Statistical analyses

The overall sample was divided into the following age groups: adults, teenagers 13–17 years of age and their caregivers, children 8–12 years of age, and caregivers of children 6–12 years of age. All analyses were completed in each of these subsamples. Statistical analyses were performed on an intention-to-treat basis and included all participants who had undergone randomization. The treatment effects of the BP versus SC groups on psychosocial outcomes were assessed using linear mixed-effects regression models adjusting for baseline value of the metric, age, and center (random effect). For the psychosocial outcomes with a significant treatment group difference, moderation of the treatment effect by participant demographics was assessed by including an interaction term and the main effect of the moderator in the linear mixed-effects regression model.

In addition, the treatment effect on A1c was assessed by age group using a linear mixed-effects regression model adjusting for baseline A1c, age, and center (random effect). The baseline psychosocial outcomes were examined as a moderator of A1c by including an interaction term and the main effect of the moderator in the linear mixed-effects regression model. Perceived benefits and burdens of the BP as measured by the BP user opinion survey (BPUOS) were compared using paired t-tests within each subgroup. Missing data were handled with the use of direct-likelihood analyses. The type I error was controlled with the use of the Benjamini–Hochberg false discovery rate correction procedure. All analyses were performed in SAS version 9.4 or R version 3.6.2 statistical software.¹⁹

Results

Randomization group differences on psychosocial outcomes

The psychosocial outcomes at baseline, week 6, and week 13 by treatment group are shown in Table 3. For adults, there were significant treatment group differences at week 13 for the diabetes distress scale (T1-DDS), hypoglycemia fear survey (HFS), and World Health Organization Well-Being Index (WHO-5). The T1-DDS score decreased from 1.9 at baseline to 1.6 at week 13 in the BP group and did not change (was 1.8 at both time points) in the SC group (adjusted difference in mean change in T1-DDS = −0.21, confidence interval [95% CI] −0.32 to −0.11, P < 0.001, Table 3). By week 13, the HFS was 6 points lower in the BP group compared with the SC group (P = 0.005), and the WHO-5 total score was 5 points higher in the BP group compared with the SC group (P = 0.02).

Table 3.

Treatment Group Differences in Psychosocial Outcomes

Measure	Baseline		6 Weeks		13 Weeks		13-Week adjusted group difference (95% CI)	P ^a
Measure	BP	SC	BP	SC	BP	SC	13-Week adjusted group difference (95% CI)	P ^a
Adults	N = 221	N = 54	N = 221^b	N = 54^b	N = 221^b	N = 54^b
T1-DDS Overall, mean ± SD	1.9 ± 0.7	1.8 ± 0.7	1.6 ± 0.5	1.8 ± 0.8	1.6 ± 0.5	1.8 ± 0.7	−0.21 (−0.32 to −0.11)	<0.001
HFS Overall, mean ± SD	38 ± 20	37 ± 23	29 ± 19	34 ± 25	28 ± 17	34 ± 24	−6 (−11 to −2)	0.005
HCS Overall, mean ± SD	3.2 ± 0.6	3.3 ± 0.6	3.4 ± 0.5	3.3 ± 0.6	3.3 ± 0.5	3.4 ± 0.5	0.0 (−0.1 to 0.1)	0.99
DSAT Overall, median (IQR)	24 (21, 25)	25 (20, 25)	25 (22, 25)	25 (22, 25)	25 (22, 25)	25 (23, 25)	−0.1 (−0.5 to 0.3)	0.59
EQ5D-5L Summary, median (IQR)	0.940 (0.877, 1.000)	0.940 (0.817, 1.000)	0.943 (0.883, 1.000)	0.943 (0.877, 1.000)	0.943 (0.940, 1.000)	0.943 (0.877, 1.000)	0.001 (0.000 to 0.057)	0.02
WHO-5 Total, mean ± SD	63 ± 17	65 ± 20	66 ± 19	63 ± 21	70 ± 17	66 ± 21	5 (1 to 9)	0.02
Teenagers	N = 48	N = 23	N = 48^d	N = 23^d	N = 48^d	N = 23^d
HFS Overall, mean ± SD
Teenagers	36 ± 13	35 ± 12	32 ± 12	33 ± 15	31 ± 12	32 ± 12	−1 (−6 to 4)	0.56
Parents of Teenager	44 ± 16	41 ± 9	40 ± 16	38 ± 12	42 ± 17	36 ± 10	3 (−2 to 9)	0.18
PAID Total, mean ± SD
Teenagers	35 ± 14	36 ± 17	31 ± 12	32 ± 16	31 ± 14	33 ± 16	−1 (−8 to 6)	0.74
Parents of Teenager	48 ± 18	43 ± 18	44 ± 18	40 ± 13	41 ± 17	43 ± 19	−4 (−13 to 4)	0.29
WHO-5 Total, Teenagers mean ± SD	65 ± 19	65 ± 18	64 ± 20	69 ± 19	68 ± 18	71 ± 19	−3 (−11 to 5)	0.41
Children^e	N = 64^e	N = 30^e	N = 64^e	N = 30^e	N = 64^e	N = 30^e
HFS Overall, mean ± SD
Children	35 ± 14	37 ± 13	31 ± 12	29 ± 13	33 ± 13	31 ± 14	2 (−4 to 8)	0.41
Parents of Child	50 ± 15	48 ± 17	39 ± 15	45 ± 11	43 ± 15	43 ± 15	−2 (−7 to 4)	0.51
PAID Total, mean ± SD
Children	39 ± 18	40 ± 15	38 ± 17	35 ± 12	36 ± 15	34 ± 13	3 (−4 to 9)	0.37
Parents of Child	43 ± 16	43 ± 18	41 ± 17	43 ± 14	40 ± 18	43 ± 15	−2 (−10 to 5)	0.46
WHO-5 Total, Children, mean ± SD	73 ± 18	71 ± 13	72 ± 16	73 ± 18	72 ± 17	74 ± 13	−2 (−9 to 5)	0.60

P-values of treatment group difference computed using a mixed model adjusting for age, baseline scores, and center (random effect). Multiple comparisons were adjusted using the Benjamini–Hochberg adaptive false discovery rate correction procedure. Due to a skewed distribution, T1-DDS, DSAT, and EQ5D-5L scores were rank-transformed.

