Predicting the prevalence of a disease in a cohort at risk

A non-homogeneous Markov chain model is used to describe the evolution of a progressive chronic disease. Transition probabilities are estimated from data on coronary heart disease in Australia. By varying the transition probabilities, the effects of changes in medical management or in the natural history of the disease are examined. For reducing rates of fatal and non-fatal acute myocardial infarction a multifaceted strategy of primary and secondary prevention and improved acute care is found most effective. Primary prevention reducing the risk of early stages of the disease has the potential to yield the highest proportion of disease-free people. By contrast, intensive medical treatment of acute disease would reduce mortality but increase the total burden of disease in the population.