Abstract
Background
To evaluate the long-term glycemic outcomes of automated insulin delivery (AID) systems in young people (age ≤ 25 years) with type 1 diabetes.
Methods
We searched PubMed, Embase, the Cochrane Library, and ClinicalTrials.gov from inception to October 25, 2025. Clinical trials and real-world studies with intervention periods of ≥ 12 months were included. The primary outcomes were changes in time in range (TIR, 3.9–10.0 mmol/L) and HbA1c. Secondary outcomes included time in tight range (TITR, 3.9–7.8 mmol/L), time below/above range metrics (TBR1 < 3.9 mmol/L; TBR2 < 3.0 mmol/L; TAR1 > 10.0 mmol/L; TAR2 > 13.9 mmol/L), mean sensor glucose, glucose coefficient of variation (CV), and glucose standard deviation (SD). Data were pooled using random-effects models to calculate mean differences (MDs) with 95% confidence intervals (CIs).
Results
A total of 26 studies involving 16,596 participants were included. The use of AID systems significantly increased TIR by 11.56% (MD 11.56%, 95% CI 10.20 to 12.93; p < 0.001), equivalent to 166 additional minutes per day, and reduced HbA1c by 0.52% (MD: -0.52%; 95% CI: -0.63 to -0.40; p < 0.001). AID systems significantly increased TITR by 9.11% (MD 9.11%, 95% CI 7.59 to 10.63; p < 0.001). Significant reductions were also observed in TBR1 (MD -0.69%, 95% CI -1.33 to -0.06), TBR2 (MD -0.31%, 95% CI -0.54 to -0.08), TAR1 (MD -9.87%, 95% CI -11.56 to -8.17), and TAR2 (MD -5.21%, 95% CI -6.39 to -4.03). Mean sensor glucose decreased by 15.05 mg/dL (95% CI -17.99 to -12.11). Glucose CV was reduced by 0.99% (MD -0.99%, 95% CI -1.76 to -0.21) and glucose SD decreased by 5.80 mg/dL (MD -5.80 mg/dL, 95% CI -7.00 to -4.59).
Conclusions
AID systems provide substantial and sustained improvements in glycemic control over ≥ 12 months in young people with type 1 diabetes, supporting their long-term clinical implementation.
Trial registration
This study was registered on PROSPERO (CRD420251150271).
Supplementary Information
The online version contains supplementary material available at 10.1186/s13098-026-02194-0.
Introduction
Type 1 diabetes (T1D) imposes a substantial lifelong management burden on young people and their families [1]. Maintaining optimal glycemic control remains challenging despite advances in insulin formulations, delivery devices, and continuous glucose monitoring (CGM) technologies [2]. The American Diabetes Association recommends a target glycated hemoglobin (HbA1c) level of less than 7% for most pediatric patients [3], corresponding to more than 70% time in range (TIR, 3.9–10.0 mmol/L) as measured by CGM [4, 5]. The International Society for Pediatric and Adolescent Diabetes further recommends an HbA1c target of < 6.5% for individuals using AID systems under appropriate management conditions [6]. However, real-world data indicate that fewer than 20% of young people with T1D achieve these goals [7].
Automated insulin delivery (AID) systems represent a transformative technological advancement in diabetes management [8]. These systems integrate CGM, insulin pumps, and control algorithms to automatically modulate insulin infusion in response to real-time glucose levels [9]. Previous meta-analyses, including our own work, have demonstrated that AID systems significantly improve glycemic control in young people (age ≤ 25 years) with T1D compared to conventional treatment methods [10–14]. However, these analyses primarily focused on short-term outcomes (typically 3 months or less) from randomized controlled trials (RCTs) conducted in controlled settings.
The long-term effectiveness of AID systems in real-world clinical practice remain insufficiently investigated [15]. While RCTs provide evidence of efficacy under ideal conditions, real-world evidence (RWE) studies offer valuable insights into how these systems perform in diverse clinical settings with varied patient populations and practice patterns [16]. This is especially important for young people, a population with distinct physiological and behavioral characteristics that may influence technology use and outcomes [17]. Understanding the sustained benefits and potential risks of AID technology over extended periods is crucial for clinical decision making and patient education [18]. Given the limited evidence regarding long-term outcomes of AID systems in pediatric populations, we conducted this systematic review and meta-analysis to evaluate the glycemic outcomes of AID systems over extended periods (≥ 12 months) in young people with T1D by synthesizing evidence from both clinical trials and real-world studies.
Methods
Search strategy and study selection
This systematic review was prospectively registered on the International Prospective Register of Systematic Reviews (PROSPERO) under registration number CRD420251150271. We conducted a comprehensive literature search across multiple electronic databases, including PubMed, Embase, the Cochrane Central Register of Controlled Trials, and ClinicalTrials.gov, from inception until October 25, 2025. The search strategy incorporated a combination of Medical Subject Headings (MeSH) and keywords related to “type 1 diabetes”, “closed-loop system”, “automated insulin delivery”, “artificial pancreas”, “child”, and “adolescent”. The full search strategy is available in the Supplementary Material. No language restrictions were applied. Additionally, the reference lists of retrieved articles and relevant reviews were manually screened to identify additional eligible publications.
Studies were included if they met the following criteria: (1) participants were young people (age ≤ 25 years) diagnosed with T1D; (2) the intervention involved any AID system; (3) the comparator was conventional insulin therapy, such as multiple daily injections (MDI), insulin pump without automation (CSII), or sensor-augmented pump therapy (SAP) with or without predictive low-glucose suspend (PLGS); (4) reported outcomes included TIR or HbA1c; (5) the study design was a RCT, cohort study, or before-after study; and (6) the study duration was at least 12 months. We excluded studies with intervention periods shorter than 12 months, those with insufficient outcome data, studies involving fewer than 10 participants using AID. If multiple reports from the same study with different follow-up durations were identified, only the one with the longest follow-up was included to avoid duplication.
