Additional file 1: Table S1. of Mercury exposure and risk of cardiovascular disease: a nested case-control study in the PREDIMED (PREvention with MEDiterranean Diet) study

Relative Risk of Cardiovascular Disease, according to quartiles of toenail mercury, among case participants and matched controls in the PREDIMED trial, restricted to those who were not diagnosed with cardiovascular disease within one year of follow-up. Table S2. Relative Risk of Cardiovascular Disease, according to quartiles of toenail mercury, among case participants and matched controls in the PREDIMED trial, restricted to those who were diagnosed with cardiovascular disease within 5 years of toenail collection. Table S3. Relative Risk of Cardiovascular Disease, according to quartiles of toenail mercury, among case participants and matched controls in the PREDIMED trial, restricted to participants who were below the 90th percentile of toenail selenium. Table S4. Relative Risk of Cardiovascular Disease, according to quartiles of toenail mercury, among case participants and matched controls in the PREDIMED trial, stratified by baseline median fish intake (below/above the median). Table S5. Relative Risk of Cardiovascular Disease, according to quartiles of toenail mercury, among case participants and matched controls in the PREDIMED trial, restricted to participants who were at very high risk of cardiovascular disease. Table S6. Relative Risk of Cardiovascular Disease, according to quartiles of toenail mercury, among case participants and matched controls in the PREDIMED trial, stratified by baseline aspirin use (yes/no). Table S7. Bias analysis (simulation study). Sensitivity analysis to assess the degree of plausible bias due to residual confounding, assuming an unknown binary confounder (n-3 PUFA consumption; dichotomized at median) with different prevalence among exposed (highest quartile of mercury) and unexposed (lowest quartile of mercury) controls and with a relatively strong association (even after adjusting for the known and measured confounders) with cardiovascular disease. This correction was applied to our observed OR=0.66 for the highest versus lowest quartile of mercury exposure. (DOC 153 kb