INTRODUCTION
Diabetes is associated with a large clinical and economic burden in Japan1, 2, 3, 4. The International Diabetes Federation estimates that, in 2024, there were a total of 79,463 people with type 1 diabetes (T1D) in Japan, including 9,382 individuals <20 years of age1. Some studies also suggest that the incidence of T1D among children in the country has been rising in recent decades5, 6.
The impact of T1D on those living with it is substantial, with a large retrospective cohort study conducted in the country finding that people with T1D have a life expectancy 8.6 years shorter than those without diabetes7. Long‐term diabetes‐related complications have also been found to be common, with a sub‐analysis of adults with T1D in the Study of Adults' Glycemia in T1D (SAGE) finding that 36.7% of participants report at least one microvascular complication8. Similarly, a cross‐sectional study in Saitama prefecture found that 26.3% of people with T1D experience retinopathy, 20.8% experience nephropathy, and 21.5% have neuropathy9.
The achievement of good glycemic management is important in order to attempt to prevent these complications10, 11, 12 and in reducing mortality13, with the Japanese Clinical Practice Guidelines for Diabetes recommending a target of a glycated hemoglobin (HbA1c) <7.0% for preventing complications14. Achievement of HbA1c management targets has also been shown to be associated with some aspects of treatment satisfaction among people with T1D in Japan15. However, studies in Japan have also found that only a small proportion of people with T1D (23.0–25.8%) achieve an HbA1c target of <7.0%8, 9.
Improvements in technology for diabetes management in recent years, particularly continuous glucose monitoring (CGM) and automated insulin delivery (AID) systems, have been found to improve the glycemic management and quality of life (QoL) of people with T1D16. In particular, hybrid‐closed loop systems, which integrate both a CGM and an insulin pump, can automatically adjust insulin delivery to promote glycemic management. More recently, advanced hybrid closed‐loop (AHCL) devices that also incorporate automated correctional boluses have become available17.
In Japan specifically, the MiniMed 780G system became available in November 2023. The MiniMed 780G system includes automated correction boluses in addition to automated basal insulin delivery, and allows the user to select the target glucose level; it replaces the MiniMed 770G system, which has a fixed glucose target and adjusts basal insulin only18. In a trial comparing the MiniMed 780G system with multiple daily insulin injections (MDI) in combination with intermittent‐scanning continuous glucose monitoring (isCGM), the MiniMed 780G system was found to confer better glycemic outcomes, including HbA1c and time‐in‐range (TIR), than MDI plus isCGM, in addition to improvements in QoL and treatment satisfaction, and reductions in fear of hypoglycemia19, 20.
Furthermore, real‐world evidence from 101,629 users across 34 countries has demonstrated that the majority of MiniMed 780G system users achieve glycemic targets21. Switching from the MiniMed 770G to the MiniMed 780G system has also been associated with improvements in glycemic outcomes and treatment‐related QoL in studies conducted in Japan22, 23. However, despite such potential benefits associated with AID usage, the majority of people (72.1–85.6%) with T1D in Japan continue to receive treatment with MDI rather than insulin pumps8, 9.
The health economic analysis presented here was performed to provide robust evidence relating to the long‐term clinical effectiveness and economic outcomes associated with the use of the MiniMed 780G specifically in Japan. Such long‐term analyses provide insights that may inform decision makers as well as supporting reimbursement and policy discussions, which may ultimately improve access, enabling more people to benefit from the MiniMed 780G system. Here, long‐term outcomes with the MiniMed 780G system were compared with MDI in combination with CGM in adults with T1D and a HbA1c above target levels in Japan.
METHODS
Model structure
The IQVIA CORE Diabetes Model (CDM) (version 10) was used to perform a cost‐effectiveness analysis comparing the use of the MiniMed 780G system with MDI and CGM24, 25. The CDM consists of 17 interdependent Markov sub‐models and can be used to predict long‐term health outcomes and costs associated with diabetes care and complications. These outcomes include life expectancy, quality‐adjusted life years (QALYs), complication incidences, time free of complications, direct costs, indirect costs, and incremental cost‐effectiveness ratios (ICERs).
