This is another suspect observational study, IMO. The 0-5 Magnesium Depletion Score is computed from variables that pre-correlate with Metabolic Syndrome irresp

This is another suspect observational study, IMO. The 0-5 Magnesium Depletion Score is computed from variables that pre-correlate with Metabolic Syndrome irrespective of Magnesium: 2 points for kidney disease [1], 1 for alcohol use, 1 for GERD pills, 1 for magic-diet pills. When they try to control for alcohol use, they find another MetS correlation: non-drinkers. But people with diagnosed MetS/(pre-)diabetes/obesity often quit drinking, so they should have checked for heteroscedasticity wrt alcohol use. I'd wait for a randomized study before stocking up on Magnesium tablets. But a multivitamin supplement is usually a good idea anyways. [1] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4220353/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4220353/