Continuing automated insulin delivery systems during hospitalisation with type 1 diabetes: A thematic literature review

AIMS: A thematic review that identifies and summarises available evidence for people with Type 1 diabetes (T1D) continuing automated insulin delivery (AID) systems in the hospital setting, primarily with a focus on assessing the in-hospital safety and efficacy of AID use. METHODS: A thematic review was conducted searching Embase, MEDLINE and EBSCO for English language publications from 2014 to 2025 using keywords including AID system brands and terms related to insulin pumps and to inpatients. Eligible studies included original research, retrospective observational studies and case reports and series. RESULTS: Of 1043 articles identified, 1037 did not meet the inclusion criteria. Six articles were reviewed in detail, and a further two papers were identified by screening the reference list of the six papers. There is a paucity of evidence, with heterogeneous methodology regarding the safety and efficacy of continuing AID in hospitalised people with T1D. Results support the feasibility of continuing AID use from the ambulatory to the inpatient setting in hospitalised people with T1D. There is a trend towards improved time in recommended glucose ranges or mean glucose levels without increased hypoglycaemia. There were no adverse glucose outcomes or diabetic ketoacidosis reported. CONCLUSIONS: Whilst continuing AID in the inpatient setting appears promising, our review identified significant heterogeneity in patient populations, device types, as well as limited data on healthcare professional perspectives and person-reported outcomes. Further studies and guidelines are merited. Until then, inpatient use of AID should be guided by specialist diabetes healthcare teams with expertise in diabetes technology.