The structure of the model and its validation have been described in further detail by Palmer et al.26 and McEwan et al.27
Baseline cohort characteristics
The baseline characteristics were primarily derived from the Advanced Hybrid Closed Loop Study in Adult Population with Type 1 Diabetes (ADAPT) study19, and CDM defaults25. The ADAPT study was a randomized controlled trial (RCT) carried out at 14 centers across France, Germany, and the UK that compared MiniMed 780G with isCGM plus MDI in adults with T1D who had a HbA1c of ≥8%19. ADAPT was used as the source of the HbA1c treatment effects (described below) and was therefore also used to inform baseline cohort characteristics as changes in HbA1c are well established to correlate with baseline HbA1c values28, 29, 30.
Cohort characteristics at baseline are listed in Table 1. Briefly, the mean age of the cohort was 40.6 ± 12.4 years and the average duration of diabetes was 18 ± 10.7 years19. The mean HbA1c was 9.04 ± 0.85% (based on average HbA1c across the trial)19.
One key difference from the ADAPT cohort characteristics was the assumption of a baseline body mass index (BMI) of 22 kg/m2. This was based on the use of such a BMI as a reference in weight‐focused public health studies conducted in Japan31 and is supported by cohort data from T1D studies such as the Japan Diabetes Complication and its Prevention Prospective Study 2 population32. Additionally, background mortality rates were sourced from life tables published by the Japanese Ministry of Health, Labour and Welfare33.
Treatment effectiveness inputs
Treatment effects considered in the base‐case analysis included changes in HbA1c, and differences in the incidence of non‐severe hypoglycemic events, severe hypoglycemic events (SHEs) requiring non‐medical third‐party intervention, SHEs requiring medical assistance, and diabetic ketoacidosis (DKA). Changes in HbA1c were sourced from the ADAPT RCT, with reductions of −1.54% and −0.2% assumed for AHCL and MDI plus CGM, respectively19. Although the ADAPT RCT compared the MiniMed 780G system with isCGM, it was assumed for the purpose of this analysis that the efficacy of real‐time CGM and isCGM in terms of HbA1c was equivalent; this was based on the findings of a 2024 meta‐analysis of clinical trials34.
Moreover, the effect of assumptions around the HbA1c treatment effect with CGM was explored in sensitivity analyses. For the base‐case, hypoglycemic event and DKA rates were obtained from a Singaporean cost‐effectiveness analysis comparing AHCL with standard care published by Gardner et al.17, which in turn were based on an average of values reported from three studies35, 36, 37. For MDI plus CGM, the rates were 1,941.68 per 100 patient years for non‐severe hypoglycemic events, 64.6 per 100 patient years for SHEs requiring non‐medical assistance, and 13 per 100 patient years for SHEs requiring medical assistance.
A rate of 2.93 events per 100 patient years was assumed for DKA. Hypoglycemia and DKA rates were assumed to be 50% lower with the MiniMed 780G system compared with MDI plus CGM17.
Utilities
Utility values used in this study are provided in Table S1. The baseline utility value for people with T1D free of complications was 0.896, based on a study that reported EQ‐5D‐5L utility values from a Japanese population including a sub‐population with T1D38. Health state utility values used in the base‐case analysis were primarily sourced from a study reported by Ishii et al.39, which converted EQ‐5D‐5L scores from people with diabetes in Japan to utility values using an algorithm from a Japanese cost‐effectiveness analysis40.
Disutilities for myocardial infarction, heart failure hospitalization, stroke, active ulcer, amputation, and severe hypoglycemic events were converted from European to Japanese values using a ratio of 0.66 based on previous observations that the utility values reported by people with diabetes in Japan experiencing complications tend to be higher than those reported outside of the country39. Disutilities for non‐severe hypoglycemic events were calculated by taking the ratio between non‐severe and severe event disutilities in the CDM default values and applying it to the severe hypoglycemic event disutility. For DKA, a disutility of 0.0367 was sourced from a Chinese study41.
In the base‐case, a utility benefit from reductions in fear of hypoglycemia (FoH) was incorporated with values of 0.0544 for the MiniMed 780G system and 0.0160 for MDI plus CGM based on data from the ADAPT study with a conversion to utility values using an equation reported by Currie et al.19, 42
Costs of complications
For treatments, costs including devices and consumables were sourced from the Japan National Health Insurance system43, 44. The costs of diabetes complications were obtained primarily from a 2018 cost‐effectiveness analysis of treatments for type 2 diabetes in Japan45. These costs were adjusted to 2024 values using a multiplier from the Medical Payment System Revision Rates published by the Ministry of Health, Labour and Welfare of Japan46.
