Natural history of idiopathic dilated cardiomyopathy: effect of referral bias and secular trend

Objectives. The current study was designed to determine the effect of secular trend and referral bias on the natural history of idiopathic dilated cardiomyopathy. Background. In a previous study of 104 patients with idiopathic dilated cardiomyopathy conducted in a referral population at the Mayo Clinic between 1960 and 1973, the 1- and 5-year mortality rates were 31% and 64%, respectively. A recent study of 40 patients with idiopathic dilated cardiomyopathy conducted in a. population-based cohort at the Mayo Clinic between 1975 find 1984 reported 1- and 5-year mortality rates of 5% and 20%, respectively. We hypothesized that improvements in diagnosis and therapy have occurred since the original referral cohort was described and that these improvements have altered the apparent natural history of the disease. We refer to this effect as secular trend. Alternatively, the presence of more advanced disease in the referral population (referral bias) may also contribute to the differences in survival. Methods. Two sequential referral populations with idiopathic dilated cardiomyopathy seen at the Mayo Clinic between 1976 and 81 (n = 85) and 1982 and 1987 (n = 137) were identified. Outcome was compared between these cohorts and the 1960-1973 referral cohort to examine the effect of secular trend. Outcomes were compared with that of the population-based cohort to examine the effect of referral bias. Results. Survival in the 1976-1981 referral cohort did not differ from that in the 1960-1973 referral cohort, suggesting little impact of secular trend during these time periods. Survival in the more recent 1982-1987 referral cohort was significantly better than that in the earlier referral cohorts, suggesting that improvements in diagnosis and treatment in the 1980s altered the natural history of idiopathic dilated cardiomyopathy. Survival in the 1982-1987 referral cohort was still worse than that of the population-based cohort, suggesting an effect of referral bias on studies of the natural history of idiopathic dilated cardiomyopathy. Conclusions. The current study demonstrates that secular trend and referral bias affect the apparent natural history of idiopathic dilated cardiomyopathy. Survival in referral patients with this disease is significantly better than previously described.