Fig. 5

Mediation analysis of serum 25(OH)D on the interaction between heavy metals and CVD. A Mediation models of 25(OH)D, Cadmium, and CVD: direct effect (TE = 0.018056; P = 0.008) of 25(OH)D (exposure) toward CVD (outcome), and 25(OH)D medication proportion is 1.45%; indirect effect (IE = 0.000294; P < 0.001) of 25(OH)D (exposure) toward 25(OH)D (mediator) and effect CVD (DE = 0.017762; P < 0.001), from 25(OH)D (mediator) toward CVD (outcome). B Mediation models of 25(OH)D, Lead, and CVD: direct effect (TE = 0.00348; P = 0.140) of 25(OH)D (exposure) toward CVD (outcome), and 25(OH)D medication proportion is 2.56%; indirect effect (IE = 0.000114; P < 0.240) of 25(OH)D (exposure) toward 25(OH)D (mediator) and effect CVD (DE = 0.00337; P < 0.150), from 25(OH)D (mediator) toward CVD (outcome). C Mediation models of 25(OH)D, Mercury, and CVD: direct effect (TE = -0.000893; P = 0.510) of 25(OH)D (exposure) toward CVD (outcome), and 25(OH)D medication proportion is 2.21%; indirect effect (IE = −0.000057; P < 0.220) of 25(OH)D (exposure) toward 25(OH)D (mediator) and effect CVD (DE = −0.000836; P < 540), from 25(OH)D (mediator) toward CVD (outcome). D Mediation models of 25(OH)D, Manganese, and CVD: direct effect (TE = −0.0046; P < 0.001) of 25(OH)D (exposure) toward CVD (outcome), and 25(OH)D medication proportion is 6.7%; indirect effect (IE = −0.0003; P < 0.001) of 25(OH)D (exposure) toward 25(OH)D (mediator) and effect CVD (DE = −0.0043; P < 0.001), from 25(OH)D (mediator) toward CVD (outcome). E Mediation models of 25(OH)D, Manganese, and CVD: direct effect (TE = −0.0000649; P < 0.620) of 25(OH)D (exposure) toward CVD (outcome), and 25(OH)D medication proportion is 6.7%; indirect effect (IE = −0.0000075; P = 0.023) of 25(OH)D (exposure) toward 25(OH)D (mediator) and effect CVD (DE = −0.0000574; P = 0.650), from 25(OH)D (mediator) toward CVD (outcome). Abbreviations: 25(OH)D, 25-hydroxyvitamin D; CVD, Cardiovascular disease