> Previously believed by whom? The previous data/methodology. The whole point of the study was to look at three untracked externalities[0] in BMI-outcome resea

> Previously believed by whom? The previous data/methodology. The whole point of the study was to look at three untracked externalities[0] in BMI-outcome research. They used historical NHANES data combined 2015 data on mortality data. > (1) confounding bias from heterogeneity in body shape; (2) positive survival bias in high-BMI samples due to recent weight gain; and (3) negative survival bias in low-BMI samples due to recent weight loss. https://www.tandfonline.com/doi/abs/10.1080/00324728.2023.2168035?journalCode=rpst20 https://www.tandfonline.com/doi/abs/10.1080/00324728.2023.21...