Some adults did not complete 6-week or 13-week questionnaires. The N's for each questionnaire at 6 weeks for BP group is as follows: T1-DDS—N = 201, HFS—N = 200, HCS—N = 201, DSAT—N = 201, EQ5D-5L—N = 202, WHO-5—N = 201. The N's at 13 weeks for the BP group is as follows: T1-DDS—N = 213, HFS—N = 213, HCS—N = 214, DSAT—N = 213, EQ5D-5L—N = 215, WHO-5—N = 213. The N's at 6 weeks for the SC group is as follows: T1-DDS—N = 49, HFS—N = 48, HCS—N = 49, DSAT—N = 49, EQ5D-5L—N = 49, WHO-5—N = 49. All adults in the SC group completed 13-week questionnaire.

Some teenagers and caregivers did not complete 6-week or 13-week questionnaires. The N's for each questionnaire at 6 weeks for BP group is as follows: HFS—N = 41 teenagers/40 caregivers, PAID—N = 42 teenagers/42 caregivers, WHO-5—N = 42 teenagers. All teenagers and caregivers in the BP group completed 13-week questionnaire. There were 18 teenagers and 18 caregivers in the SC group that completed 6-week questionnaires. There was one teenager and one caregiver in the SC group who did not complete 13-week HFS questionnaire.

Children 6 and 7 years old did not complete the questionnaires (14 in BP group and 2 in SC group), but their caregivers still completed the questionnaire. Some children 8–12 years of age and caregivers of children 6–12 years of age did not complete 6-week or 13-week questionnaires. The N's for each questionnaire at 6 weeks for BP group is as follows: HFS—N = 44 children/61 caregivers, PAID—N = 49 children/61 caregivers, WHO-5—N = 48 children. One child 8–12 years of age in the BP group did not complete 13-week PAID questionnaire. The N's for each questionnaire at 6 weeks for SC group is as follows: HFS—N = 23 children/23 caregivers, PAID—N = 23 children/24 caregivers, WHO-5—N = 23 children. One child 8–12 years of age in the SC group did not complete 13-week WHO-5 questionnaire.

CI, confidence interval; SC, standard of care.

The psychosocial outcomes at weeks 6 and 13 were similar, which suggest the improved psychosocial outcomes occurred soon after randomization. For teenagers and children, the HFS and problem areas in diabetes (PAID) scores were low at baseline with little change during follow-up. There was no significant treatment effect on any psychosocial outcome for teenagers or children.

Caregivers of children had a more positive change in diabetes treatment satisfaction questionnaire (DTSQ) in the BP group compared with the SC group (adjusted difference in mean DTSQ-C 9.1, 95% CI 4.5–13.6, P < 0.001, Table 4). Caregivers of teenagers had a mean DTSQ-C-adjusted treatment group difference of 7.2 points, but this did not reach statistical significance (P = 0.06). No other psychosocial outcome for caregivers of children or caregivers of teenagers was found to be significantly different by treatment.

Table 4.

Diabetes Treatment Satisfaction Questionnaire Results Summary

	Baseline DTSQ-s		Week 13 DTSQ-c		13-Week adjusted group difference (95% CI)	P
	BP, mean ± SD	SC, mean ± SD	BP, mean ± SD	SC, mean ± SD	13-Week adjusted group difference (95% CI)	P
Adults	26 ± 6	27 ± 6	10 ± 8	9 ± 7	1.4 (−1.0 to 3.9)	0.22
Teenagers	36 ± 7	34 ± 7	14 ± 9	10 ± 6	3.7 (−1.4 to 8.8)	0.16
Parents of Teenager	42 ± 10	42 ± 10	16 ± 15	9 ± 10	7.2 (−0.2 to 14.5)	0.06
Parents of Child	43 ± 7	43 ± 7	19 ± 10	11 ± 10	9.1 (4.5 to 13.6)	<0.001

Baseline DTSQ-s scores represent initial treatment satisfaction with higher scores corresponding to higher satisfaction (total score range from 0 to 36 for adults, 0 to 48 for teenagers, and 0 to 54 for caregivers of teenagers and children). Week-13 DTSQ-c scores reflect perceived change in treatment satisfaction with negative scores reflecting decreases, zero reflecting no change, and positive scores reflecting increases in satisfaction (total score range from −18 to 18 for adults, −24 to 24 for teenagers, and −27 to 27 for the caregivers). P-values of treatment group difference in DTSQ-c scores are computed from a mixed model adjusting for age and baseline DTSQ-s scores and center (random effect). Multiple comparisons were adjusted using the Benjamini–Hochberg adaptive false discovery rate correction procedure.

DTSQ-c, Diabetes Treatment Satisfaction Questionnaire-change version; DTSQ-s, Diabetes Treatment Satisfaction Questionnaire-status version.

Participant characteristics as moderators of psychosocial outcomes

The three psychosocial outcomes that demonstrated a significant treatment group difference for the participants (T1-DDS adults, HFS adults, and WHO-5 adults) were further explored to determine if any participant characteristic was a treatment moderator on the psychosocial outcomes. No participant characteristic was determined to be a treatment moderator on T1-DDS adults or HFS adults (Supplementary Tables S1 and S2). Prestudy CGM use was not a significant moderator for the T1-DDS, HFS, or WHO-5 adult scores, nor was any combination of CGM, pump, PGLS, or HCL use at baseline. Participant sex was found to be a moderator on WHO-5 adults (Table 5). The mean WHO-5 for women increased in the BP group and decreased in the SC group, while for men, there were similar increases between groups (P-value for interaction = 0.03). Thus, the BP led to a significantly greater impact on the WHO-5 than SC for women compared to men.

Table 5.

