Methods/Statistics
Body Mass Index as a Negative Confounder? Evidence and solutions Zhu Liduzi Jiesisibieke* Zhu Liduzi Jiesisibieke Zhu Liduzi Jiesisibieke C Mary Schooling
Background: Adequate control for confounding is key to many observational study designs. Confounders are often identified based on subject matter knowledge from empirical investigations. Negative confounders, which typically generate type 2 error, i.e., false nulls, can be elusive. Such confounders can be identified comprehensively by using Mendelian randomization to search the wealth of publicly available data systematically. Here, to demonstrate the concept, we examined whether a common positive confounder, body mass index (BMI), was also a negative confounder of any common physiological attributes on health outcomes, overall and sex-specifically.
Methods: We used a Mendelian Randomization (MR) study to assess, overall and sex-specifically, whether BMI is a negative confounder potentially obscuring effects of potentially harmful physiological attributes. Inverse variance weighting was the main method. We assessed sex differences using a z-test.
Results: BMI was a potential negative confounder for Apolipoprotein B and total testosterone in men, and in both sexes for low-density lipoprotein-cholesterol, choline, phosphate, linoleic acid, polyunsaturated fatty acids and cholesterol.
Conclusions: Using BMI as an illustrative example, we demonstrate that negative confounding is an easily overlooked bias. Given negative confounding is not always obvious or known, using MR to identify potential negative confounders in observational studies may be helpful.