In addition to these costs, severe hypoglycemia and DKA costs were sourced from a 2019 study focused on the medical costs and resource utilization associated with severe hypoglycemic events in Japan47. All costs associated with complications used in this analysis are listed in Table S2.
Perspective, time horizon and discounting
For the base‐case analysis, a time horizon of 55 years was used to reflect the remaining lifetime of the population being modeled, with a discount rate of 2.0% per annum applied to both future costs and clinical outcomes as per the Japanese HTA guidelines48. The analysis was carried out from the perspective of payers in Japan, with a willingness‐to‐pay (WTP) threshold of JPY 5,000,000 per quality‐adjusted life year (QALY) considered when assessing cost‐effectiveness49, 50.
Sensitivity and scenario analyses
A probabilistic sensitivity analysis and one‐way sensitivity analyses were used to investigate the impact of uncertainty in input parameters on cost‐effectiveness outcomes. In the one‐way sensitivity analyses, parameters varied included MiniMed 780G cost (±10%), time horizon (10, 25, and 40 years), complication costs (±10%), MiniMed 780G HbA1c effect magnitude (±10%), discount rate (0%, 4%), and FoH utilities (±10%). Additional simulations were also performed in which the FoH utility benefit was either removed completely or reduced by 50%.
Simulations were also performed in which a lower baseline HbA1c value of 7.8% was assumed, based on values reported in studies conducted in people with T1D in Japan by Sato et al.51 and Oikawa et al.9 Two different simulations were performed using the lower baseline HbA1c value. In the first one, a treatment effect of −1.54% was assumed for MiniMed 780G, in line with the base‐case. In the second, a lower treatment effect of −0.854% was used, which was calculated based on the findings of Konantambigi et al.52 The rationale for this study was that the analysis by Konantambigi et al. allows for a treatment effect to be estimated for different baseline HbA1c levels, based on real‐world data.
Further simulations were performed in which the HbA1c effect for MDI plus CGM was doubled, as well as simulations that involved the exclusion of DKA and non‐severe hypoglycemic events, the exclusion of all hypoglycemic events and DKA, use of a baseline utility value obtained from Ishii et al.39, and an alternative source for SHE rates (from a study conducted in Japan)53. Specifically, SHE rates were sourced from Iwase et al.53 and the overall event rate was divided into SHE1 and SHE2 using the same multiplier as per the base‐case (Table S3). NSHE rates were also calculated from SHE2 rates using the same multiplier as in the base‐case.
As per the base‐case, in this simulation acute event rates (SHE, NSHE, and DKA) were assumed to be 50% lower with MiniMed 780G versus the comparator.
Ethical approval
The analysis was performed using a computer simulation model and all input data were sourced from previously published studies (or used reasonable assumptions where no local data were available). As such, ethical approval was not required for this analysis.
RESULTS
Base‐case results
The base‐case results are displayed in Table 2. The use of the MiniMed 780G system was associated with a mean life expectancy gain of 1.51 years (95% confidence interval [95% CI]: 1.49–1.53) and a mean gain in quality‐adjusted life expectancy of 2.83 QALYs (95% CI: 2.81–2.84) compared with MDI plus CGM. Mean costs of treatment were JPY 16,226,167 higher for MiniMed 780G, but costs associated with the management of complications including severe hypoglycemia and adverse events were lower, resulting in a cost difference of JPY 13,350,942 (95% CI: 13,317,960–13,383,928).
This resulted in an ICER of JPY 4,723,991 (95% CI: 4,700,587–4,747,395) per QALY, which is below the WTP threshold.
The use of the MiniMed 780G system was also predicted to increase the time spent without complications compared with MDI plus CGM, as shown in Figure S1. The time spent free of any complications was estimated to be 2.94 years higher with MiniMed 780G (5.25 vs. 2.31 years).
When considering specific complications, all showed numerical increases in time free from them, with the largest observed for proliferative retinopathy (11.31 years), macular edema (10.83 years), microalbuminuria (10.40 years), severe vision loss (7.46 years), and gross proteinuria (7.29 years). As described above, these reductions in the incidences of complications were associated with decreases in the costs of managing them, with a JPY 1,290,471 reduction in the costs associated with renal complications, a JPY 799,829 reduction for cardiovascular disease (CVD) costs, a JPY 208,832 reduction for eye complications, and a combined JPY 115,489 decrease for ulcer/amputation/neuropathy (Table 2).
