A new study indicates that for older US veterans with long-standing diabetes, adding GLP-1 receptor agonists does not significantly reduce the need for insulin, highlighting the importance of comprehensive treatment plans.
For many people with type 2 diabetes, GLP-1 medicines have come to represent more than a way to lower blood sugar. They can also aid weight loss, and that has led to hopes that they might eventually reduce or even replace the need for insulin. A new study published in the Annals of Internal Medicine suggests the reality is more limited: in a large group of older US veterans, adding a GLP-1 drug did not meaningfully improve the chances of stopping basal insulin compared with other common glucose-lowering medicines. In practical terms, the research indicates that starting a GLP-1 receptor agonist is not, by itself, a reliable route off insulin for patients with long-standing, poorly controlled diabetes.
The study analysed electronic health records from nearly 9,000 patients in the US Department of Veterans Affairs system who were taking either a GLP-1 agonist, an SGLT2 inhibitor or a DPP-4 inhibitor. SGLT2 inhibitors help the body remove excess glucose through urine, while DPP-4 inhibitors work by helping control blood sugar after meals. The researchers focused on basal insulin, the long-acting form taken to keep glucose steady between meals and overnight. Lead author Kasia Lipska, a Yale School of Medicine endocrinologist, said in a press release that prescribing a GLP-1 drug may be only the first step, adding that clinicians still lack clear guidance on when and how to withdraw insulin safely.
The findings also have to be read in context. Only about 17% of patients in the study stopped insulin, and the veterans were generally older than 65, had A1c levels above 9% and were already in a group likely to have more advanced beta-cell damage, making insulin withdrawal harder. The care was also largely delivered in primary care rather than endocrinology clinics, which may have made doctors more cautious about changing treatment. Experts quoted by Medscape said the study did not include tirzepatide, a newer and often more potent incretin medicine, and that a broader, better-supported approach may matter more than simply adding one drug.
That broader picture is important because real-world use of GLP-1 therapies is often fragile. A 2026 review in Nature Reviews Endocrinology noted that many patients stop GLP-1 receptor agonists or tirzepatide within the first year, often because of gastrointestinal side effects, cost or concerns about rare harms, and that discontinuation can reverse gains in weight and cardiometabolic health. Other recent studies have found similar patterns, with substantial early dropout and rebound weight regain after treatment ends. Taken together, the evidence suggests GLP-1 medicines can be valuable tools in diabetes care, but they work best as part of a wider plan that includes careful insulin management, lifestyle change and sustained follow-up rather than as a simple substitute for insulin on their own.
Disclaimer: This content is for informational purposes only and is not intended to be a substitute for professional medical judgment, advice, diagnosis, or treatment.
