It certainly does not. To make that claim, the study would need a control group of people who had never taken the drug. They didn't have that:

> Participants Veterans Affairs users with type 2 diabetes who started treatment with GLP-1RAs (n=132 551) or sulfonylureas (n=201 136), followed up for three years. Veterans Affairs users were defined as having at least two visits to Veterans Affairs and having used the Veterans Affairs outpatient pharmacy within a year before receiving treatment with GLP-1RAs or sulfonylureas.

▲

embedding-shape 5 hours ago | parent [-]

> They didn't have that

So, why not? Seems very obvious to everyone here on HN that it's "kind of useless" unless they did have that, yet they didn't. What reason would there be for ignoring that?

	▲	post-it an hour ago \| parent [-]
		They were testing what happens when you stop taking Ozempic compared to what happens when you don't stop taking it, and also what happens when you start taking it again. Assembling a control group of people who have never taken Ozempic could be difficult. How do you control for the fact that people not on Ozempic are less likely to need Ozempic? You'd need to figure out some criteria by which to include and exclude patients before sorting by whether they take Ozempic or not, so you'd have a smaller sample size of people who are taking Ozempic. Best not to allow scope creep.