Real‐world glycaemic outcomes of automated insulin delivery in type 1 diabetes: A meta‐analysis

Abstract

AbstractAimTo evaluate the real‐world effectiveness of automated insulin delivery (AID) systems in patients with type 1 diabetes (T1D).Materials and MethodsPubMed, Embase, the Cochrane Library, and ClinicalTrials.gov were searched for studies published up until 2 March 2024. We included pragmatic randomized controlled trials (RCTs), cohort studies, and before–after studies that compared AID systems with conventional insulin therapy in real‐world settings and reported continuous glucose monitoring outcomes. Percent time in range (TIR; 3.9–10 mmol/L), time below range (TBR; <3.9 mmol/L), time above range (TAR; >10 mmol/L), and glycated haemoglobin (HbA1c) level were extracted. Data were summarized as mean differences (MDs) with 95% confidence interval.ResultsA total of 23 before–after studies (101 704 participants) were included in the meta‐analysis. AID systems were associated with an increased percentage of TIR (11.61%, 10.47 to 12.76; p < 0.001). The favourable effect of AID systems was consistently observed when used continuously for 6 (11.76%) or 12 months (11.33%), and in both children (12.16%) and adults (11.04%). AID systems also showed favourable effects on TBR (−0.53%, −0.63 to −0.42), TAR (−9.65%, −10.63 to −8.67) and HbA1c level (−0.42%, −0.47 to −0.37) when compared with previous treatments.ConclusionsSimilar improvements in glycaemic parameters were observed in real‐world settings in RCTs using AID systems in T1D. AID systems benefit both children and adults by increasing TIR for both short‐ and long‐term interventions.