Outcome measures
The primary outcomes were the TIR (3.9–10.0 mmol/L) and HbA1c. Secondary glycemic outcomes included time in tight range (TITR; 3.9–7.8 mmol/L), time below range Level 1 (TBR1; <3.9 mmol/L), time below range Level 2 (TBR2; <3.0 mmol/L), time above range Level 1 (TAR1; >10.0 mmol/L), time above range Level 2 (TAR2; >13.9 mmol/L), mean sensor glucose, coefficient of variation (CV, %) of glucose, and standard deviation (SD, mg/dL) of glucose. The definitions of CGM-derived glycemic metrics were based on the international consensus recommendations [4].
Data extraction and quality assessment
Two reviewers (ZS and SY) independently screened titles and abstracts, and potentially relevant full-text articles were retrieved for further evaluation. Discrepancies were resolved through discussion or by consulting a third reviewer (ZS). Data were extracted using a standardized form, which included first author, publication year, country, study design, sample size, participant characteristics (e.g., age, sex, diabetes duration, baseline HbA1c), type of AID system and comparator, study duration, and outcome data.
The methodological quality of RCTs was assessed using the Cochrane Risk of Bias Tool (RoB 2) [19], while non-randomized studies were evaluated using the Risk of Bias in Non-randomized Studies of Interventions (ROBINS-I) tool [20].
Data synthesis and analysis
Meta-analyses were performed using DerSimonian and Laird random-effects models to calculate pooled mean differences (MDs) and 95% confidence intervals (CIs) for continuous outcomes. For dichotomous outcomes, we calculated pooled relative risks (RRs) with 95% CIs. When studies reported medians and interquartile ranges (IQRs) instead of means and standard deviations (SDs), we assumed the median to be equivalent to the mean and estimated SDs by dividing IQRs by 1.35, as recommended by the Cochrane Handbook [21]. For before-after studies that did not report 95% CI or standard error (SE) of MD, we calculated SEs from pre- and post-intervention values based on group means and SDs assuming a correlation coefficient of 0.5 [12, 22].
Heterogeneity among studies was assessed using the Cochran’s Q test and quantified with the I²statistic. I²values of 25%, 50%, and 75% were interpreted as indicating low, moderate, and high heterogeneity, respectively [23]. Subgroup analyses for TIR and HbA1c were conducted to explore potential sources of heterogeneity based on study design, type of AID system, study duration, type of comparator, age, and mean baseline HbA1c. Potential publication bias was assessed visually using funnel plots and statistically using Egger’s test when at least 10 studies were available for an outcome. All statistical analyses were conducted using Stata version 17.0 (StataCorp, College Station, TX).
Results
Literature search and study characteristics
The initial systematic literature search identified 2219 records from electronic databases and other sources. After removing 686 duplicates, 1533 records underwent title and abstract screening. Following this, 105 full-text articles were assessed for eligibility. Ultimately, 26 studies, comprising a total of 16,596 participants, met the inclusion criteria and were included [24–49]. Two studies that exclusively enrolled youth with elevated baseline HbA1c were excluded from the quantitative meta-analysis due to a high risk of bias [35, 39]. The detailed study selection process is illustrated in a PRISMA flow diagram (Figure S1).
The characteristics of the included studies are summarized in Table 1. Among the 26 studies, 2 were RCTs, 6 was single-arm trial, while 18 were RWE studies, including both prospective and retrospective designs. All RWE studies were uncontrolled before-after designs. The sample sizes ranged from 19 to 13,922 participants. Study duration varied from 12 to 48 months, with 6 studies conducted over 24 months or longer. The mean age of participants ranged from 2.4 to 21.0 years, and baseline HbA1c values ranged from 7.1% to 10.5%. Various AID systems were evaluated, including MiniMed 780G, Control-IQ, MiniMed 670G, CamAPS FX, Omnipod 5, and others. Conventional therapy comparators included SAP, CSII, MDI, and mixed insulin regimens.
Table 1.
Study characteristics and patient demographics of included studies
| Study | Participant characteristics* | Study design | Type of automated insulin delivery | Comparison/Prior therapy | No of patients | Country | Study duration |
|---|---|---|---|---|---|---|---|
| McVean 2023 [38] | Age 11.8 (2.8) [7–17] years; 64 male, 49 female; newly diagnosed; HbA1c 10.3% (1.4) | Parallel RCT, NCT04233034 | Control-IQ or MiniMed 670G or MiniMed 780G | CSII or MDI | 113 | USA | 12 months |
| Ware 2024a [49] | Age 14 (2) [10–16.9] years; 47 male, 34 female; diabetes duration NR; HbA1c 7.3% (1.2) | Parallel RCT (extension phase), NCT02871089 | FlorenceM or/and CamAPS FX | MDI | 81 | UK | 48 months |
| Ware 2024b [48] | Age 6.6 (1.5) [1–7] years; 23 male, 26 female; diabetes duration NR; HbA1c 7.1% (0.6) | Single-arm trial, NCT03784027 | CamAPS FX | SAP | 49 | Multiple countries (UK, Austria and Germany) | 18 months |
| Beato-Víbora 2022 [24] | Age ≤ 25 years; 49 male, 86 female; diabetes duration 21 (12) years; HbA1c 7.30% (0.89) | Prospective before–after RWE | MiniMed 780G | CSII (73.3%) or MDI (26.7%) | 135 | Spain | 12 months |
| Del Valle Rolón 2023 [28] | Median age 13.3 [IQR 9.9–15.5] years; 56 male, 59 female; median diabetes duration 5 (IQR 1.7–8) years; median HbA1c 7.8% (6.9–8.8) | Retrospective before–after RWE | Control-IQ | CSII (82%) or MDI (18%) | 115 | USA | 12 months |
