AimThe study aim was to evaluate the feasibility, safety and efficacy of automated insulin delivery (AID) assisted by home health care (HHC) services in people with type 2 diabetes unable to manage multiple daily insulin injections (MDI) at home on their own.Patients and MethodsThis was an open label, multicentre, randomized, parallel group trial. In total, 30 adults with type 2 diabetes using MDI and requiring nursing support were randomly allocated to AID or kept their usual therapy over a 12‐week period. Both treatments were managed with the support of HHC services. The primary outcome was the percentage time in the target glucose range of 70‐180 mg/dl (TIR). Secondary outcomes included other continuous glucose monitoring metrics, glycated haemoglobin (HbA1c) levels, daily insulin doses, body weight, and of quality of life scores, fear of hypoglycaemia and satisfaction questionnaires.ResultsAge (69.7 vs. 69.3 years) and HbA1c (9.25 vs. 9.0) did not differ in MDI and AID at baseline. Compared with MDI, AID resulted in a significant increase in TIR by 27.4% [95% CI (15.0‐39.8); p < .001], a decrease in time above range by 27.7% and an unchanged time below range of <1%. A between‐group difference in HbA1c was 1.3% favouring AID. Neither severe hypoglycaemia nor ketoacidosis occurred in either group. Patient and caregiver satisfaction with AID was high.ConclusionsAID combined with tailored HHC services significantly improved glycaemic control with no safety issues in people with type 2 diabetes previously under an MDI regimen with HHC. AID should be considered a safe option in these people when lacking acceptable glucose control.