BACKGROUND: Diastolic dysfunction (DD) is highly prevalent and associated with increased morbidity and mortality, but its natural history remains poorly defined. OBJECTIVE: This cohort study sought to characterise the influence of clinical features, medical therapy and echocardiographic parameters on the progression of DD. METHODS: We identified 926 consecutive patients (aged 62 (14) years, 221 women) with DD and preserved systolic function. A repeat echocardiogram was performed in 199 patients > or =1 year after the baseline study (average 3.6 (1.4) years). Follow-up for 4.8 (2.5) years was 97% complete for the major endpoint of all-cause mortality. Cox regression analyses were performed to identify the associations of mortality. RESULTS: Over follow-up, 142 patients died and 22 were admitted with heart failure. The independent predictors of death were age, hyperlipidaemia, co-morbid disease and restrictive filling. The degree of diastolic dysfunction remained stable in 52%, deteriorated in 27% and improved in 21%. There was a greater use of medical therapy in those with stable or worsening diastolic function; when the protective effects of these agents were taken into account in a multivariate model, improvement in diastolic dysfunction was associated with a survival benefit. CONCLUSION: DD is associated with all-cause mortality, independent of the presence of a major co-morbidity. The degree of DD remains stable in about 50% of patients, the population whose diastolic function improves over time has a more favourable outcome.