| Varimo 2021 [47] | Age 9.7 (3.2) [3–16] years; 67 male, 44 female; diabetes duration 5.1 (2.5) years; HbA1c 7.4% (0.8) | Retrospective before–after RWE | MiniMed 670G | CSII (77.5%) or MDI (22.5%) | 111 | Finland | 12 months |
| Petrovski 2021 [42] | Age 10.2 (2.6) [7–18] years; 15 male, 15 female; diabetes duration 2.8 (1.7) years; HbA1c 8.2% (1.4) | Prospective before–after RWE | MiniMed 670G | MDI | 30 | Qatar | 12 months |
| Ng 2024 [40] | Age 12.3 (3.5) [2–19] years; 147 male, 104 female; diabetes duration 6.6 (3.7) years; HbA1c 7.9% (3.3) | Prospective before–after RWE | Control-IQ or MiniMed 780G or CamAPS FX | CSII or MDI | 251 | England | 12 months |
| Passanisi 2024 [41] | Median age 12.9 [7–18] years; 192 male, 176 female; median diabetes duration 5.2 years; median HbA1c 7.3% | Prospective before–after RWE | MiniMed 780G | CSII (66.6%) or MDI (33.4%) | 368 | Italy | 12 months |
| Henry 2024 [33] | Age range [10–25] years; NR diabetes duration NR; HbA1c 7.9% (IQR 7.7–8.0) | Retrospective before–after RWE | Control-IQ or MiniMed 780G | CSII | 48 | France | 12 months |
| Gruber 2024 [32] | Median age 15.1 [IQR 12.9–17.0] years; 63 male, 40 female; diabetes duration 6.6 (4.0) years; HbA1c 8.65% (0.12) | Prospective before–after RWE | MiniMed 780G | SAP | 93 | Israel | 12 months |
| Rabbone 2025 [44] | Age 7.9 (3.9) [1–18] years; 58 male, 26 female; diabetes duration NR; HbA1c 12% (2) | Prospective before–after RWE | MiniMed 780G or Tandem Control-IQ or MyLife Loop | PLGS or MID | 84 | Italy | 12 months |
| Ung 2025 [46] | Age 21.0 (3.0) [15–25] years; 40 male, 61 female; diabetes duration 12 (5) years; HbA1c 9.6% (2.3) | Retrospective before–after RWE | MiniMed 780G or Tandemt: slimX2 or DBLG1 (Diabeloop) | Before AID (not specified) | 101 | France | 12 months |
| De Meulemeester 2025 [27] | Age 12.0 (3.2) [6–18] years; 44 male, 70 female; diabetes duration 6.1 (3.6) years; HbA1c 7.8% (1.3) | Prospective before–after RWE | Control-IQ | SAP | 114 | Belgium | 12 months |
| Granados 2025 [31] | Age 7.9 (3.7) [2–22] years; 92 male, 82 female; diabetes duration 5.3 (0.4) years; HbA1c 8.04% (1.7) | Retrospective before–after RWE | Omnipod 5 | Before AID (not specified) | 174 | USA | 12 months |
| Karges 2025 [34] | Age 12.8 (3.9) [2–20] years; 7097 male, 6825 female; diabetes duration 6.2 (3.8) years; HbA1c 7.48% (1.1) | Prospective before–after RWE | AID (not specified) | SAP | 13,922 | Multiple countries | 18 months |
| Michaels 2024 [39] | Age 18.8 [13–25] years; 8 male, 12 female; diabetes duration 9.7 (5.4) years; HbA1c 10.5% (2.1) | Single-arm trial, ACTRN12621000556842 | MiniMed 780G | MDI | 20 | New Zealand | 12 months |
| Schneidewind 2025 [45] | Age 12.2 (3.8) [3.4–19.1] years; 60 male, 51 female; diabetes duration 6.1 (4.0) years; HbA1c 8.1% (1.1) | Retrospective before–after RWE | Control-IQ or MiniMed 780G or CamAPS FX | SAP (32%) or CSII + CGM (44%) or MDI + CGM (23%) | 111 | Germany | 24 months |
| Kiilavuori 2025 [35] | Age 11.7 (3.4) [7–16] years; 52 male, 27 female; diabetes duration 6.4 (3.4) years; HbA1c 8.3% (1.0) | Retrospective before–after RWE | MiniMed 780G | CSII (66%) or MDI (27%) or HCL (670G, 7%) | 79 | Finland | 24 months |
| López-López 2024 [36] | Age 2.4 (1.4) [1–6] years; 35 male, 26 female; diabetes duration NR; HbA1c 7.1% (0.8) | Retrospective before–after RWE | MiniMed 780G | MDI | 35 | Spain | 12 months |
| Franzone 202530 | Age 4.4 (1.6) [1–6] years; 26 male, 15 female; diabetes duration 1.6 (1.4) years; HbA1c 7.5% (1.1) | Retrospective before–after RWE | Control-IQ or MiniMed 780G | PLGS (68%) or MDI (22%) | 41 | Italy | 12 months |
| Pulkkinen 2024 [43] | Age 4.3 (1.3) [2–6] years; 18 male, 17 female; diabetes duration 2.3 (1.3) years; HbA1c 7.4% (0.7) | Single-arm trial, NCT04949022 | MiniMed 780G | AID (37%) or SAP (54%) or MDI (9%) | 35 | Finland | 18 months |
| Lührs 2025 [37] | Median age 12.0 [IQR 9.3–14.7] years; 168 male, 137 female; diabetes duration 5.3 (IQR 2.7–8.3) years; HbA1c 7.4% (IQR 6.8–8.2) | Retrospective before–after RWE | Control-IQ or MiniMed 780G or MiniMed 670G | PLGS (36%) or CSII (38%) or MDI (26%) | 305 | Germany | 24 months |
| Criego 2024 [26] | Age 10.4 (2.1) [6–13.9] years; 51 male, 59 female; diabetes duration 4.7 (2.6) years; HbA1c 7.7% (0.9) | Single-arm trial (extension phase), NCT04196140 | Omnipod 5 | CSII (89%) or MDI (12%) | 110 | USA | 24 months |
| DeSalvo 2024 [29] | Age 4.7 (1.0) [2–6] years; 46 male, 34 female; diabetes duration 2.3 (1.1) years; HbA1c 7.4% (1.0) | Single-arm trial (extension phase), NCT04476472 | Omnipod 5 | CSII (85%) or MDI (15%) | 80 | USA | 15 months |
| Bismuth 2024 [25] | Age 10.1 (1.7) [6.8–14] years; 70 male, 48 female; diabetes duration 5.2 (2.3) years; HbA1c 7.7% (0.7) | Single-arm trial (extension phase), NCT04476472 | Control-IQ | CSII | 118 | France | 24 months |
HbA1c units reported as mean or median in % (SD or IQR). AID=automated insulin delivery. SAP=sensor augmented pump therapy. CSII=continuous subcutaneous insulin infusion. PLGS=predictive low glucose suspend. MDI=multiple daily injections. RWE=real-world evidence. RCT=randomised controlled trial. IQR=interquartile range. *Data are mean (SD) unless otherwise indicated
One study reported data for two age subgroups (10–17 years and 18–25 years), which were treated as two separate comparisons in our analysis [33]. Similarly, the study by Lührs et al. was split into comparisons evaluating different AID systems [37]. Another study compared AID against both PLGS therapy and MDI; we also split this into two comparisons (AID vs. PLGS and AID vs. MDI) [44]. Consequently, a total of 26 comparisons from 24 studies were included in the final meta-analysis.