Participant Characteristics Examined as Treatment Moderators on WHO-5 in Adults

Demographic	WHO-5 at baseline				WHO-5 at week 6				WHO-5 at week 13				Interaction P^a
	BP		SC		BP		SC		BP		SC
	N	Mean ± SD	N	Mean ± SD	N	Mean ± SD	N	Mean ± SD	N	Mean ± SD	N	Mean ± SD
Overall	221	63 ± 17	54	65 ± 20	201	66 ± 19	49	63 ± 21	213	70 ± 17	54	66 ± 21
Age at randomization
18–25 years	39	66 ± 19	7	62 ± 14	35	67 ± 19	6	63 ± 14	39	72 ± 18	7	58 ± 20	0.38
26–49 years	99	59 ± 15	28	61 ± 21	91	62 ± 17	25	59 ± 21	95	67 ± 16	28	63 ± 23
≥50 years	83	67 ± 18	19	71 ± 18	75	71 ± 19	18	69 ± 22	79	73 ± 16	19	73 ± 17
Diabetes duration
<20 years	91	62 ± 19	15	59 ± 22	82	65 ± 20	13	52 ± 22	87	70 ± 17	15	59 ± 22	0.45
≥20 years	130	64 ± 16	39	67 ± 18	118	67 ± 17	36	67 ± 19	125	70 ± 16	39	69 ± 21
HbA1c at randomization (%)
<7.0%	59	64 ± 16	16	68 ± 19	49	67 ± 16	12	61 ± 19	54	68 ± 14	16	67 ± 20	0.90
7.0 to <8.0%	83	61 ± 17	19	65 ± 17	76	65 ± 18	18	68 ± 20	81	69 ± 16	19	70 ± 16
8.0 to <9.0%	52	66 ± 18	12	65 ± 25	49	65 ± 20	12	61 ± 25	51	72 ± 19	12	63 ± 27
≥9.0%	26	60 ± 20	6	58 ± 20	26	66 ± 21	6	61 ± 22	26	73 ± 18	6	67 ± 19
Sex
Female	114	62 ± 16	26	64 ± 19	105	66 ± 18	22	58 ± 23	109	71 ± 17	26	60 ± 21	0.03
Male	107	64 ± 19	28	66 ± 20	96	66 ± 19	27	67 ± 19	104	69 ± 16	28	72 ± 20
Race
White non-Hispanic	183	63 ± 17	47	65 ± 19	164	65 ± 19	42	64 ± 21	176	69 ± 16	47	67 ± 21	0.88
Other race	38	66 ± 18	7	61 ± 24	37	70 ± 17	7	58 ± 20	37	76 ± 18	7	62 ± 25
Pre-study CGM use
CGM	186	63 ± 17	45	67 ± 18	170	66 ± 18	40	65 ± 20	178	69 ± 16	45	68 ± 19	0.60
No CGM	35	65 ± 17	9	54 ± 24	31	67 ± 21	9	56 ± 26	35	74 ± 20	9	56 ± 28
Pre-study technology use
MDI+No CGM	20	60 ± 17	6	54 ± 21	18	64 ± 23	6	57 ± 20	20	74 ± 20	6	61 ± 18	0.88
MDI+CGM	53	61 ± 18	12	69 ± 13	51	68 ± 16	10	58 ± 24	50	70 ± 18	12	68 ± 14
Pump+No CGM	15	73 ± 14	3	55 ± 36	13	71 ± 19	3	55 ± 41	15	74 ± 21	3	48 ± 45
Pump+CGM	43	64 ± 17	14	60 ± 20	38	66 ± 20	12	60 ± 21	42	71 ± 16	14	62 ± 24
Pump+CGM+PLGS	11	66 ± 14	2	68 ± 11	10	64 ± 16	1	60	11	64 ± 12	2	76 ± 0
Pump+CGM+HCL	79	62 ± 18	17	70 ± 20	71	65 ± 19	17	72 ± 14	75	68 ± 15	17	72 ± 20
Family income
<$50,000	22	64 ± 15	8	51 ± 26	19	66 ± 17	6	45 ± 17	22	73 ± 20	8	53 ± 27	0.88
$50,000 to <$100,000	60	60 ± 16	12	65 ± 18	57	64 ± 20	11	67 ± 17	59	70 ± 17	12	65 ± 22
≥$100,000	111	63 ± 17	24	67 ± 18	103	66 ± 18	23	62 ± 23	106	69 ± 16	24	68 ± 18
Insurance
Private	192	62 ± 17	45	67 ± 19	175	65 ± 18	41	64 ± 22	186	69 ± 16	45	67 ± 21	0.90
Other insurance	25	69 ± 20	9	54 ± 18	22	73 ± 20	8	61 ± 18	24	74 ± 24	9	60 ± 22

P-values of treatment group by moderator interaction computed using a mixed effect linear model adjusting for age, baseline scores, main effect of moderator, and center (random effect). Missing data were handled by direct likelihood, and multiple comparisons were adjusted using the Benjamini–Hochberg adaptive false discovery rate correction procedure.

Randomization group differences on A1c and psychosocial outcomes as moderators of A1c

The BP Pivotal Trial¹⁷ found a significant reduction in A1c among all study participants (children, teenagers, and adults). Based on these significant changes in A1c, we examined the psychosocial moderators of this change. Compared to adults with high Diabetes Specific Attitudes about Technology (DSAT) scores (more positive attitudes toward diabetes technology), adults with low DSAT scores had higher mean A1c at baseline overall and a greater BP treatment effect of reducing A1c (P < 0.001, Table 6). Children with caregivers having high PAID scores (more distress) and children with lower WHO-5 scores (worse well-being) also had higher mean A1c at baseline overall and a greater BP treatment effect of reducing A1c (P = 0.002 and P = 0.006, respectively). Teenagers showed similar trends with worse psychosocial outcomes tending to have greater BP treatment effect compared to teenagers with better psychosocial outcomes, but none of the interactions reached significance.

Table 6.

