This supplementary appendix is part of the original submission and has been peer reviewed. It provides additional information on the study titled "The Global Burden of Metabolic Risk Factors for Chronic Diseases Collaboration (BMI Mediated Effects)." The study investigates the metabolic mediators of the effects of body mass index (BMI), overweight, and obesity on coronary heart disease (CHD) and stroke, using a pooled analysis of 97 prospective cohorts with 1.8 million participants. The study used systematic literature searches in Medline (PubMed) and Embase databases, as well as personal communications with researchers. The search included articles published up to March 2010, focusing on adiposity measures, cardiovascular outcomes, and prospective studies. Articles were evaluated by two reviewers, and the number of retained articles is summarized in a flow diagram. Cohort studies were eligible if they had a prospective design with at least one year of follow-up, participants not selected based on prior history of CHD or stroke, and baseline measurements of height, weight, and at least one mediator (blood pressure, serum cholesterol, and blood glucose or diabetes). Fatal and/or non-fatal CHD or stroke were ascertained during follow-up. Webappendix 2 describes the estimation of uncertainty in the percentage of excess risk mediated (PERM). This was done by randomly drawing 5000 pairs of hazard ratios (HRs) from their corresponding uncertainty distributions, accounting for their correlations. The bootstrap method was used to estimate the correlations between HRs adjusted for confounders and HRs adjusted for confounders and mediators. Webtable 1 summarizes the correlations between HRs adjusted for confounders and HRs adjusted for confounders and mediators from 7 selected cohorts. The estimated correlations were used to incorporate a conservative estimate of 0.95 in the uncertainty calculation. Webtable 2 provides characteristics of the cohorts included in the pooled analysis. Webtable 3 presents stratified analyses of hazard ratios (HRs) for overweight and obesity vs. normal weight. Webfigures 1-9 show various hazard ratios (HRs) for CHD and stroke per 5 kg/m² higher BMI, stratified by BMI categories and adjusted for different confounders and mediators. The figures also include heterogeneity measures (I²) to assess the variability in results across studies.This supplementary appendix is part of the original submission and has been peer reviewed. It provides additional information on the study titled "The Global Burden of Metabolic Risk Factors for Chronic Diseases Collaboration (BMI Mediated Effects)." The study investigates the metabolic mediators of the effects of body mass index (BMI), overweight, and obesity on coronary heart disease (CHD) and stroke, using a pooled analysis of 97 prospective cohorts with 1.8 million participants. The study used systematic literature searches in Medline (PubMed) and Embase databases, as well as personal communications with researchers. The search included articles published up to March 2010, focusing on adiposity measures, cardiovascular outcomes, and prospective studies. Articles were evaluated by two reviewers, and the number of retained articles is summarized in a flow diagram. Cohort studies were eligible if they had a prospective design with at least one year of follow-up, participants not selected based on prior history of CHD or stroke, and baseline measurements of height, weight, and at least one mediator (blood pressure, serum cholesterol, and blood glucose or diabetes). Fatal and/or non-fatal CHD or stroke were ascertained during follow-up. Webappendix 2 describes the estimation of uncertainty in the percentage of excess risk mediated (PERM). This was done by randomly drawing 5000 pairs of hazard ratios (HRs) from their corresponding uncertainty distributions, accounting for their correlations. The bootstrap method was used to estimate the correlations between HRs adjusted for confounders and HRs adjusted for confounders and mediators. Webtable 1 summarizes the correlations between HRs adjusted for confounders and HRs adjusted for confounders and mediators from 7 selected cohorts. The estimated correlations were used to incorporate a conservative estimate of 0.95 in the uncertainty calculation. Webtable 2 provides characteristics of the cohorts included in the pooled analysis. Webtable 3 presents stratified analyses of hazard ratios (HRs) for overweight and obesity vs. normal weight. Webfigures 1-9 show various hazard ratios (HRs) for CHD and stroke per 5 kg/m² higher BMI, stratified by BMI categories and adjusted for different confounders and mediators. The figures also include heterogeneity measures (I²) to assess the variability in results across studies.