Improved Glycemia and Quality of Life With Open-Source AID: Real-World Data From 1840 Patient-Years
Ali Alqahtani1,2, Nabeel Khan1, Kate Hawke1 , Gerri Klein1 and Tom Elliott1,2
1 BC Diabetes, Vancouver, British Columbia, Canada. 2 The University of British Columbia, Vancouver, British Columbia, Canada.
Introduction
Automated insulin delivery (AID) systems have revolutionized Type 1 Diabetes management. However, access remains limited in many healthcare settings due to cost and regulatory barriers. Since 2020, BC Diabetes, a public Canadian diabetes clinic, has offered open-source AID (OS-AID) systems—Loop, iAPS, and AAPS—to eligible patients with ongoing clinical support & software at no cost to the client. These systems leverage widely available hardware (Dexcom CGM, Omnipod pump) and smartphone apps, enabling broader and equitable access to AID.
Methods
We conducted a retrospective cohort study of T1D patients using OS-AID at BC Diabetes between January 2020– April 2025. Eligible patients were Omnipod and Dexcom CGM users who provided informed consent to install an OS-AID app (Loop, iAPS, or AAPS) on their smartphones.
Data sources included 14-day ambulatory glucose profiles (AGPs) before and 90 days after OS-AID initiation. Safety endpoints (death, DKA, severe hypoglycemia) and patient-reported quality of life (QoL) outcomes (diabetes distress and device satisfaction surveys) were also evaluated. QoL metrics were assessed using validated questionnaires, with paired pre/post analysis.
Results
Out of 1442 clients (1840 patient-years), 30% were pediatric (ages 2–17) and 70% adults (ages ≥18), with a 46% male. Adults were mean age 42 years (range 18–85) and diabetes diabetes duration 22 years; pediatric clients were mean age 13 years and 6 years duration. App usage: Loop (78%), AAPS (13%), iAPS/Trio (9%).
Glycemia: Among 1,442 clients (1,013 adults and 429 pediatric clients), initiation of OS-AID was associated with significant improvements in glycemic control. Specifically, time in range (TIR, 3.9–10.0 mmol/L) increased, while mean glucose and Glycemic Management Indicator (GMI) decreased. Subgroup analyses revealed no statistically significant differences in glycemic outcomes across the three OS-AID systems (Loop, iAPS, and AAPS), supporting pooled analysis. Improvements were observed across all age groups and baseline TIR strata and were sustained over time.
Safety: Zero deaths were reported. There were four documented DKA episodes and fourteen patient-reported severe hypoglycemic events across the full cohort.
These findings suggest that OS-AID systems, when supported by an experienced team, offer safe, durable and reproducible improvements in glycemic outcomes in both pediatric and adult populations.
Conclusion
Open-source AID systems, when paired with structured clinical support, were safe and effective in improving glycemic outcomes and psychosocial wellbeing across pediatric and adult populations. These findings support the integration of OS-AID into clinical practice to promote equitable access to diabetes technology.
CGM metrics | Group | Baseline | 1M | 2M | 3M | 6M | 12M |
Time in range, 3.9-10 mmol/L(70-180 mg/dl) ,% | Adult | 58 | 73 | 72 | 72 | 71 | 70 |
Peds | 52 | 63 | 63 | 62 | 62 | 62 | |
Time Below Range,<3.9 mmol/L (70 mg/dl),% | Adult | 3 | 2 | 2 | 2 | 2 | 2 |
Peds | 3 | 2 | 2 | 2 | 2 | 2 | |
Time Above Range,>10 mmol/L (200 mg/dl),% | Adult | 38 | 25 | 26 | 26 | 27 | 27 |
Peds | 44 | 34 | 35 | 35 | 36 | 36 | |
Time Above Range,>14 mmol/L (250 mg/dl),% | Adult | 10 | 7 | 7 | 7 | 8 | 8 |
Peds | 14 | 13 | 13 | 13 | 14 | 14 | |
Glucose Management Indicator,% | Adult | 7.5 | 6.9 | 6.9 | 7.0 | 7.0 | 7.0 |
Peds | 7.8 | 7.3 | 7.3 | 7.3 | 7.4 | 7.4 | |
Average Glucose, mmol/L | Adult | 9.5 | 8.4 | 8.4 | 8.5 | 8.5 | 8.6 |
Peds | 10.3 | 9.2 | 9.3 | 9.3 | 9.4 | 9.4 | |
Standard Deviation, mmol/L | Adult | 3.3 | 2.9 | 2.9 | 3.0 | 3.0 | 3.0 |
Peds | 3.8 | 3.6 | 3.6 | 3.6 | 3.7 | 3.7 |
Quality of Life: We evaluated changes in the Diabetes Distress Scale (DDS), Device Satisfaction (DIDS DS), and INSPIRE scores pre and 3 months post system initiation. Among 279 adult users, mean DDS scores decreased from 2.47 ± 0.73 to 1.92 ± 0.63 (p < 0.0001), while DIDS DS scores improved from 6.37 ± 0.80 to 8.27 ± 1.30 (p < 0.0001), indicating reduced distress and increased device satisfaction. Similar patterns were observed in the parent/pediatric cohort (n=96), with DDS decreasing from 2.35 ± 0.67 to 1.97 ± 0.60 (p < 0.0001) and DIDS DS increasing from 6.39 ± 0.82 to 8.22 ± 1.20 (p < 0.0001). INSPIRE scores remained high post OS-AID (adults: 80.1 ± 13.6; parents: 83.4 ± 13.2), reflecting strong user confidence in OS-AID technology.
Table 1. Glycemic Metrics Before and After OS-AID Initiation at Multiple Time Points in Adults and Pediatric Cohorts
Figure 1. Nighttime Time in Range (12:00 AM – 6:00 AM) Before and 90-days After OS-AID Initiation
Figure 2. CGM Outcomes at 90 Days Post-OS-AID Across the Entire Cohort
Scan this QR code to access a calculator that estimates the likelihood of success and predicted outcomes with OS-AID based on our cohort data.