Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Can you perform cost benefit analysis in the long term on a drug that has only been used in the short term?


It has been pretty heavily researched since the 90s.


yes. everyone knows chemotherapy agents are going to be bad, and that they can’t be continued forever etc, and yet the long-term prognosis is absolutely clear as to what happens if you don’t take them.

drugs don’t need to be used long-term to have a pretty good picture of the long-term tradeoffs, particularly when it’s something like cancer or obesity that has a clearly negative long-term outcome in the a sense of a successful intervention. Most interventions for obesity are ultimately not successful and to have one that is is a game changer, even if it has some minor side effects (and burden of proof cuts both ways - it’s not clear that they even exist etc, only that it would still be worth it if they did).

in the real world, things usually don’t turn out to have some stealth side effect that takes 30 years to manifest, and nobody is telling you to take them anyway, just to stop injecting yourself into private medical decisions around an approved, well-studied drug with a seemingly quite benign profile.

(why oh why do I suspect that semaglutide skepticism correlates heavily with the “respect my freedom to not wear a mask” crowd…)


Not a great analogy, as not many people use chemotherapy agents daily, for life (unless you die while taking them, of course) and those can have crippling side effects that are way worse in some populations than simply dying from cancer.

> stealth side effect that takes 30 years to manifest

Heavy metal toxicity, cirrhosis, lung cancer, persistent infections that lead to cancer (HPV, Hep B, etc), silicosis, COPD, skin cancer, all sorts of diseases result from cumulative exposure.

> (why oh why do I suspect that semaglutide skepticism correlates heavily with the “respect my freedom to not wear a mask” crowd…)

Can you approach skepticism without ad hominem attacks?


liraglutide was developed in 1998 by Novo Nordisk. define short term.


Yeah, and there is some literature questioning its safety profile.

See this disproportionality analysis from a nationwide population-based study of the French (full text, figure 2):

https://diabetesjournals.org/care/article/46/2/384/147888/GL...


You know that GLP-1s aren't new, right? There are diabetics who have been on them continuously for decades; it's only the realization that they can treat obesity which is new. If there were common and severe side-effects from continuous long-term use, they'd already have shown up in diabetic patients.


What is better about the newer ones that there is so much hype? Or did it take 30 years to realise the other positive effects of these?


A few things:

1. Weekly vs daily administration, liraglutide and exenatide had to be injected daily which made the drug less attractive for patients and doctors vs metformin which is well known and a cheap pill.

2. Greater magnitude of effect. Liraglutide had less impact at prescribed dosage for both lowering A1C and weight loss so in combo with (1) made it less attractive than semaglutide.

3. Magnitude of weight loss from (2) now so significant it easily beats all previously approved weight loss drugs with few side effects. So entire GLP-1 agonist class of drugs reexamined as direct weight loss treatments vs just a modest but very welcome side effect of diabetes treatment. And once weekly administration meant it would have much better patient compliance than a daily injection.


Great summary thanks




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: