High serum calcium and phosphate levels have been linked to cardiovascular diseases and all-cause mortality but evidence from longitudinal studies is scarce, especially among patients with pre-existing coronary heart disease. The association between baseline calcium and phosphate and prognosis was examined in a cohort study of patients with stable coronary heart disease.Serum calcium and phosphate were measured in a cohort of initially 1206 patients undergoing a 3 week rehabilitation programme after an acute cardiovascular event and subsequently being followed-up for 8 years. Multivariate Cox regression was employed to assess the association of quartiles and continuous levels of calcium and phosphate with secondary cardiovascular events and all-cause mortality.No significant risk elevations were observed for secondary cardiovascular event incidence in models adjusted for a variety of potential confounders. High calcium levels, however, were strongly associated with mortality risk in adjusted models (HR(Q4vsQ1)=2.39 (1.22 to 4.66)). In additional multivariable analyses, the calcium/albumin ratio was predictive for all-cause mortality (HR(Q4vsQ1)=2.66 (1.35 to 5.22)) and marginally predictive for cardiovascular event incidence (HR(Q4vsQ1)=1.74 (1.00 to 3.05)).Calcium and the ratio of calcium with albumin, its major binding protein, were strongly associated with all-cause mortality among patients with coronary heart disease. The underlying mechanisms and the clinical implications of these findings deserve further study.