Automated insulin delivery in people with type 2 diabetes undergoing elective hip or knee arthroplasty

Intervention delivery: Participants randomised to the intervention group will use the OmniPod 5™ automated insulin delivery (AID) system integrated with the FreeStyle Libre 2 Plus continuous glucose monitoring (CGM) sensor. The OmniPod 5 system is a tubeless, wearable insulin pump (pod) worn on the body (typically arm or abdomen) that delivers rapid-acting insulin continuously for up to three days per pod. The system automatically adjusts basal insulin delivery every five minutes based on real-time CGM glucose measurements. Participants administer prandial insulin using pre-set meal bolus options (small, medium, or large). Rapid-acting insulin (NovoRapid® or Humalog®) is used throughout the study. Device replacement: Participants replace the OmniPod 5 pod themselves approximately every three days for the duration of the intervention. The CGM sensor is replaced according to manufacturer instructions. Participants receive education and written instructions to support independent device changes. Run-in period Participants complete a two-week run-in period prior to formal intervention commencement. During this period, participants wear the CGM sensor and continue usual diabetes care to assess baseline glycaemic patterns, ensure device familiarity, and confirm study eligibility prior to initiation of OmniPod 5. Following the run-in period, participants commence OmniPod 5 approximately three to four weeks prior to elective hip or knee arthroplasty and continue use for approximately ten weeks, including the pre-operative period, hospital admission (including surgery), and post-discharge period. Education and training The intervention is delivered individually. Initial education: One to two individual face-to-face sessions at randomisation (approximately 60 minutes per session) with a credentialled diabetes clinician. These sessions include device setup, insulin parameter configuration, troubleshooting, and safety education. Dietitian education: A face-to-face session with a dietitian (approximately 30 minutes) occurs during the initial education period. This session focuses on carbohydrate awareness, meal bolus selection (small, medium, large), peri-operative nutrition considerations, and use of the OmniPod 5 system in relation to food intake. Early follow-up: A supervised pod change within the first few days of initiation (approximately 20 to 30 minutes) to ensure correct technique and reinforce education. Ongoing review: Follow-up reviews at scheduled study visits (pre-operative, hospital admission, and post-discharge), each lasting approximately 20 to 30 minutes, with additional contacts provided via face-to-face, telehealth, or telephone as clinically indicated. Setting The intervention is delivered across multiple settings, including: Outpatient clinical trials unit (education, initiation, and follow-up), Participant’s home environment (daily system use), Hospital inpatient setting during admission for surgery, and Virtual platforms or telephone for interim support. Participants may continue using the OmniPod 5 system during hospital admission unless clinically contraindicated. Adherence monitoring Adherence will be monitored using objective device-generated data. Measures include CGM wear time, percentage of time the OmniPod 5 system operates in automated insulin delivery mode, and the number of days with equal to or more than 70 percent valid CGM data. Device data will be uploaded and reviewed at scheduled study visits and during hospital admission.