Baseline Psychosocial Outcomes Examined as Treatment Moderators on A1c

Baseline PRO	HbA1c at baseline				HbA1c at week 6				HbA1c at week 13				Interaction P^a
	BP		SC		BP		SC		BP		SC
	N	Mean ± SD	N	Mean ± SD	N	Mean ± SD	N	Mean ± SD	N	Mean ± SD	N	Mean ± SD
Adults
Overall	220	7.7 ± 1.2	53	7.6 ± 1.2	217	7.2 ± 0.7	54	7.5 ± 1.0	213	7.1 ± 0.6	53	7.5 ± 0.9
T1-DDS Overall
<1.5	74	7.4 ± 0.9	20	7.2 ± 0.8	72	7.1 ± 0.6	21	7.3 ± 0.7	71	7.0 ± 0.6	21	7.4 ± 0.8	0.23
1.5 to <2.5	112	7.8 ± 1.3	26	7.6 ± 1.1	111	7.3 ± 0.7	26	7.4 ± 0.9	108	7.1 ± 0.6	26	7.5 ± 1.0
≥2.5	34	7.8 ± 1.6	7	8.5 ± 1.7	34	7.2 ± 0.8	7	8.2 ± 1.4	34	7.1 ± 0.5	6	8.0 ± 1.2
HFS Overall
<25	68	7.5 ± 1.1	19	7.4 ± 0.7	66	7.1 ± 0.7	20	7.4 ± 0.6	64	7.0 ± 0.6	20	7.6 ± 0.8	0.81
25 to <50	96	7.7 ± 1.2	18	7.4 ± 1.2	95	7.2 ± 0.7	18	7.2 ± 1.0	94	7.1 ± 0.6	18	7.4 ± 0.9
≥50	56	7.8 ± 1.3	16	8.0 ± 1.4	56	7.3 ± 0.7	16	7.8 ± 1.2	55	7.2 ± 0.5	15	7.6 ± 1.1
HCS Overall
<3	63	7.7 ± 0.9	12	7.7 ± 1.5	63	7.3 ± 0.7	12	7.6 ± 1.2	62	7.1 ± 0.6	11	7.4 ± 0.9	0.42
3 to <3.5	79	7.8 ± 1.4	21	7.5 ± 1.3	77	7.2 ± 0.7	21	7.3 ± 1.1	77	7.1 ± 0.5	21	7.4 ± 1.1
≥3.5	78	7.6 ± 1.2	20	7.6 ± 0.8	77	7.2 ± 0.7	21	7.6 ± 0.6	74	7.0 ± 0.7	21	7.7 ± 0.7
DSAT Overall
<25	84	8.0 ± 1.4	18	8.3 ± 1.2	83	7.3 ± 0.6	19	8.0 ± 0.9	84	7.1 ± 0.5	18	8.2 ± 0.9	<0.001
25 to <30	65	7.7 ± 1.2	12	7.2 ± 1.0	66	7.2 ± 0.7	12	7.1 ± 1.0	62	7.0 ± 0.6	12	7.0 ± 0.7
≥30	71	7.4 ± 1.0	23	7.2 ± 0.9	68	7.1 ± 0.7	23	7.2 ± 0.8	67	7.0 ± 0.6	23	7.3 ± 0.8
WHO-5 Total
<60	77	7.7 ± 1.1	14	7.8 ± 1.6	77	7.3 ± 0.6	15	7.6 ± 1.2	77	7.1 ± 0.6	15	7.7 ± 1.2	0.56
60 to <70	65	7.8 ± 1.5	17	7.4 ± 0.8	65	7.2 ± 0.7	17	7.4 ± 0.8	62	7.0 ± 0.5	16	7.2 ± 0.6
≥70	78	7.6 ± 1.1	22	7.5 ± 1.1	75	7.2 ± 0.7	22	7.5 ± 0.9	74	7.1 ± 0.6	22	7.6 ± 1.0
Teenagers
Overall	48	8.5 ± 1.4	23	7.7 ± 1.3	48	7.7 ± 0.8	23	7.8 ± 1.3	46	7.6 ± 0.7	22	7.7 ± 1.2
HFS Teenager Overall
<30	16	8.3 ± 1.4	5	8.1 ± 1.8	16	7.8 ± 0.8	5	8.2 ± 2.0	15	7.7 ± 0.7	4	7.7 ± 1.4	0.07
30 to <40	15	8.5 ± 1.4	12	7.7 ± 0.8	15	7.7 ± 0.8	12	7.8 ± 0.7	15	7.5 ± 0.7	12	8.1 ± 1.1
≥40	17	8.5 ± 1.4	6	7.5 ± 1.7	17	7.7 ± 0.9	6	7.2 ± 1.5	16	7.5 ± 0.8	6	7.1 ± 1.2
HFS Parents Overall
<40	15	8.9 ± 1.7	12	7.8 ± 1.3	15	7.8 ± 0.8	12	7.8 ± 1.5	15	7.5 ± 0.7	11	7.9 ± 1.5	0.10
40 to <50	15	8.3 ± 1.3	8	7.6 ± 0.8	15	7.9 ± 0.8	8	7.6 ± 0.6	13	7.8 ± 0.6	8	7.5 ± 0.6
≥50	18	8.2 ± 1.1	3	8.1 ± 2.4	18	7.6 ± 0.8	3	8.0 ± 2.1	18	7.5 ± 0.8	3	7.9 ± 1.7
PAID Teenager Total
<25	14	8.5 ± 1.5	6	7.9 ± 0.7	14	7.7 ± 0.9	6	8.1 ± 0.6	14	7.6 ± 0.7	6	8.4 ± 1.5	0.43
25 to <40	18	8.1 ± 1.2	10	7.0 ± 1.1	18	7.6 ± 0.6	10	7.0 ± 1.0	17	7.5 ± 0.7	10	7.3 ± 1.1
≥40	16	8.9 ± 1.5	7	8.7 ± 1.3	16	7.9 ± 0.9	7	8.6 ± 1.5	15	7.5 ± 0.7	6	7.9 ± 1.0
PAID Parents Total
<40	18	8.5 ± 1.7	11	7.0 ± 0.8	18	7.6 ± 0.9	11	7.2 ± 0.8	18	7.4 ± 0.8	11	7.2 ± 0.7	0.07
40 to <60	17	8.5 ± 1.2	8	8.2 ± 1.2	17	7.8 ± 0.7	8	8.1 ± 1.5	16	7.6 ± 0.5	7	7.7 ± 1.0
≥60	13	8.3 ± 1.2	4	8.9 ± 1.4	13	7.9 ± 0.8	4	8.8 ± 1.2	12	7.7 ± 0.8	4	9.3 ± 1.6
WHO-5 Teenager Total
<55	15	8.0 ± 1.2	9	7.2 ± 1.1	15	7.5 ± 0.7	9	7.1 ± 0.9	14	7.5 ± 0.8	9	7.0 ± 0.8	0.22
55 to <70	14	8.8 ± 1.5	3	9.2 ± 1.3	14	7.9 ± 0.9	3	9.3 ± 1.8	13	7.5 ± 0.7	2	8.9 ± 1.1
≥70	19	8.6 ± 1.4	11	7.8 ± 1.2	19	7.8 ± 0.8	11	7.9 ± 1.1	19	7.6 ± 0.7	11	8.1 ± 1.3
Children
Overall	64	7.9 ± 1.0	30	7.8 ± 0.9	64	7.6 ± 0.6	30	7.8 ± 0.9	64	7.5 ± 0.6	29	7.8 ± 0.9
HFS Children Overall
<30	16	7.6 ± 0.9	8	7.9 ± 0.7	16	7.6 ± 0.6	8	7.8 ± 0.7	16	7.4 ± 0.5	8	7.9 ± 0.7	0.43
30 to <40	16	7.8 ± 0.8	12	8.0 ± 0.9	16	7.6 ± 0.5	12	8.1 ± 0.7	16	7.4 ± 0.6	11	8.1 ± 0.9
≥40	18	8.2 ± 1.1	8	7.9 ± 1.0	18	7.7 ± 0.4	8	7.7 ± 1.0	18	7.6 ± 0.4	8	7.7 ± 1.0
HFS Parents Overall
<45	28	7.9 ± 0.8	16	7.8 ± 1.0	28	7.7 ± 0.6	16	7.9 ± 0.9	28	7.6 ± 0.6	15	7.8 ± 1.0	0.20
45 to <60	15	7.5 ± 0.6	8	7.6 ± 1.1	15	7.4 ± 0.4	8	7.5 ± 1.0	15	7.3 ± 0.5	8	7.7 ± 1.1
≥60	21	8.2 ± 1.2	6	8.1 ± 0.8	21	7.6 ± 0.7	6	8.0 ± 0.8	21	7.4 ± 0.6	6	8.1 ± 0.7
PAID Children Total
<30	17	7.5 ± 0.9	10	8.0 ± 0.7	17	7.3 ± 0.5	10	7.8 ± 0.7	17	7.3 ± 0.5	10	7.8 ± 1.0	0.25
30 to <45	17	7.9 ± 1.0	7	8.0 ± 0.6	17	7.7 ± 0.5	7	8.0 ± 0.6	17	7.5 ± 0.5	7	8.1 ± 0.5
≥45	16	8.4 ± 0.7	11	7.8 ± 1.1	16	7.8 ± 0.5	11	7.9 ± 1.0	16	7.5 ± 0.5	10	7.9 ± 1.1
PAID Parents Total
<35	26	7.5 ± 0.9	13	7.5 ± 0.9	26	7.5 ± 0.6	13	7.5 ± 0.9	26	7.5 ± 0.6	12	7.3 ± 0.9	0.002
35 to <55	22	8.0 ± 1.1	7	8.2 ± 0.9	22	7.7 ± 0.7	7	8.0 ± 0.8	22	7.5 ± 0.6	7	8.1 ± 0.8
≥55	16	8.3 ± 0.7	10	8.1 ± 1.0	16	7.7 ± 0.5	10	8.0 ± 1.0	16	7.5 ± 0.5	10	8.2 ± 0.9
WHO-5 Children Total
<70	17	8.2 ± 0.8	13	8.3 ± 0.8	17	7.7 ± 0.4	13	8.2 ± 0.8	17	7.5 ± 0.5	13	8.3 ± 0.8	0.006
70 to <80	11	8.0 ± 1.3	5	8.0 ± 0.7	11	7.7 ± 0.7	5	8.0 ± 0.7	11	7.4 ± 0.4	5	7.7 ± 1.0
≥80	22	7.6 ± 0.9	10	7.5 ± 0.8	22	7.5 ± 0.5	10	7.5 ± 0.8	22	7.4 ± 0.5	9	7.5 ± 0.8

