Among nearly 10 million US infants born between 2016 and 2018, breastfed babies were 33% less likely to die during the post-perinatal period (day 7−364) than infants who were not breastfed, reports a new study in the American Journal of Preventive Medicine, published by Elsevier. The findings build on previous US research with smaller datasets, which documented the association between the initiation of breastfeeding and the reduction of post-perinatal infant mortality by a range of 19% to 26%.
The problem with this data is that there are a lot of external factors that are usually not considered in the data-set, like for example that a poorer family won’t have time to breast feed because the mom has to go back to work asap.
You will not always accurately flatten out socioeconomic, etc factors because there are no test-subjects per se, but rather just pooling.
Interesting video on the topic (Sci Show)
https://youtu.be/i1UMnKduosE