3 ms·
What's the line? If we found that some intervention group had 90% mortality with the control group having 10% mortality in some experiment, I think you would ag
by joshgel 8y ago
What's the line? If we found that some intervention group had 90% mortality with the control group having 10% mortality in some experiment, I think you would agree we shouldn't repeat that experiment with a lot of people as you would be randomizing to something likely to kill them. What about 20% mortality to 10%? What about 11% to 10%? What degree of certainty about the results do you need? Obviously there is a line somewhere. But, its not easy, so I wouldn't recommend dismissing such concerns glibly. Is that the same as claiming we answered the question definitively, or is that just ethical research?
- quotemstr 8y agoWe're not talking about administering arsenic to people just in case we're wrong about thousands of years of experience. We're talking about randomizing exposure to an activity that billions of people choose voluntarily, an activity that at worst has an effect of a few parts per 100,000. That's on the same order of risk as being struck by lightning. Unnecessary x-rays are likely worse. Why the outrage over this risk in particular? Declaring this level of risk to be outside the bounds of permissible research is an attitude incompatible with scientific progress. It's elevating short-term harm prevention to an unjustifiable extreme that makes it impossible to determine whether we have an accurate model of the world.