The breastfeeding research seems like peak "bad science", because there's so much obvious confounding between the causes of breastfeeding as a behavior and the

The breastfeeding research seems like peak "bad science", because there's so much obvious confounding between the causes of breastfeeding as a behavior and the properties of the child. I probably would suspect minimal effect unless there were (1) a clear mechanism of action, (2) really strong associations even after adding controls, (3) a causal study, perhaps on non-humans.