PRO, Patient Reported Outcomes.

Acceptability of the BP

With respect to the measure assessing perceived benefits and burdens of the BP (BPUOS), all respondent groups (adults, teenagers, children, and caregivers) endorsed significantly greater benefits (mean = 3.50–3.89) versus burdens (mean = 2.15–2.34, all P-values <0.001). There were common trends among the respondents. With respect to perceived benefits (21 items), the categories assessed included feeling freer to live their lives, spending less time thinking about diabetes, feeling less worried and burdened by diabetes, preventing out of range glucose levels, trusting the device, and wanting to continue/recommend the device to others.

Most participants reported strongly recommending the BP (74%–81%), trusting the device (70%–82%), worrying less about high glucose levels (68%–75%), wanting to continue device use after the study (67%–75%), and feeling freer to live their lives (55%–77%). Teenagers thought less about diabetes (71%) and children felt they could eat more freely (85%). Caregivers endorsed having greater peace of mind with device use (77% parent of teen and 84% parent of child), felt more relaxed (77%–79%), had fewer worries about blood glucose levels (75%–79%), felt freer to allow their child to do what they wanted to do (73%–77%), and perceived their children as having more freedom to live their lives (80%–81%).

With respect to perceived burdens (14 items), the categories assessed included hassles with device use, including alarms, feeling the system was intrusive, the length of time the system took to respond to out-of-range glucose levels, not trusting the system, and worrying about glucose levels while sleeping. For all respondent groups, only a minority endorsed any burden with the device, reporting that the alarms were annoying (adults, 39%; teenagers, 36%; and children, 35%), wanting the device to respond faster to high (adults, 40%; caregivers of teens, 41%; and caregivers of children, 44%) and low (adults, 42% and caregivers of younger children, 41%) glucose levels, and continuing to worry about having a low while sleeping (adults, 31%; caregivers of teens, 43%; and caregivers of children, 35%).

Participants in the BP group were asked about their past experience using an AID system and how they perceived the BP in comparison. There were 92 adults, 18 teenagers, and 10 children who reported using an AID system in the past. For adults, 60% (n = 55) reported that the BP was better or much better than their past system and 53% (n = 49) reported that they prefer the BP over not using an AID system or their current system.

