Hacker News Reader

Actually, I can recall reading about a few double-blind medical studies that were terminated early, because of this very reason. The efficacy of the treatment was so overwhelmingly positive in the variable group, that it was deemed unethical to not give the treatment to the control group. I don't have a citation right this second, but if you want one, I can do some searching. If anyone remembers a particular case of this happening, I would appreciate a citation.

Circumcision to prevent HIV infection is one example.

https://en.wikipedia.org/wiki/Circumcision_and_HIV

This strongly anti-circumcision site has some more info, but the ranting gets a bit much: http://www.circumstitions.com/HIV-SA.html

Placebo doesn't mean "no treatment" - it could be the standard therapy, while an experimental drug is matched up against it.

Then again, "nominal monthly income supplement" is the standard treatment for poverty in many places.

Yup, so not unethical really. Especially when a higher monthly supplement might not have benefit over placebo.

The difference is that with cancer trials one is close to death already (as far as I understand)