The risk of bias assessment for the included studies is provided in the Supplementary Table S1-S2. Two RCTs were rated as low risk. Among the non-randomized studies, two were rated as high risk and the remainder as moderate risk.
Primary outcomes
Meta-analysis of 22 studies (yielding 25 independent comparisons) reporting TIR demonstrated that AID system use significantly increased time in the TIR compared to control treatments (Fig. 1), with a pooled MD of 11.56% (95% CI: 10.20 to 12.93; p < 0.001), equivalent to approximately 166 additional minutes per day (95% CI: 147 to 186 min). Substantial heterogeneity was observed among the studies (I² = 82%). For HbA1c, pooling of 21 studies (23 comparisons) demonstrated a significant reduction with AID use (MD: −0.52%; 95% CI: −0.63 to −0.40; p < 0.001; Fig. 2). Heterogeneity was also substantial (I² = 93%). Leave-one-out sensitivity analyses confirmed the robustness of both findings, with no single study driving the overall estimates (Figure S2 and S3).
Fig. 1.
Forest plot for time in range (3.9–10 mmol/L)
Fig. 2.
Forest plot for HbA1c
Secondary outcomes
Other glycemic outcomes are summarized in Table 2. Use of AID systems significantly increased TITR by a MD of 9.11% (95% CI 7.59 to 10.63; p < 0.001; Figure S4), with low heterogeneity (I²=33%). AID systems also significantly significantly reduced TBR1 by an MD of −0.69% (95% CI: −1.33 to −0.06; p = 0.032; Fig. 3), equivalent to approximately 10 fewer minutes per day (95% CI: 0.9 to 19 min). For hyperglycemia, AID use reduced TAR1 by an MD of −9.87% (95% CI: −11.56 to −8.17; p < 0.001; Figure S5), equivalent to approximately 142 fewer minutes per day (95% CI: 118 to 166 min).
Table 2.
Results of primary and secondary outcomes
| Outcomes | Comparisons | MD (95% CI) | P | I2 | P interaction |
|---|---|---|---|---|---|
| TIR (3.9–10.0 mmol/L) | |||||
| All comparisons | 25 | 11.56 (10.2 to 12.93) | < 0.001 | 82% | |
| Study design | 0.386 | ||||
| RCT | 2 | 14.67 (9.77 to 19.57) | < 0.001 | 0 | |
| RWE | 18 | 11.65 (9.89 to 13.41) | < 0.001 | 85% | |
| Single-arm trial | 5 | 10.84 (8.47 to 13.21) | < 0.001 | 81% | |
| Type of AID | 0.006 | ||||
| Control-IQ | 3 | 12.86 (11.47 to 14.24) | < 0.001 | 0 | |
| FlorenceM or CamAPS FX | 2 | 8.54 (6.87 to 10.21) | < 0.001 | 0 | |
| MiniMed 670G | 2 | 11.57 (9.81 to 13.33) | < 0.001 | 0 | |
| MiniMed 780G | 5 | 12.38 (9.78 to 14.99) | 71% | ||
| Omnipod 5 | 2 | 11.83 (8.4 to 15.26) | 68% | ||
| Mixed or others | 11 | 11.41 (8.43 to 14.38) | < 0.001 | 90% | |
| Study duration | 0.290 | ||||
| < 12 months | 20 | 12.08 (10.81 to 13.34) | < 0.001 | 67% | |
| ≥24 months | 5 | 9.96 (6.27 to 13.66) | < 0.001 | 94% | |
| Comparator | 0.375 | ||||
| CSII or MDI | 14 | 12.4 (11.72 to 13.09) | < 0.001 | 0 | |
| SAP | 5 | 12.04 (8.83 to 15.25) | < 0.001 | 85% | |
| Mixed or others | 6 | 9.64 (5.08 to 14.2) | < 0.001 | 93% | |
| Age | NA | ||||
| ≤ 7 years | 6 | 9.07 (7.82 to 10.33) | < 0.001 | 0 | |
| ≤ 19 years | 16 | 11.65 (10.59 to 12.7) | < 0.001 | 53% | |
| Mean HbA1c (%) | 0.329 | ||||
| ≥8.0% | 6 | 13.74 (8.48 to 19.01) | < 0.001 | 89% | |
| <8.0% | 19 | 11.04 (9.74 to 12.35) | < 0.001 | 79% | |
| HbA1c (%) | |||||
| All comparisons | 23 | −0.52 (−0.63 to −0.4) | < 0.001 | 93% | |
| Study design | 0.105 | ||||
| RCT | 2 | −0.76 (−1.1 to −0.41) | < 0.001 | 0 | |
| RWE | 16 | −0.54 (−0.69 to −0.4) | < 0.001 | 36% | |
| Single-arm trial | 5 | −0.43 (−0.52 to −0.34) | < 0.001 | 95% | |
| Type of AID | 0.521 | ||||
| Control-IQ | 3 | −0.6 (−0.75 to −0.45) | < 0.001 | 50% | |
| FlorenceM or CamAPS FX | 2 | −0.55 (−0.99 to −0.1) | 0.016 | 54% | |
| MiniMed 670G | 3 | −0.26 (−0.8 to 0.29) | 0.352 | 94% | |
| MiniMed 780G | 4 | −0.48 (−0.78 to −0.19) | 0.001 | 91% | |
| Omnipod 5 | 2 | −0.46 (−0.58 to −0.34) | < 0.001 | 0 | |
| Mixed or others | 9 | −0.66 (−0.91 to −0.4) | < 0.001 | 94% | |
| Study duration | 0.013 | ||||
| < 12 months | 17 | −0.63 (−0.77 to −0.48) | < 0.001 | 94% | |
| ≥24 months | 6 | −0.27 (−0.51 to −0.03) | 0.029 | 93% | |
| Comparator | 0.791 | ||||
| CSII or MDI | 11 | −0.49 (−0.61 to −0.37) | < 0.001 | 79% | |
| SAP | 5 | −0.61 (−0.95 to −0.27) | < 0.001 | 95% | |
| Mixed or others | 7 | −0.5 (−0.87 to −0.12) | 0.01 | 95% | |
| Age | NA | ||||
| ≤ 7 years | 5 | −0.41 (−0.6 to −0.23) | < 0.001 | 74% | |
| ≤ 19 years | 17 | −0.49 (−0.59 to −0.39) | < 0.001 | 77% | |
| Mean HbA1c (%) | < 0.001 | ||||
| ≥8.0% | 8 | −0.92 (−1.18 to −0.66) | < 0.001 | 79% | |
| <8.0% | 15 | −0.35 (−0.47 to −0.24) | < 0.001 | 93% | |
| Secondary outcomes | |||||
| TTIR (3.9–7.8 mmol/L) | 8 | 9.11 (7.59 to 10.63) | < 0.001 | 33% | |
| TBR1 (< 3.9 mmol/L) | 22 | −0.69 (−1.33 to −0.06) | 0.032 | 96% | |
| TBR2 (< 3.0 mmol/L) | 17 | −0.31 (−0.54 to −0.08) | 0.009 | 95% | |
| TAR1 (> 10 mmol/L) | 20 | −9.87 (−11.56 to −8.17) | < 0.001 | 77% | |
| TAR2 (> 13.9 mmol/L) | 15 | −5.21 (−6.39 to −4.03) | < 0.001 | 75% | |
| Mean sensor glucose (mg/dL) | 17 | −15.05 (−17.99 to −12.11) | < 0.001 | 84% | |
| CV of glucose (%) | 18 | −0.99 (−1.76 to −0.21) | 0.013 | 91% | |
| SD of glucose (mg/d) | 10 | −5.80 (−7.00 to −4.59) | < 0.001 | 62% |
TAR=time above range. TBR=time below range. TIR=time in range. RWE=real-world evidence. RCT=randomised controlled trial. SAP=sensor augmented pump therapy. CSII=continuous subcutaneous insulin infusion. MDI=multiple daily insulin injection. MD=mean difference. CV=coefficient of variation (%). SD=standard deviation. TTIR=time in tight range
Fig. 3.
Forest plot for time below range (< 3.9 mmol/L)
AID systems reduced TBR2 by an MD of −0.31% (95% CI: −0.54 to −0.08; p = 0.009; Figure S6) and TAR2 by an MD of −5.21% (95% CI: −6.39 to −4.03; p < 0.001; Figure S7). AID systems were also associated with a significant reduction in mean sensor glucose (MD: −15.05 mg/dL; 95% CI: −17.99 to −12.11; p < 0.001; Figure S8). Additionally, AID systems significantly improved glycemic variability parameters, with a reduction in glucose CV by 0.99% (95% CI −1.76 to −0.21; p = 0.013; Figure S9) and glucose SD by 5.80 mg/dL (95% CI −7.00 to −4.59; p < 0.001; Figure S10).
Subgroup analyses
Subgroup analyses for TIR and HbA1c (25 comparisons) were performed to explore potential sources of heterogeneity (Table 2). The improvement in TIR was consistent across most subgroups, including study design (RCT, RWE, or single-arm trial), type of AID system (MiniMed 780G, Control-IQ, MiniMed 670G, Omnipod 5, CamAPS FX, and mixed or others), study duration (≥ 24 months or < 24 months), type of comparator (CSII or MDI, SAP, and mixed or others), age (≤ 7 years or ≤ 19 years), and mean baseline HbA1c (≥ 8.0% or < 8.0%).
For HbA1c, subgroup analyses revealed significant interactions for study duration (pinteraction= 0.013) and baseline HbA1c level (pinteraction< 0.001). The reduction in HbA1c seemed to be greater in studies of shorter duration (< 24 months: MD −0.63% vs. ≥24 months: MD −0.27%) and in those enrolling participants with higher baseline HbA1c (≥ 8.0%: MD −0.92% vs. <8.0%: MD −0.35%).
Publication bias
For the primary outcomes, visual inspection of funnel plots and Egger’s test indicated no significant publication bias for TIR (Figure S11; p = 0.453). However, significant publication bias was detected for HbA1c (Figure S12; p < 0.001). Among the secondary outcomes, statistically significant publication bias was suggested for mean sensor glucose (Figure S13; p = 0.038), but not for the other metrics (Figure S14-S20).
Discussion
This systematic review and meta-analysis of 26 studies demonstrates that AID systems significantly improve long-term glycemic outcomes in young people with T1D. The use of AID systems was associated with an 11.56% increase in TIR, equivalent to approximately 166 additional minutes per day, and a clinically meaningful 0.52% reduction in HbA1c, while simultaneously reducing time in hypoglycemia and hyperglycemia. The demonstration of significant benefit for both key metrics in the overall analysis and across most predefined subgroups underscores the robustness of these findings.
Our findings extend the evidence from previous meta-analyses that primarily focused on short-term outcomes [10–12]. The magnitude of TIR improvement observed in our study is remarkably consistent with previous reports in children and adolescents [12, 13], suggesting that the glycemic benefits of AID systems are not only substantial but also sustainable over extended periods. This consistency across different study durations strengthens the clinical validity of AID technology for long-term diabetes management in young people.
The glycemic outcomes demonstrated a high degree of consistency between clinical trials and RWE studies, indicating that the efficacy observed in clinical trial settings translates effectively to routine clinical practice. Similarly, the HbA1c reduction was consistent across different study designs and patient populations. This convergence of evidence from different study designs provides robust support for the effectiveness of AID systems in diverse clinical settings.
