Table 5.

Relative Risk (95% CI) of CHD by Quartile (Q) of Plasma Magnesium

	Plasma Magnesium
	Q1	Q2	Q3	Q4	P–Linear Trend*	RR Comparing ≥2.1 With <2.1 mg/dL
Range of magnesium, mg/dL	≤2.0	2.1 to 2.1	2.2 to 2.3	≥2.4
Cases	98	55	204	101
Controls	66	70	201	121
Model 1*	1.0 (ref)	0.51 (0.31 to 0.82)	0.65 (0.44 to 0.96)	0.54 (0.35 to 0.82)	0.04	0.59 (0.41 to 0.85)
Model 2*	1.0 (ref)	0.51 (0.30 to 0.85)	0.73 (0.48 to 1.11)	0.61 (0.39 to 0.97)	0.22	0.65 (0.44 to 0.96)
Model 2+magnesium intake	1.0 (ref)	0.50 (0.30 to 0.84)	0.74 (0.49 to 1.12)	0.61 (0.39 to 0.97)	0.23	0.65 (0.44 to 0.96)
Model 2+diabetes, hypertension, and high cholesterol at baseline	1.0 (ref)	0.47 (0.27 to 0.81)	0.76 (0.48 to 1.20)	0.60 (0.37 to 0.98)	0.30	0.65 (0.42 to 0.99)
Model 2+HDL‐C, LDL‐C, hsCRP, adiponectin, and HbA1c	1.0 (ref)	0.50 (0.28 to 0.88)	0.78 (0.49 to 1.25)	0.63 (0.38 to 1.05)	0.37	0.67 (0.44 to 1.04)

CI indicates confidence interval; CHD, coronary heart disease; RR, relative risk; HDL‐C, high‐density lipoprotein cholesterol; LDL‐C, low density lipoprotein cholesterol; hsCRP, high‐sensitivity C‐reactive protein; HbA1c, hemoglobin A1c; BMI, body mass index; MI, myocardial infarction; eGFR, estimated glomerular filtration rate.

^*Test for linear trend calculated by assigning the median value of magnesium in each quintile and modeling this as a continuous variable in regression models.

^*Model 1: RR estimated from logistic regression models, conditioned on matching factors (age, smoking status, month of blood draw, and fasting status).

^*Model 2: RR further adjusted for BMI, exercise, alcohol intake, family history of MI, eGFR, menopausal therapy, multivitamin use, and intake of omega‐3 fats, polyunsaturated:saturated fat, trans fat, dietary cholesterol, cereal fiber, calcium, potassium, and vitamin D.