Automated Insulin Delivery Boosts Time-in-Range by 2.7 Hours Daily in Older Adults with Type 1 Diabetes
A meta-analysis of 23 studies finds automated insulin delivery safely improves glycemic control in older and high-risk type 1 diabetes patients.
Summary
A systematic review and meta-analysis pooling 23 studies and 967 participants examined whether automated insulin delivery (AID) — closed-loop systems that continuously adjust insulin based on real-time glucose readings — works safely in older adults (60+) and high-risk individuals with type 1 diabetes. These groups often have impaired hypoglycemia awareness or cognitive and functional limitations, making glucose management especially challenging. AID increased time in the healthy glucose range by about 11.4 percentage points, equivalent to nearly 2.7 extra hours per day. Benefits occurred both during the day and overnight. Hyperglycemia time fell by nearly 8% and hypoglycemia time by over 1%. Serious adverse events were rare. Psychosocial scores did not significantly improve. Findings suggest AID is a viable, safe option even for traditionally vulnerable older patients.
Detailed Summary
Managing type 1 diabetes in older adults presents unique challenges. Aging is associated with reduced hypoglycemia awareness, cognitive decline, and functional limitations — all of which complicate self-managed insulin regimens and raise the stakes of glucose excursions. Automated insulin delivery (AID), or closed-loop technology, continuously senses blood glucose and adjusts insulin delivery without manual input, potentially transforming care for this vulnerable population. Yet rigorous evidence for older and high-risk groups has been lacking.
This systematic review and meta-analysis searched five major databases through March 2026, ultimately including 23 studies — 11 randomized controlled trials, 3 single-arm trials, and 9 cohort studies — involving 967 participants aged 60 and older or classified as high-risk due to impaired hypoglycemia awareness, cognitive impairment, or functional limitations.
The primary outcome, time in range (70–180 mg/dL), improved by 11.38 percentage points (95% CI: 8.07–14.70), translating to roughly 2.7 additional hours per day within the healthy glucose window. Gains were seen both during daytime hours (+8.70%) and overnight (+13.35%), with overnight improvements being particularly pronounced. Time spent in hyperglycemia fell by 7.89% and time in hypoglycemia decreased by 1.16%. Both commercial and non-commercial AID systems delivered comparable benefits, broadening the practical relevance of these findings.
Importantly, rates of serious adverse events — including diabetic ketoacidosis and severe hypoglycemia — were not significantly elevated, supporting the safety profile of AID in these populations. However, no statistically significant improvements in psychosocial outcomes (fear of hypoglycemia, quality of life) were detected.
Caveats include substantial heterogeneity (I² = 70%), a very low certainty of evidence rating, small total sample size, and the fact that this summary is based on the abstract only. Clinicians should interpret psychosocial findings cautiously given limited measurement tools and follow-up durations across included studies.
Key Findings
- AID increased time in glucose target range by 11.4% — roughly 2.7 extra hours per day — in adults 60+ with type 1 diabetes.
- Overnight time in range improved most dramatically (+13.35%), addressing a key vulnerability in older patients.
- Hyperglycemia time dropped ~7.9% and hypoglycemia time fell ~1.2%, improving both ends of glucose control.
- Serious adverse events including diabetic ketoacidosis and severe hypoglycemia were not significantly increased.
- Both commercial and non-commercial closed-loop systems produced comparable glycemic benefits.
Methodology
Systematic review and meta-analysis of 23 studies (11 RCTs, 3 single-arm trials, 9 cohort studies) involving 967 participants drawn from five databases through March 2026. Pooled effect sizes were calculated using random-effects models; the protocol was pre-registered on PROSPERO (CRD420251273526). Overall certainty of evidence was rated very low, and heterogeneity was substantial (I² = 70%).
Study Limitations
The overall certainty of evidence was rated very low, and substantial heterogeneity (I² = 70%) limits confidence in pooled estimates. The total sample size of 967 participants is modest for a meta-analysis, and psychosocial outcome tools varied across studies. This summary is based on the abstract only, as the full text was not available.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.
Enter your email to subscribe: