Aims Systematic reviews report an association between poorer oral health and an increased risk of coronary heart disease. This contentious relationship may not be causal but existing studies have been insufficiently well powered comprehensively to examine the role of confounding, particularly by cigarette smoking. Accordingly, we sought to examine the role of smoking in generating the relationship between oral health and coronary heart disease in life-long non-smokers. Methods and results In the Korean Cancer Prevention Study, 975,685 individuals (349,579 women) aged 30-95 years had an oral examination when tooth loss, a widely used indicator of oral health, was ascertained. Linkage to national mortality and hospital registers over 21 years of follow-up gave rise to 64,784 coronary heart disease events (19,502 in women). In the whole cohort, after statistical adjustment for age, there was a moderate, positive association between tooth loss and coronary heart disease in both men (hazard ratio for seven or more missing teeth vs. none; 95% confidence interval 1.08; 1.02, 1.14; Ptrend across tooth loss groups <0.0001) and women (1.09; 1.01, 1.18; Ptrend 0.0016). Restricting analyses to a subgroup of 464,145 never smokers (25,765 coronary heart disease events), however, resulted in an elimination of this association in men (1.01; 0.85, 1.19); Ptrend 0.7506) but not women (1.08; 0.99, 1.18; Ptrend 0.0086). Conclusion In men in the present study, the relationship between poor oral health and coronary heart disease risk appeared to be explained by confounding by cigarette smoking so raising questions about a causal link.
Keywords: Oral health; coronary heart disease; epidemiology.