For adolescents, 67% (n = 12) reported that the BP was better or much better than their past system, and 61% (n = 11) reported that they prefer the BP over not using an AID system or their current system. For children, 90% (n = 9) reported that the BP was better or much better than their past system and 80% (n = 8) reported that they prefer the BP over not using an AID system or their current system.

The INSPIRE measure assesses positive expectancies about an AID system with respect to decreasing glycemic variability, making diabetes management easier, and improving health-related quality of life. Baseline and endpoint INSPIRE scores for the BP system for each subsample (Table 7) were high (mean score ranges from 79 to 82 at baseline and from 69 to 77 at endpoint). Participants had positive expectations before and after use of the BP in terms of its utility and benefit. The perceived health status for teenagers and children was high based on the EuroQOL 5 Dimension Youth Version (EQ5D-Y) measure with most participants having no problem doing usual activities and having no pain or discomfort at baseline and during follow-up (Supplementary Table S3). Further, the adult-reported EuroQOL 5 Dimension 5 Level Adult Version (EQ5D-5L) demonstrated significant differences between treatment groups, favoring BP, for improvement in perceived health status (Table 3).

Table 7.

INSPIRE Results Summary for Bionic Pancreas

Cohort	Baseline, mean ± SD	Week 13, mean ± SD	P
Adults	79 ± 14	69 ± 21	<0.001
Teenagers	80 ± 14	73 ± 19	0.054
Parents of Teenagers	81 ± 13	72 ± 26	0.121
Children	82 ± 11	73 ± 17	0.001
Parents of Children	80 ± 15	77 ± 18	0.175

Discussion

We assessed the psychosocial outcomes of the insulin-only configuration of the BP in a multicenter, randomized trial of adults and youth with T1D. Most participants in all age groups would recommend using the BP. Likewise, the majority who had used another automated delivery device preferred use of the BP over the other AID devices that they had tried previously. Adult participants reported significant decreases in fear of hypoglycemia and in diabetes-specific emotional distress, as well as improvements in their perceived well-being. In comparison to most other studies assessing AID systems, our study offered an in-depth assessment of perceived benefits and burdens, with benefits significantly outweighing burdens. Our findings regarding treatment satisfaction and improved well-being are consistent with prior studies of HCL systems.^6,12,14,15

Adults reported significant improvements in diabetes-specific emotional distress, which was also seen in Tandem Control IQ studies^4,5 but not in a study of the Medtronic 670G system.¹² The observed treatment group reduction of 0.21 on diabetes distress exceeded the minimal clinically important difference of 0.19 proposed for an individual,⁶ suggesting the BP treatment effect was clinically significant. The BP also showed a significant reduction in fear of hypoglycemia for adults, a finding only reported in one other study.⁵

When we assessed differences in outcomes based on participant characteristics, our findings suggest that women showed a significant improvement in perceived well-being while using the BP and a slight decline in the SC group, whereas men in both treatment groups improved in well-being. Further studies comparing the varied experiences between men and women when using AID systems may be helpful in better understanding individual differences that may impact both uptake and continued use of these systems. Importantly, no other demographic factor moderated the impact on any of the key psychosocial outcomes, suggesting that the BP leads to a broad impact on diabetes distress, fear of hypoglycemia, and well-being for adults across demographic and diabetes-related factors.

Consistent with the adult experiences, children and adolescents reported a high level of perceived benefits and a low number of perceived burdens. Differences in other psychosocial measures pre- and post-BP use were not found, which is consistent with many other published trials assessing a pediatric population.^7
–9,11,13 In contrast to this study, which found significant psychosocial impact for adults relative to parents, children, and adolescents, no other study that included both adults and youth found differences between age groups in psychosocial impact.^2,14,16 Of note, at baseline, our adolescent participants reported lower levels of diabetes distress²⁰ and fear of hypoglycemia²¹ than the normative population, suggesting that there were few concerns at baseline and observing improvements in these outcomes over time would be unusual. However, adult participants also had relatively low levels of diabetes distress and fear of hypoglycemia at baseline and still demonstrated significant change over time in the BP group.

Finally, our data suggest that adults with more negative attitudes toward diabetes technologies (lower DSAT), children whose caregivers reported higher levels of diabetes-specific emotional distress (high DDS), and children with worse perceived well-being (lower WHO-5) at baseline showed a greater improvement in A1c if the participant was in the BP group than if the participant was in the SC group. Further studies are warranted to better understand the potential impact of improving attitudes toward technology, caregiver distress, and children's well-being on the participant's glycemic control.

There are a few limitations in this study. This study reflects a highly diverse group for a clinical trial in T1D (i.e., about 23% identify as not White, 34% were using multiple daily injection therapy, and 37% with A1cs over 8.0% at baseline); however, further studies of the BP in even more diverse samples are necessary. The sample was composed of majority White participants who were relatively well resourced (e.g., most were on private insurance). Assessing the BP in cohorts with greater racial representation as well as assessing psychosocial impacts and attitudes of the BP in less resourced populations would be beneficial.

Second, compared to the T1D Exchange data,⁶ the young adult and adolescent participants had relatively lower mean A1c values at baseline. Testing the BP in samples with higher mean blood glucose is warranted to evaluate its efficacy and effectiveness in the broader diabetes community.

Third, many of the participants had used an insulin pump at baseline and a large majority were already using CGM. Investigating the BP in device-naive or new technology users would provide more evidence for the utility of the BP in this population. Finally, this sample reported lower psychosocial concerns at baseline compared to normative samples of the measures used (particularly for youth). Understanding of the BP in higher distress samples or who have more fears related to diabetes would be useful and more representative of many living with diabetes.

Conclusions

These analyses demonstrate the acceptability and the psychosocial impact of the insulin-only BP. Overall, the BP was well accepted and demonstrated positive outcomes for adults. Children and teenagers reported high acceptability and benefit of the BP with less psychosocial outcome improvement; however, the youth samples also demonstrated floor effects with respect to measured outcomes at baseline (i.e., low diabetes distress and fear of hypoglycemia), which may have reduced the likelihood of observing a significant change.