The glycemic benefits observed in our analysis are particularly relevant for youth with elevated baseline HbA1c. While our study did not specifically focus on this population, evidence from targeted studies suggests that AID systems offer substantial benefits to individuals with elevated HbA1c levels [39, 50]. These findings align with our subgroup analysis showing greater HbA1c reductions in youth with higher baseline HbA1c, indicating that AID systems may be particularly effective in higher-risk populations who have the most room for improvement. This supports prioritizing AID system access for youth with elevated baseline HbA1c to improve long-term outcomes.
The safety profile of AID systems requires careful consideration in the future. While two RCTs showed no significant difference in severe hypoglycemia or diabetic ketoacidosis (DKA) rates between AID and conventional therapy [38, 49], data from a large cohort study indicated a potentially increased risk of diabetic ketoacidosis with closed-loop therapy [34]. This finding underscores the importance of ongoing education and monitoring, particularly during the transition to AID systems. However, the overall benefits in glycemic control were maintained across both study designs. Future research to reduce DKA risk in AID users may include integrating ketone sensors into AID systems, developing algorithmic alerts for prolonged hyperglycemia unresponsive to AID, and establishing standardized sick-day management protocols with targeted user education on pump disconnection and illness-related risks.
Several limitations should be acknowledged. First, substantial heterogeneity was observed for the primary outcomes, which may be attributable to differences in study design, patient characteristics, AID systems, comparator treatments, and study duration across studies. Second, all real-world studies were uncontrolled before-after designs, which are susceptible to confounding. Third, the generalizability of findings to low-income countries may be limited, as nearly all of the included studies were conducted in high-income countries. Finally, the long-term cost-effectiveness of AID systems requires further evaluation.
Our findings that various AID systems produced comparable long-term glycemic benefits are supported by a recent network meta-analysis, which found that hybrid, advanced hybrid, and full closed-loop systems all significantly improved TIR compared to conventional insulin therapy [51]. This consistency across system types suggests that the core closed-loop principle, rather than a specific algorithm, is the key driver of glycemic improvement, reassuring clinicians that the clinical benefits of AID technology are robust across different commercial systems.
The long-term clinical benefits demonstrated in our analysis must be considered alongside economic implications for widespread AID implementation. Recent cost-effectiveness analyses specifically focused on pediatric populations provide compelling evidence. In Australia, HCL therapy showed favorable cost-effectiveness with an incremental cost-effectiveness ratio of AUD $32,789 per QALY gained [52]. Similarly, Swedish analyses found the Control-IQ system to be a dominant strategy compared with multiple daily injections or conventional pump therapy across all time horizons [53]. Most notably, a binational trial of the Cambridge hybrid closed-loop system projected cost-effectiveness below the £20,000/QALY threshold in the UK and potential cost savings in the US setting [54]. These consistent findings across different healthcare systems and AID technologies suggest that the improved glycemic control observed in our meta-analysis may translate into economically efficient long-term outcomes for pediatric populations. While initial acquisition costs remain substantial, the converging evidence indicates that AID systems represent good value for money in youth with T1D through prevention of complications and improved quality of life.
Conclusions
This comprehensive systematic review and meta-analysis provides robust evidence that AID systems maintain significant glycemic benefits for young people with T1D in the long-term. The consistent outcomes observed across clinical trials and RWE studies support the long-term use of AID technology in diabetes management for young people.
Supplementary Information
Author contributions
B.Z. and D.L. conceived and designed the study. Z.S., and S.Y. undertook the literature review and extracted the data. B.Z., S.Y., and L.G. coded the statistical analysis, figures, and appendix. Z.S., B.Z., and L.G. interpreted the data and wrote the first draft of the manuscript. All authors reviewed and revised subsequent drafts and approved the final version. B.Z. 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 the data and the accuracy of the data analysis.
Funding
This work was supported by the Tianjin Key Medical Discipline Construction Project (TJYXZDXK-3-003D) and National Key Clinical Specialty Construction Project for Emergency Medicine (No. 2023283).
Data availability
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Declarations
Ethics approval and consent to participate
Not applicable.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Zhihui Sun and Sihan Yang served as co-first authors.
Contributor Information
Le Gao, Email: gaole0513@connect.hku.hk.
Lin Dou, Email: doulint@163.com.
Baoqi Zeng, Email: zengbaoqi@126.com.
References
- 1.DiMeglio LA, Evans-Molina C, Oram RA. Type 1 diabetes. Lancet (London, England). 2018;391(10138):2449–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Miller KM, Foster NC, Beck RW, et al. Current state of type 1 diabetes treatment in the U.S.: updated data from the T1D Exchange clinic registry. Diabetes Care. 2015;38(6):971–8. [DOI] [PubMed] [Google Scholar]
- 3.Association AD. Children and adolescents: standards of medical care in Diabetes-2020. Diabetes Care. 2020;43(Suppl 1):S163-s82. [DOI] [PubMed] [Google Scholar]
- 4.Battelino T, Danne T, Bergenstal RM, et al. Clinical targets for continuous glucose monitoring data interpretation: recommendations from the International Consensus on Time in Range. Diabetes Care. 2019;42(8):1593–603. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Vigersky RA, McMahon C. The Relationship of Hemoglobin A1C to Time-in-Range in Patients with Diabetes. Diabetes Technol Ther. 2019;21(2):81–5. [DOI] [PubMed] [Google Scholar]
