> more severe unknown incidents like thyroid and pancreatic cancer
Alcoholism, diabetes and obesity don't give you long enough to get these cancers. Even if we had evidence GLP-1 agonists massively increased the risk of these cancers, which we don't, despite them having been studied and prescribed for decades [1][2], the tradeoff makes sense for much of the population.
To put that into perspective, annual cost of diabetes to America is $1,200 per capita ($1,300 adjusted for inflation [2]. 40% more than a month of Ozempic [3].
One twelfth of Americans have diabetes [4]. If the government could negotiate just a 10% bulk discount, the payback period on buying every American with diabetes Ozempic would be around a year. Even if everyone ceased treatment after that, even if you only did this for a couple years, we'd see a permanent boost in productivity and health across the American population.
Government could build their own plants, these drugs are not under patent, just the drug+delivery mechanism is. This says $5 per month as actual costs.
"Semaglutide is expected to become patent-free in the United States no earlier than December 2031" [1].
We could probably strike a volume deal for ~$200 for 5 years and then $75 (the upper end of your study’s estimate for manufacturing cost) for another 5.
This really depends how you calculate cost to society. Everyone dies, and the healthiest people die the slowest. So when you look at the cost of the alternatives the difference looks much smaller or non-existant. If the alternative of getting a heart attack at 57 is retiring at 67, collecting social security for 20 years, getting Alzheimers at 73 and spending 10 years in a Medicare funded memory care facility, obesity and alcoholism don't look as expensive.
From your third link, the metastudy with the most power: "GLP-1 receptor agonists had no significant effects on the occurrence of thyroid cancer (RR 1.30, 95% CI 0.86-1.97), hyperthyroidism (RR 1.19, 95% CI 0.61-2.35), hypothyroidism (RR 1.22, 95% CI 0.80-1.87), thyroiditis (RR 1.83, 95% CI 0.51-6.57), thyroid mass (RR 1.17, 95% CI 0.43-3.20), and goiter (RR 1.17, 95% CI 0.74-1.86)."
The only significant finding was "an increased risk of overall thyroid disorders (RR 1.28, 95% CI 1.03-1.60)."
Scientific literacy is important. The contributions to that meta-study were predominantly looking at 1-2 year outcomes, with a few outliers, which gives means it draws little insight with regard to what might apply from long-term (or lifetime!) use.
The abstract of the paper itself concludes "However, due to the low incidence of these diseases, these findings need to be examined further." suggesting that the authors didn't even find their own work to be conclusive about the short/medium-term risks at the scale of widespread use, let alone the long-term risks that they weren't even analyzing.
Straw man. Nobody claimed these studies prove GLP-1 agonists are totally safe. Just that they don't show "severe unknown incidents like thyroid and pancreatic cancer" [1].
> suggesting that the authors didn't even find their own work to be conclusive about the short/medium-term risks at the scale of widespread use, let alone the long-term risks that they weren't even analyzing
Sure. That's a straightforward reading of the paper. Again, nobody argued these aren't thesre. Just that present data don't sustain it as a serious problem.
>However, no major safety concerns have arisen to date, although definitive conclusions for pancreatic cancer, thyroid cancer and DRP complications cannot be drawn at this point. When compared with the beneficial effects of these drugs on glucose metabolism, blood pressure, body weight and cardiovascular (and potentially even renal) endpoints, these agents have an overall beneficial risk/benefit-profile for treatment of patients with T2D.