While a device may demonstrate significant improvements in metabolic management of diabetes, it is imperative to assess the attitudes and perceptions of the device as this ultimately affects uptake and continued use of the device. The insulin-only BP was well accepted by study participants. Translational studies that examine the dissemination and implementation process of the BP and perceptions of the device in the broader community will be essential to ongoing quality improvement and optimizing the experiences of people with diabetes.

Footnotes

Acknowledgments

A complete listing of the Bionic Pancreas Research Group appears in the Online Supplementary Material S1. We thank the participants for their time and effort; the diabetes care providers who referred potential participants to the study staff; and the multicenter research center study staff.

Authors' Contributions

J.W.-B. wrote and edited the article. A.T.V., J.S., and P.C. performed statistical analysis and wrote/edited the article. Z.L. performed statistical analysis and reviewed/edited the article. E.R.D., S.J.R., F.H.E.-K., K.J.R., C.A.B., and R.W.B. contributed to discussion and reviewed/edited the article.

Guarantor Statement

J.W.-B. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of data and the accuracy of data analysis.

Author Disclosure Statement

J.W.-B. reports consulting payments from Beta Bionics. A.T.V. has no personal financial disclosures. J.S. has no personal financial disclosure. P.C. is a former Dexcom employee, and his current employer has received consulting payments on his behalf from vTv Therapeutics, Beta Bionics, Dexcom, and Diasome. E.R.D. has issued patents and pending patents on aspects of bionic pancreas, and is an employee, the Executive Chair of the Board of Directors, and shareholder of Beta Bionics. S.J.R. has issued patents and pending patents on aspects of bionic pancreas that are assigned to Massachusetts General Hospital and licensed to Beta Bionics, has received honoraria and/or travel expenses for lectures from Novo Nordisk, Roche, and Ascensia, serves on the scientific advisory boards of Unomedical, served on scientific advisory board and had stock in Companion Medical that was bought out by Medtronic, has received consulting fees from Beta Bionics, Novo Nordisk, Senseonics, and Flexion Therapeutics, has received grant support from Zealand Pharma, Novo Nordisk, and Beta Bionics, and has received in-kind support in the form of technical support and/or donation of materials from Zealand Pharma, Ascencia, Senseonics, Adocia, and Tandem Diabetes. Z.L. has no personal financial disclosure, but reports that her employer has received grant support from Beta Bionics, Dexcom, and Tandem Diabetes Care. F.H.E.-K. has issued patents and pending patents on aspects of bionic pancreas and is an employee and shareholder of Beta Bionics. K.J.R. has no personal financial disclosure, but reports that her employer has received grant support from Beta Bionics, Dexcom, and Tandem Diabetes Care. C.A.B. reports receiving consulting payments from Beta Bionics, Novo Nordisk, and Zealand Pharma. R.W.B. reports no personal financial disclosure, but reports that his institution has received funding on his behalf as follows: grant funding and study supplies from Tandem Diabetes Care, Beta Bionics, and Dexcom; study supplies from Medtronic, Ascencia, and Roche; consulting fees and study supplies from Eli Lilly and Novo Nordisk; and consulting fees from Insulet, Bigfoot Biomedical, vTv Therapeutics, and Diasome.

Funding Information

This study was funded by the National Institute of Diabetes and Digestive and Kidney Diseases (grant number 1UC4DK108612-01), by an Investigator-Initiated Study award from Novo Nordisk, and by Beta Bionics, Inc., which also provided the bionic pancreas devices used in the study. Fast-acting insulin aspart and insulin aspart were provided by Novo Nordisk and insulin lispro was provided by Eli Lilly. Blood glucose meters and test strips were provided by Ascensia Diabetes Care. Continuous glucose monitor sensors and transmitters were purchased from Dexcom, Inc., at a discounted price.

Supplementary Material

Supplementary Data S1

Supplementary Table S1

Supplementary Table S2

Supplementary Table S3

References

Bekiari

, Kitsios

, Thabit

, et al. Artificial pancreas treatment for outpatients with type 1 diabetes: Systematic review and meta-analysis. BMJ, 2018; 361:k1310; doi: 10.1136/bmj.k1310

Adams

, Tanenbaum

, Hanes

, et al. Psychosocial and human factors during a trial of a AIDAID system for type 1 diabetes management. Diabetes Technol Ther, 2018; 20(10):648–653; doi: 10.1089/dia.2018.0174

Bisio

, Brown

, McFadden

, et al. Sleep and diabetes-specific psycho-behavioral outcomes of a new automated insulin delivery system in young children with type 1 diabetes and their parents. Pediatr Diabetes, 2021; 22(3):495–502; doi: 10.1111/pedi.13164

Bisio

, Gonder-Frederick

, McFadden

, et al. The impact of a recently approved automated insulin delivery system on glycemic, sleep, and psychosocial outcomes in older adults with type 1 diabetes: A pilot study. J Diabetes Sci Technol, 2022; 16(3):663–669; doi: 10.1177/1932296820986879

Kudva

, Laffel

, Brown

, et al. Patient-reported outcomes in a randomized trial of closed-loop control: The pivotal international diabetes closed-loop trial. Diabetes Technol Ther, 2021; 23(10):673–683; doi: 10.1089/dia.2021.0089

Weissberg-Benchell

, Hessler

, Fisher

, et al. Impact of an automated bihormonal delivery system on psychosocial outcomes in adults with type 1 diabetes. Diabetes Technol Ther, 2017; 19(12):723–729; doi: 10.1089/dia.2017.0174

Abraham

, de Bock

, Smith

, et al. Effect of a hybrid closed-loop system on glycemic and psychosocial outcomes in children and adolescents with type 1 diabetes: A randomized clinical trial. JAMA Pediatr, 2021; 175(12):1227–1235; doi: 10.1001/jamapediatrics.2021.3965

Berget

, Messer

, Vigers

, et al. Six months of AIDAID in the real-world: An evaluation of children and young adults using the 670G system. Pediatr Diabetes, 2020; 21(2):310–318; doi: 10.1111/pedi.12962

Cobry

, Bisio

, Wadwa

, et al. Improvements in parental sleep, fear of hypoglycemia, and diabetes distress with use of an advanced hybrid closed-loop system. Diabetes care, 2022; 45(5):1292–1295; doi: 10.2337/dc21-1778

10.