- 6.de Bock M, Agwu JC, Deabreu M, et al. ISPAD Clinical Practice Consensus Guidelines 2024: Glycemic Targets. Hormone Res Paediatrics. 2024;97(6):546–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Foster NC, Beck RW, Miller KM, et al. State of type 1 diabetes management and outcomes from the T1D Exchange in 2016–2018. Diabetes Technol Ther. 2019;21(2):66–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Haidar A. The Artificial Pancreas: How Closed-Loop Control Is Revolutionizing Diabetes. IEEE Control Syst. 2016;36:28–47. [Google Scholar]
- 9.Moon SJ, Jung I, Park CY. Current advances of artificial pancreas systems: a comprehensive review of the clinical evidence. Diabetes Metab J. 2021;45(6):813–39. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Bekiari E, Kitsios K, Thabit H, et al. Artificial pancreas treatment for outpatients with type 1 diabetes: systematic review and meta-analysis. BMJ (Clinical Res ed). 2018;361:k1310. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Weisman A, Bai JW, Cardinez M, Kramer CK, Perkins BA. Effect of artificial pancreas systems on glycaemic control in patients with type 1 diabetes: a systematic review and meta-analysis of outpatient randomised controlled trials. The lancet Diabetes & endocrinology. 2017;5(7):501–12. [DOI] [PubMed] [Google Scholar]
- 12.Zeng B, Gao L, Yang Q, Jia H, Sun F. Automated insulin delivery systems in children and adolescents with type 1 diabetes: a systematic review and meta-analysis of outpatient randomized controlled trials. Diabetes care. 2023;46(12):2300–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.de Visser HS, Waraich S, Chhabra M, et al. Automated insulin delivery systems and glucose management in children and adolescents with Type 1 Diabetes: a systematic review and meta-analysis. JAMA Pediatr. 2025;179(11):1162–71 10.1001/jamapediatrics.2025.2740 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Godoi A, Reis Marques I, Padrão EMH, et al. Glucose control and psychosocial outcomes with use of automated insulin delivery for 12 to 96 weeks in type 1 diabetes: a meta-analysis of randomised controlled trials. Diabetol Metab Syndr. 2023;15(1):190. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Phillip M, Nimri R, Bergenstal RM, et al. Consensus recommendations for the use of automated insulin delivery technologies in clinical practice. Endocr Rev. 2023;44(2):254–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Chodankar D. Introduction to real-world evidence studies. Perspect Clin Res. 2021;12(3):171–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Kimbell B, Lawton J, Boughton C, Hovorka R, Rankin D. Parents’ experiences of caring for a young child with type 1 diabetes: a systematic review and synthesis of qualitative evidence. BMC Pediatr. 2021;21(1):160. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Considine EG, Sherr JL. Real-World Evidence of Automated Insulin Delivery System Use. Diabetes Technol Ther. 2024;26:53–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Sterne JAC, Savović J, Page MJ, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ (Clinical Res ed). 2019;366:l4898. [DOI] [PubMed] [Google Scholar]
- 20.Sterne JA, Hernán MA, Reeves BC, et al. ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. BMJ (Clinical Res ed). 2016;355:i4919. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Higgins JP, Thomas J, Chandler J. Cochrane handbook for systematic reviews of interventions. John Wiley & Sons; 2011. [Google Scholar]
- 22.Yang Q, Zeng B, Hao J, Yang Q, Sun F. Real-world glycaemic outcomes of automated insulin delivery in type 1 diabetes: a meta-analysis. Diabetes Obes Metab. 2024;26(9):3753–63. [DOI] [PubMed] [Google Scholar]
- 23.von Hippel PT. The heterogeneity statistic I(2) can be biased in small meta-analyses. BMC Med Res Methodol. 2015;15:35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Beato-Víbora PI, Ambrojo-López A, Fernández-Bueso M, Gil-Poch E, Arroyo-Díez JF. Long-term outcomes of an advanced hybrid closed-loop system: a focus on different subpopulations. Diabetes Res Clin Pract. 2022. 10.1016/j.diabres.2022.110052. [DOI] [PubMed] [Google Scholar]
- 25.Bismuth É, Tubiana-Rufi N, Rynders CA, et al. Sustained 3-year improvement of glucose control with hybrid closed loop in children with type 1 diabetes while going through puberty. Diabetes Care. 2024;47(9):1696–703. [DOI] [PubMed] [Google Scholar]
- 26.Criego AB, Carlson AL, Brown SA, et al. Two years with a tubeless automated insulin delivery system: a single-arm multicenter trial in children, adolescents, and adults with type 1 diabetes. Diabetes Technol Ther. 2024;26(1):11–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.De Meulemeester J, Valgaerts L, Massa G, et al. Real-world glycemic and person-reported outcomes after Tandem Control-IQ initiation in children with Type 1 Diabetes. J Clin Endocrinol Metab. 2025. 10.1210/clinem/dgaf265. [DOI] [PubMed] [Google Scholar]
- 28.Del Valle Rolón ME, Brown EA, Wolf RM. Real-World Glycemic Improvements with Hybrid Closed Loop Pumps in Youth with Type 1 Diabetes. Pediatric Diabetes. 2023; 2023. [DOI] [PMC free article] [PubMed]
- 29.DeSalvo DJ, Bode BW, Forlenza GP, et al. Glycemic Outcomes Persist for up to 2 Years in Very Young Children with the Omnipod(®) 5 Automated Insulin Delivery System. Diabetes Technol Ther. 2024;26(6):383–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Franzone D, Spacco G, Piano A, et al. Real-world evaluation of safety and efficacy of AHCL systems in young children with type 1 diabetes: a 1-year assessment. Front Endocrinol. 2025;16:1590964. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Granados A, Castellanos JO, Sanchez AM, Giraldo MJ, Carrillo-Iregui A. Assessing the efficacy of a hybrid closed loop system in a racial-ethnic minority cohort of children and adolescents with type 1 diabetes. J Pediatr Endocrinol Metabolism. 2025;38(4):340–4. [DOI] [PubMed] [Google Scholar]
- 32.Gruber N, Wittenberg A, Brener A, et al. Real-life achievements of MiniMed 780G advanced closed-loop system in youth with type 1 diabetes: AWeSoMe Study Group multicenter prospective trial. Diabetes Technol Ther. 2024;26(11):869–80. [DOI] [PubMed] [Google Scholar]
