A modelling approach is used to analyse the cost effectiveness of prescribing angiotensin converting enzyme (ACE) inhibitors, compared with standard practice, as first-line therapeutic agents in the treatment of heart failure in The Netherlands. Data concerning costs, incidence, prevalence and survival are used to construct an age-dependent semi-Markov-chain model. Two scenarios are compared. The first reflects the continuation of common practice. The second, containing assumptions made on the basis of results from randomised clinical trials, reflects the situation in which ACE inhibitors are given as first-line pharmacotherapy. Conditional on the estimates and assumptions made, it is shown that prescribing ACE inhibitors as first-line pharmacotherapy will improve survival by about 4% over the first 10 years, and will save about 17% in costs over the first 10 years. Sensitivity analysis shows the robustness of the conclusions to all major parameters.