Abstract
In order to investigate the joint effect of history of diabetes and blood transfusion on ischemic heart disease mortality, we analyzed data for 7,759 subjects enrolled in a cohort study in Fukuoka Prefecture, Japan. All deaths due to ischemic heart disease were recorded. The Cox proportional hazards model was used to estimate hazard ratios (HR) and 95% confidence intervals (95% CI) of ischemic heart disease with history of diabetes and blood transfusion. We also tested the null hypothesis of the multiplicative model to assess the potential joint effect of these two factors. During 102,800 person-years of follow-up, we observed 66 deaths due to ischemic heart disease. Adjusted HRs of diabetes only as well as transfusion only showed no statistical association, with HRs of 0.93 and 1.18, respectively. In contrast, an increased HR of 9.49 (95% CI: 2.82-32.00) was observed in subjects having both diabetes and blood transfusion, suggesting a positive interaction between them on the risk of death from ischemic heart disease (p=0.02). To conclude: We identified a multiplicative effect of history of diabetes and blood transfusion on the risk of death from ischemic heart disease.