- 33.Henry Z, Villar Fimbel S, Bendelac N, Perge K, Thivolet C. Beneficial effects of automated insulin delivery over one-year follow-up in real life for youths and adults with type 1 diabetes irrespective of patient characteristics. Diabetes Obes Metab. 2024;26(2):557–66. [DOI] [PubMed] [Google Scholar]
- 34.Karges B, Rosenbauer J, Stahl-Pehe A, et al. Hybrid closed-loop insulin therapy and risk of severe hypoglycaemia and diabetic ketoacidosis in young people (aged 2–20 years) with type 1 diabetes: a population-based study. Lancet Diabetes Endocrinol. 2025;13(1):88–96. [DOI] [PubMed] [Google Scholar]
- 35.Kiilavuori M, Varimo T, Tuomaala AK, Pulkkinen MA. Children and adolescent with suboptimal control of type 1 diabetes improve during the first 2 years on automated insulin delivery system. Diabetes Obes Metab. 2025;27(1):134–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.López-López S, Díaz-Martín C, García-de Pablo I, et al. Real world efficacy and safety of the advanced hybrid closed-loop system MiniMed 780G (SmartGuard) in children under 7 years of age. Front Med (Lausanne). 2024;11:1465800. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Lührs T, Kordonouri O, von dem Berge T, et al. Long-Term Glycemic Outcomes in Pediatric Type 1 Diabetes: A 2-Year Real-World Analysis of Different Automated Insulin Delivery Systems Generations. Diabetes technology & therapeutics; 2025. [DOI] [PubMed]
- 38.McVean J, Forlenza GP, Beck RW, et al. Effect of Tight Glycemic Control on Pancreatic Beta Cell Function in Newly Diagnosed Pediatric Type 1 Diabetes: A Randomized Clinical Trial. JAMA. 2023;329(12):980–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Michaels VR, Boucsein A, Watson AS, et al. Glucose and psychosocial outcomes 12 months following transition from multiple daily injections to advanced hybrid closed loop in youth with type 1 diabetes and suboptimal glycemia. Diabetes Technol Ther. 2024;26(1):40–8. [DOI] [PubMed] [Google Scholar]
- 40.Ng SM, Wright NP, Yardley D, et al. Long-term assessment of the NHS hybrid closed-loop real-world study on glycaemic outcomes, time-in-range, and quality of life in children and young people with Type 1 Diabetes. BMC Med. 2024;22(1):175. 10.1186/s12916-024-03396-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Passanisi S, Salzano G, Bombaci B, et al. Sustained Effectiveness of an Advanced Hybrid Closed-Loop System in a Cohort of Children and Adolescents With Type 1 Diabetes: A 1-Year Real-World Study. Diabetes Care. 2024;47(6):1084–91. [DOI] [PubMed] [Google Scholar]
- 42.Petrovski G, Al Khalaf F, Campbell J, et al. One-year experience of hybrid closed-loop system in children and adolescents with type 1 diabetes previously treated with multiple daily injections: drivers to successful outcomes. Acta Diabetol. 2021;58(2):207–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Pulkkinen MA, Varimo TJ, Hakonen ET, Hero MT, Miettinen PJ, Tuomaala AK. During an 18-month course of automated insulin delivery treatment, children aged 2 to 6 years achieve and maintain a higher time in tight range. Diabetes Obes Metab. 2024;26(6):2431–8. [DOI] [PubMed] [Google Scholar]
- 44.Rabbone I, Savastio S, Castorani V, et al. The Importance of Instigating Automated Insulin Delivery Systems at Onset of Type 1 Diabetes: 1-Year Follow-Up of Children and Adolescents from Two Tertiary Pediatric Diabetes Centers. Diabetes technology & therapeutics; 2025. [DOI] [PubMed]
- 45.Schneidewind J, von Sengbusch S, Schmidt D, Lange K. Human factor in the use of automated insulin delivery systems: real-world effects in children with type 1 diabetes after 2 years of use (AID-A study). Diabetes Res Clin Pract. 2025;229:9. [DOI] [PubMed] [Google Scholar]
- 46.Ung M, Penfornis A, Valentim C, Franc S, Amadou C, Eroukhmanoff J. Hybrid Closed Loop in Adolescents and Young Adults Living with Type 1 Diabetes: A Real-World Study. Diabetes technology & therapeutics; 2025. [DOI] [PubMed]
- 47.Varimo T, Pulkkinen MA, Hakonen E, Hero M, Miettinen PJ, Tuomaala AK. First year on commercial hybrid closed-loop system—experience on 111 children and adolescents with type 1 diabetes. Pediatr Diabetes. 2021;22(6):909–15. [DOI] [PubMed] [Google Scholar]
- 48.Ware J, Allen JM, Boughton CK, et al. Eighteen-Month Hybrid Closed-Loop Use in Very Young Children With Type 1 Diabetes: A Single-Arm Multicenter Trial. Diabetes Care. 2024;47(12):2189–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Ware J, Boughton CK, Allen JM, et al. Effect of 48 Months of Closed-Loop Insulin Delivery on Residual C-Peptide Secretion and Glycemic Control in Newly Diagnosed Youth With Type 1 Diabetes: A Randomized Trial. Diabetes Care. 2024;47(8):1441–48. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Marks BE, Grundman JB, Meighan S, Monaghan M, Streisand R, Perkins A. Hybrid closed loop systems improve glycemic control and quality of life in historically minoritized youth with diabetes. Diabetes Technol Ther. 2024;26(3):167–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Stahl-Pehe A, Shokri-Mashhadi N, Wirth M, et al. Efficacy of automated insulin delivery systems in people with type 1 diabetes: a systematic review and network meta-analysis of outpatient randomised controlled trials. Eclinicalmedicine. 2025;82:103190. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Pease A, Callander E, Zomer E, et al. The cost of control: costeffectiveness analysis of hybrid closed-loop therapy in youth. Diabetes Care. 2022;45(9):1971–80. [DOI] [PubMed] [Google Scholar]
- 53.Adolfsson P, Heringhaus A, Sjunnesson K, Mehkri L, Bolin K. Cost-effectiveness of the tandem t: Slim X2 with control-IQ technology automated insulin delivery system in children and adolescents with type 1 diabetes in Sweden. Diabet medicine. J Br Diabet Association. 2024;41(11):e15432. [DOI] [PubMed] [Google Scholar]
- 54.Fox DS, Ware J, Boughton CK, et al. Cost-Effectiveness of Closed-Loop Automated Insulin Delivery Using the Cambridge Hybrid Algorithm in Children and Adolescents with Type 1 Diabetes: Results from a Multicenter 6-Month Randomized Trial. J Diabetes Sci Technol. 2025;19(4):982–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.