Efficacy and Safety of Automated Insulin Delivery Systems in Children and Adolescents With Type 1 Diabetes: A Systematic Review of Randomized Controlled Trials

Automated insulin delivery (AID) systems integrate continuous glucose monitoring (CGM) with algorithmdriven insulin delivery and are increasingly used in children and adolescents with type 1 diabetes (T1D).However, the magnitude and consistency of glycemic benefit across pediatric age groups, together with the safety profile of these systems, warrant focused evaluation.We conducted a systematic review of randomized controlled trials evaluating AID systems in children and adolescents with T1D.PubMed/MEDLINE, Embase, Cochrane CENTRAL, and ClinicalTrials.govwere searched from inception through August 2026.Eligible studies compared AID with conventional insulin therapy or non-automated pump/CGM-based management and reported glycemic or safety outcomes.Risk of bias was assessed using the Cochrane Risk of Bias 2 tool.Findings were synthesized narratively without quantitative meta-analysis.Of 2,363 records identified, 11 randomized controlled trials met the eligibility criteria.Trials evaluated multiple AID platforms in participants aged 2-18 years, with intervention durations ranging from 12 to 104 weeks.AID consistently improved time in range ([TIR] 70-180 mg/dL).Individual trials reported TIR improvements ranging from approximately 7 to 12 percentage points.In one pediatric trial, TIR increased by 11 percentage points (95% CI, 7-14; P<0.001), while another demonstrated an 8.7-percentage-point increase (95% CI, 7.4-9.9;P<0.001) accompanied by a 0.4-percentage-point reduction in glycated hemoglobin.A sixmonth trial demonstrated a 6.7-percentage-point increase in TIR (95% CI, 2.7-10.8;P=0.002), with a concurrent 1.9-percentage-point reduction in time below range.Improvements were particularly pronounced during nighttime in several studies.Hyperglycemic exposure generally decreased, whereas hypoglycemia did not consistently increase.Severe hypoglycemia and diabetic ketoacidosis were uncommon, although the relatively small trial populations limited assessment of rare adverse events.Two trials were judged to have low overall risk of bias, while nine had some concerns; none was judged to have high risk of bias.AID systems consistently improve CGM-derived glycemic outcomes in children and adolescents with T1D, with the most reproducible benefit being increased TIR and reduced hyperglycemic exposure.Improvements in glycated hemoglobin were comparatively modest.Current randomized evidence does not demonstrate a consistent increase in hypoglycemia or diabetic ketoacidosis, although longer and larger studies are needed to characterize rare safety outcomes.AID should be regarded as an effective strategy for reducing glycemic burden rather than a fully autonomous treatment.Future research should emphasize long-term clinical and psychosocial outcomes, treatment burden, technology-related complications, and representation of diverse pediatric populations.