Sensitivity and scenario analyses
In the probabilistic sensitivity analysis, the MiniMed 780G system was found to have a 76.9% probability of being cost effective at a WTP threshold of JPY 5,000,000 per QALY (Figure 1).
Both baseline HbA1c and assumptions around FoH were key drivers of results. To assess the influence of baseline HbA1c two simulations were performed in which a lower baseline HbA1c of 7.8% was used, together with different treatment effects. Assuming a HbA1c reduction of 0.854% with MiniMed 780G, the ICER was JPY 7,053,369 per QALY; if the base‐case treatment effect of −1.54% was used the ICER was JPY 5,590,566 per QALY gained (Table S4).
Additionally, if the utility benefit associated with reduced FoH was removed completely, the resulting ICER was JPY 7,482,875 per QALY, which is substantially above the WTP threshold of JPY 5,000,000 per QALY. Similarly, if the FoH utility benefit with MiniMed 780G was reduced by 50%, the ICER was above the WTP threshold at JPY 6,343,094 per QALY. However, with only a 10% reduction in the FoH benefit with MiniMed 780G, the ICER was JPY 4,976,125 per QALY.
Use of alternative SHE rates, sourced from a study conducted in Japan, resulted in an ICER of JPY 5,125,784 per QALY gained (Table S3).
In other one‐way sensitivity analyses, the ICER was found to be sensitive to the exclusion of acute events, changes in MiniMed 780G costs, and discount rate (Figure 2). The ICER was found to decrease relative to the base‐case when the MiniMed 780G cost was lowered by 10% (JPY 878,106 decrease), discounting removed (JPY 595,456 decrease), the MiniMed 780G HbA1c treatment effect increased (JPY 195,410 decrease), complication costs raised by 10% (JPY 101,734 decrease), or an alternative baseline utility value used (sourced from Ishii et al.39) (JPY 83,413 decrease).
The time horizon of the analysis was also influential, with the incremental benefit in terms of quality‐adjusted life expectancy being less pronounced over shorter time horizons (resulting in ICERs exceeding JPY 5,000,000 per QALY over time horizons of 10 years and 25 years; Table S5).
DISCUSSION
This analysis shows the MiniMed 780G system was projected to be cost‐effective in the management of people with T1D with elevated HbA1c levels in Japan, with a base‐case ICER of JPY 4.7 million per QALY, which would be considered cost‐effective if using a WTP threshold of JPY 5 million per QALY. The MiniMed 780G system was found to be associated with a gain in quality‐adjusted life expectancy of 2.83 QALYs, plus an approximately 3‐year increase in the time spent free from any complications. Although the costs of treatment with the MiniMed 780G were higher, they were partly offset by savings associated with reductions in complication risk, particularly renal complications, CVD, and SHEs requiring medical assistance.
It should also be noted that, as the analysis was performed from a payer perspective, indirect costs associated with lost productivity due to complications were not included. Allied to this, even with a societal perspective, there are aspects of everyday life that are not captured in economic analyses, such as treatment burden, flexibility and confidence in daily life, sleep quality, and caregiver burden.
Sensitivity analyses revealed that the time horizon was an important driver of results, with MiniMed 780G being more cost‐effective over longer time horizons, likely due to the fact that, as a life‐long condition, the clinical benefits of MiniMed 780G in terms of delaying or preventing long‐term complications become more pronounced over longer time horizons. The importance of the time horizon observed here also raises the issue of whether earlier initiation of MiniMed 780G, including during childhood when appropriate, could further improve long‐term clinical outcomes. However, the present analysis did not assess cost‐effectiveness in a pediatric population, with the mean baseline age of the simulated cohort being 40.6 years.
Further analyses using both Japanese pediatric and adult real‐world data are needed to more fully investigate the effects of earlier initiation.
Sensitivity analyses also highlighted the importance of the reductions in hypoglycemia and DKA conferred by MiniMed 780G, as well as the data source of SHE rates, as drivers of cost‐effectiveness. Additionally, the QoL benefit associated with reduced FoH was shown to be influential. If this benefit was removed, or reduced by 50%, the incremental gain in quality‐adjusted life expectancy was reduced such that the ICER exceeded JPY 5,000,000 per QALY (JPY 7,482,875 per QALY when removed completely).
