| Popis: |
The prevailing sentiment in the general population is that a Body Mass Index (BMI) above desired levels leads to less than optimal health outcomes and increased likelihood of death. Recent medical studies, however, have found evidence that the opposite is true for a certain subset of the population, particularly those with a history of cardiovascular disease (CVD). These studies found that individuals with a BMI slightly above desired experience more optimal health outcomes compared to those with a BMI in the \textit{desired} range - a phenomenon known as the ``obesity paradox.'' However, these studies have primarily been performed using classical tools of probability and statistics. In this study, we apply tools appropriated from causal inference, with some loosened restrictions, to the age >= 65 population of the Kailuan dataset, a longitudinal dataset tracking almost 15,000 patients, 10,005 of which have a history of CVD , 4,641 have a BMI slightly above desired, 5,863 have a BMI in the desired range, and 317 a BMI slightly below desired. We ultimately find some evidence of the obesity paradox in this dataset. |
