A new study finds the popular diabetes medications may not make stopping insulin more likely
A Yale-led study looked at whether GLP-1 drugs could help people with type 2 diabetes stop using insulin.
Researchers analyzed three years of health records from nearly 9,000 matched groups of veterans with type 2 diabetes.
People taking GLP-1 drugs stopped insulin at about the same rate as those taking two other types of diabetes medications.
GLP-1 drugs have become a major topic in diabetes care, and previous research has found that these medications can reduce how much insulin some people need. But a bigger question remains: Can they actually help people with type 2 diabetes stop taking insulin altogether?
A new study led by researchers at Yale School of Medicine examined that question. The researchers compared people taking GLP-1 receptor agonists with people taking two other types of oral diabetes medications to see whether GLP-1 treatment was associated with higher rates of insulin discontinuation.
Insulin can be an effective treatment for people with type 2 diabetes, but it also comes with daily injections, frequent monitoring, and the possibility of low blood sugar reactions. That makes the possibility of safely reducing or stopping insulin an important question for some patients.
How did researchers study it?
The researchers used electronic health record data from the U.S. Department of Veterans Affairs. They created 8,869 matched sets of veterans with type 2 diabetes who were already using basal insulin and had started treatment with one of three types of medications: a GLP-1 receptor agonist, an SGLT-2 inhibitor, or a DPP-4 inhibitor.
The groups were matched based on similar health characteristics, allowing researchers to compare insulin discontinuation rates across the different treatments. The researchers then followed the patients' records for three years.
For the study, stopping insulin was defined as having a gap of at least 12 months between insulin prescription fills. Most people receiving GLP-1 treatment in the study were taking semaglutide, dulaglutide, or liraglutide.
What did they find?
Over the three-year follow-up period, 16.7% of people who started a GLP-1 drug discontinued insulin. That compared with 17.9% of those taking an SGLT-2 inhibitor and 17.1% of those taking a DPP-4 inhibitor. Researchers did not find a comparative advantage for GLP-1 drugs when it came to stopping insulin.
That doesn't mean GLP-1 drugs can't reduce insulin needs. Instead, the study suggests that simply adding a GLP-1 medication does not necessarily make a person more likely to stop insulin than taking the other medications studied.
It's very complex, researcher Kasia Lipska, M.D., MHS, said in a news release. Patients often add or switch medications over time, so it becomes very difficult to disentangle.
The researchers also pointed out some limitations. This was not a randomized clinical trial, and patients' treatments could change over time. In addition, people in the study did not advance to the full doses of GLP-1 drugs currently available, which may have affected the results.
For consumers, the findings also highlight a less-discussed issue: There isn't a standard approach for knowing when and how to safely withdraw insulin. According to the researchers, more guidance may be needed to help clinicians determine when reducing or stopping insulin is appropriate.
Most training for diabetes care is about starting and adjusting dose, not stopping or withdrawing, Dr. Lipska said.
The lesson may be that prescribing a GLP-1 receptor agonist is only the first step. If we want to help people safely come off insulin, we also need to know when and how to withdraw it. Right now, clinicians have very little guidance for doing that.
Posted: 2026-09-02 18:14:32