Additionally, in a simulation in which a lower baseline HbA1c was assumed, based on studies conducted in Japan, combined with a reduced treatment effect, the ICER was substantially higher than in the base‐case analysis (JPY 7,053,369 per QALY). This finding illustrates that baseline HbA1c, and the associated change in treatment effect, influenced cost‐effectiveness with MiniMed 780G being most cost‐effective in those with higher baseline HbA1c values. However, this simulation relied on a study conducted outside Japan for the estimated treatment effect.
Japan‐specific data for HbA1c treatment effects as well as acute event rates are still limited; further Japanese real‐world studies are needed to provide the evidence base for a more comprehensive and clinically grounded assessment of cost‐effectiveness in people with different baseline HbA1c values.
The results presented here are broadly consistent with other cost‐effectiveness analyses conducted in Greece, Singapore, Sweden, and the US, which have typically found the MiniMed 780G system to be cost‐effective when compared with MDI plus isCGM or other insulin therapies17, 54, 55, 56. Evidence of cost‐effectiveness within the Japanese healthcare context supports consideration of broader access to AID, not only for individuals who do not achieve recommended HbA1c targets with CGM alone, but for all individuals with an indication for AID to pursue better QoL.
Although the use of AIDs and CGM has been associated with numerous clinical benefits, there have been barriers to their widespread use in Japan in recent years. Studies have previously found that financial considerations are a key issue, with demand for insulin pumps steeply dropping as price increases, highlighting the need for reimbursement of such devices57, 58. Japan has a universal health insurance system, under which all residents are able to receive necessary medical care.
The standard co‐payment is generally 30%. When medical expenses exceed a certain threshold, the High‐Cost Medical Expense Benefit System covers the amount above the limit. However, because the threshold is income‐dependent and often remains high for many individuals, the practical benefit of this system can be limited for people using AID systems.
The Medical Expense Support Program for Children with Chronic Pediatric Diseases provides financial support for medical expenses, including diabetes management until 18 years of age, and may be extended until under 20 years of age when deemed necessary. However, the financial burden associated with continued use of AID systems becomes apparent after this age. In response, non‐governmental patient support initiatives—IDDM Network in Saga Prefecture and a voluntary healthcare professional (HCP) support in Okayama Prefecture have expanded financial assistance for AID‐related medical expenses up to 25 years of age59.
Additionally, in April 2026 and in response to advocacy efforts, the Tokyo Metropolitan Government introduced a targeted subsidy covering the costs of insulin pump therapy, including AID systems, during pregnancy, for Tokyo residents with T1D, representing an important step toward improving access to diabetes technology60. There is also potential for this support to be expanded in the future, including broader eligibility, longer duration of coverage, and the adoption of similar initiatives in other regions. However, although local‐level initiatives to support AID access are gradually expanding, many individuals still face substantial out‐of‐pocket costs2, 61.
Equitable access to diabetes technology and efforts to reduce disparities are essential and continued diabetes advocacy is needed to promote supportive healthcare systems and to strengthen education and training for HCPs, thereby enabling people with diabetes to benefit from advances in technology.
As the MiniMed 780G system has only been available in Japan since late 2023, comparative clinical studies are only now beginning to be published. For example, the authors of a 2026 study showed that switching to the MiniMed 780G system from the MiniMed 770G was associated with increases in TIR (71.6% to 76.1%), decreases in time‐above‐range (25.2% to 21.0%), decreases in nocturnal hypoglycemia (4.9% to 2.7%), as well as an improvement in HbA1c (7.2% to 7.0%)23. The authors noted that particular advantages of the MiniMed 780G system include not requiring routine calibration, flexibility in target glucose levels, and a SmartGuard feature that automatically adjusts and delivers correction insulin doses.
Additionally, improvements were seen in satisfaction with treatment and anxiety about treatment. In another 2026 study, also comparing MiniMed 780G with MiniMed 770G, Akiyama et al. reported improvements in HbA1c (7.5% to 7.0%), decreases in alarm frequency (11.9 to 7.1 times/day), and decreases in calibration frequency (3.6 to 0.8 times/day) over a 3‐month period22. MiniMed 780G has also been found to be well‐accepted and tolerated among children and adolescents in a small‐scale Japanese study (n = 29 people aged 2–18 years)18.
