Table 2. Associations between serum magnesium levels and Proton Pump inhibitor use as determined by regression analyses.

	Proton Pump inhibitor						H2 receptor antagonist		No acid-suppressive medications
	Total		Low-dose proton pump inhibitors		High-dose proton pump inhibitors
Magnesium±s.e. (mg/dL)	2.52±0.46		2.54±0.47		2.47±0.39		2.68±0.4		2.68±0.45
	β-coefficient±s.e.	P value	β-coefficient±s.e.	P value	β-coefficient±s.e	P value	β-coefficient±s.e	P value	Reference
Unadjusted model	-0.155±0.026	<0.001	-0.141±0.028	<0.001	-0.210±0.043	<0.001	-0.129±0.0451	0.775	-
Model I	-0.146±0.026	<0.001	-0.135±0.0283	<0.001	-0.204±0.042	<0.001	-0.143±0.0446	0.748	-
Model II	-0.147±0.029	<0.001	-0.131±0.031	<0.001	-0.185±0.046	<0.001	-0.181±0.0472	0.711

Reference category is the patients on no acid suppressive medications. β-coefficients ± s.e’s and P values are provide for each variable. Model I includes age, dialysis vintage and sex. Model II includes all variables in Model I and the addition of Diabetes mellitus, Kt/V, systolic blood pressure, albumin, potassium, C-reactive protein, sodium, blood urea nitrogen, parathyroid hormone, phosphorus, calcium, hemoglobin, furosemide, antiplatelet drug, vitamin K antagonist, angiotensin converting enzyme inhibitor or angiotensin II receptor blocker, phosphate binder, vitamin D receptor antagonist, atrial fibrillation, gastric hemorrhage, cerebral infarction, and ischemic heart disease. Conversion factors for units: serum magnesium in mg/dL to mol/L, *0.4114.