>A system that ONLY detects cancers earlier There isn't such a system, anywhere. Even if you could have a public health care that detects and tries not to trea

>A system that ONLY detects cancers earlier There isn't such a system, anywhere. Even if you could have a public health care that detects and tries not to treat, you still cannot effectively forbid people from getting treatment themselves - if you make a law against it, people will try to escape the country to get treatment. So if you have a better detection rate, there will be actions to treat. >Increased detection sometimes helps and sometimes hurts. Perhaps the fallacy here is the belief that the helps/hurts ratio of treatment is 50/50? It isn't. Of course there are cases where the treatment actually made things worse. There are more cases where the treatment has no significant effect. But it is rather silly to assume that cancer treatments would have a net negative of zero effect. Regarding the infant mortality rate, the definitions of miscarriage, stillbirth and dead newborn are not so different between developed nations that it would change this. Where IMR is lower, also stillbirth rate is lower, so it's not really about moving the boundary between these.