While this analysis demonstrates the cost‐effectiveness of the MiniMed 780G system, it has some inherent limitations. Firstly, the ADAPT RCT used to provide many of the treatment effect and baseline cohort parameters was conducted in Europe and did not include any people in Japan. This included the baseline HbA1c value, due to previous observations that HbA1c treatment effects tend to correlate with baseline HbA1c28, 29, 30.
As described above, the MiniMed 780G system has been shown to be effective in supporting glycemic management in Japanese users, even in populations with lower mean baseline HbA1c values22, 23. However, differences between European and Japanese populations in terms of factors such as insulin dose requirements, diabetes education, and healthcare systems were not accounted for. A second limitation was the use of short‐term data to predict long‐term outcomes due to the lack of existing studies of long‐term outcomes the use of MiniMed 780G.
Consequently, the ICER should be interpreted as an estimate rather than a precise value. The short‐term nature of the clinical data also means that factors such as long‐term adherence, device changes or discontinuations, or advances in technology occurring over the time horizon of the simulation are not accounted for. A further limitation was that the effectiveness of real‐time CGM in terms of HbA1c was assumed to be similar to that of isCGM, based on the findings of a meta‐analysis showing no significant difference in terms of HbA1c; although the same analysis did show a benefit in favor of real‐time CGM in terms of TIR34.
Future cost‐effectiveness analyses versus separate comparator arms for isCGM and real‐time CGM are warranted to explore the long‐term economic implications of the different types of CGM in greater detail. An additional limitation was a lack of consideration of indirect costs associated with T1D. This was largely due to a lack of recent data on such costs in Japan; again, another avenue for future research would be to consider long‐term cost‐effectiveness from both the payer and societal perspective.
In other countries, indirect costs have been found to be substantial contributors to the overall economic burden of T1D, suggesting that they may be a large factor in determining the overall economic impact of new interventions62, 63, 64.
Overall, this analysis demonstrates that the MiniMed 780G system is cost‐effective in the management of people with T1D in Japan with elevated HbA1c levels. The use of the MiniMed 780G system was also predicted to result in improvements in clinical outcomes, including increases in the time spent free from diabetes‐related complications. These improvements in outcomes partially offset the higher treatment costs associated with MiniMed 780G system use, resulting in a base‐case ICER below the Japanese WTP‐threshold of JPY 5,000,000 per QALY.
However, the ICER exceeded this threshold in some simulations performed using Japanese acute event rates, lower baseline HbA1c values, and shorter time horizons. These findings indicate that the estimated cost‐effectiveness is sensitive to the assumptions applied and should not be interpreted as a simple binary conclusion based solely on whether the ICER falls above or below the WTP threshold. These findings may also provide useful information to inform policymakers, as well as facilitate the establishment of robust support and reimbursement systems, with the aim of ultimately ensuring equitable access to this life‐changing device for all individuals in need across Japan.
FUNDING
Covalence Research Ltd received consulting fees from MiniMed in relation to the preparation of this manuscript.
DISCLOSURE
YH has received lecture fees from Eli Lilly Japan K.K., Novo Nordisk Pharma Ltd., Sanofi K.K., Sumitomo Pharma Co. Ltd., Dexcom Japan G.K., and Terumo Corp.; research expenses (including for contracted research, joint research, and clinical trials) from Sumitomo Pharma Co. Ltd., Nippon Boehringer Ingelheim Co.
Ltd., Sanofi K.K., Eli Lilly Japan K.K., and MiniMed Japan Co. Ltd. AT has received lecture fees from Sanofi, Abbott Japan, Terumo, Dexcom Japan, and Minimed Japan GK.
NK has received lecture fees from MiniMed Japan G.K., Abbott Japan LLC., Novo Nordisk Pharma Ltd., Sanofi K.K., Dexcom Japan G.K., and Terumo Corp; research expenses from Sanofi K.K., Abbott Japan LLC, and MiniMed Japan G.K.; advisory fee from MiniMed Japan G.K., and Novo Nordisk Pharma Ltd. MF and RFP are current employees of Covalence Research Ltd, which has received consulting fees from Medtronic in relation to the preparation of this manuscript. ZO, MH, and SP are employees of MiniMed.
MiniMed develops and manufactures diabetes technology such as the MM780G insulin pump and CGMs.
Approval of the research protocol: N/A.
Informed consent: N/A.
Registry and the registration no. of the study/trial: N/A.
Animal studies: N/A.
Supporting information