Cobry

, Hamburger

, Jaser

. Impact of the hybrid closed-loop system on sleep and quality of life in youth with type 1 diabetes and their parents. Diabetes Technol Ther, 2020; 22(11):794–800; doi: 10.1089/dia.2020.0057

11.

Cobry

, Kanapka

, Cengiz

, et al. Health-related quality of life and treatment satisfaction in parents and children with type 1 diabetes using closed-loop control. Diabetes Technol Ther, 2021; 23(6):401–409; doi: 10.1089/dia.2020.0532

12.

McAuley

, Lee

, Paldus

, et al. Six months of hybrid closed-loop versus manual insulin delivery with Fingerprick blood glucose monitoring in adults with type 1 diabetes: A randomized, controlled trial. Diabetes Care, 2020; 43(12):3024–3033; doi: 10.2337/dc20-1447

13.

Weissberg-Benchell

, Hessler

, Polonsky

, et al. Psychosocial impact of the bionic pancreas during summer camp. J Diabetes Sci Technol, 2016; 10(4):840–844; doi: 10.1177/1932296816640289

14.

Wheeler

, Collyns

, Meier

, et al. Improved technology satisfaction and sleep quality with Medtronic MiniMed^® Advanced Hybrid Closed-Loop delivery compared to predictive low glucose suspend in people with Type 1 Diabetes in a randomized crossover trial. Acta Diabetol, 2022; 59(1):31–37; doi: 10.1007/s00592-021-01789-5

15.

Pinsker

, Müller

, Constantin

, et al. Real-world patient-reported outcomes and glycemic results with initiation of control-IQ technology. Diabetes Technol Ther, 2021; 23(2):120–127; doi: 10.1089/dia.2020.0388

16.

Barnard

, Wysocki

, Ully

, et al. Closing the loop in adults, children and adolescents with suboptimally controlled type 1 diabetes under free living conditions: A psychosocial substudy. J Diabetes Sci Technol, 2017; 11(6):1080–1088; doi: 10.1177/1932296817702656

17.

Russell

, Beck

, Damiano

, et al. Multicenter, randomized trial of a bionic pancreas in type 1 diabetes. N Engl J Med, 2022; 387(13):1161–1172; doi: 10.1056/NEJMoa2205225

18.

Beck

, Russell

, Damiano

, et al. A multicenter randomized trial evaluating fast-acting insulin aspart in the bionic pancreas in adults with type 1 diabetes. Diabetes Technol Ther, 2022; 24(10):681–696; doi: 10.1089/dia.2022.0167

19.

R Core Team. R: A Language and Environment for Statistical Computing. 2018. Available from: https://www.R-project.org [Last accessed; May 15, 2023].

20.

Shapiro

, Vesco

, Weil

LEG

, et al. Psychometric properties of the problem areas in diabetes: Teen and parent of teen versions. J Pediatr Psychol, 2018; 43(5):561–571; doi: 10.1093/jpepsy/jsx146

21.

Gonder-Frederick

, Nyer

, Shepard

, et al. Assessing fear of hypoglycemia in children with Type 1 diabetes and their parents. Diabetes Manag (Lond), 2011; 1(6):627–639; doi: 10.2217/dmt.11.60

22.

Fisher

, Polonsky

, Hessler

, et al. Understanding the sources of diabetes distress in adults with type 1 diabetes. J Diabetes Complications, 2015; 29(4):572–577; doi: 10.1016/j.jdiacomp.2015.01.012

23.

Gonder-Frederick

, Schmidt

, Vajda

, et al. Psychometric properties of the hypoglycemia fear survey-ii for adults with type 1 diabetes. Diabetes care, 2011; 34(4):801–806; doi: 10.2337/dc10-1343

24.

Haugstvedt

, Wentzel-Larsen

, Aarflot

, et al. Assessing fear of hypoglycemia in a population-based study among parents of children with type 1 diabetes—Psychometric properties of the hypoglycemia fear survey—Parent version. BMC Endocr Disord, 2015; 15:2; doi: 10.1186/1472-6823-15-2

25.

Polonsky

, Fisher

, Hessler

, et al. Investigating hypoglycemic confidence in type 1 and type 2 diabetes. Diabetes Technol Ther, 2017; 19(2):131–136; doi: 10.1089/dia.2016.0366

26.

Evans

, Weil

LEG

, Shapiro

, et al. Psychometric properties of the parent and child problem areas in diabetes measures. J Pediatr Psychol, 2019; 44(6):703–713; doi: 10.1093/jpepsy/jsz018

27.

McCaffrey

, Kaambwa

, Currow

, et al. Health-related quality of life measured using the EQ-5D-5L: South Australian population norms. Health Qual Life Outcomes, 2016; 14(1):133; doi: 10.1186/s12955-016-0537-0

28.

Wille

, Badia

, Bonsel

, et al. Development of the EQ-5D-Y: A child-friendly version of the EQ-5D. Qual Life Res, 2010; 19(6):875–886; doi: 10.1007/s11136-010-9648-y

29.

Topp

, Østergaard

, Søndergaard

, et al. The WHO-5 Well-Being Index: A systematic review of the literature. Psychother Psychosom, 2015; 84(3):167–176; doi: 10.1159/000376585

30.

Naranjo

, Tanenbaum

, Iturralde

, et al. Diabetes technology: Uptake, outcomes, barriers, and the intersection with distress. J Diabetes Sci Technol, 2016; 10(4):852–858; doi: 10.1177/1932296816650900

31.

Bradley

The Diabetes Treatment Satisfaction Questionnaire: DTSQ. In: Handbook of Psychology and Diabetes: A Guide to Psychological Measurement in Diabetes Research and Practice. ( Bradley

ed.) Harwood Academic Publishers: London; 1994; pp. 111–132.

32.

Weissberg-Benchell

, Shapiro

, Hood

, et al. Assessing patient-reported outcomes for automated insulin delivery systems: The psychometric properties of the INSPIRE measures. Diabet Med, 2019; 36(5):644–652; doi: 10.1111/dme.13930

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

0.02 MB

0.01 MB