Method: Data from a Canadian multicentre controlled trial involving 968 post-treatment breast cancer patients who were randomly assigned to follow up by either their own FP or CC were used to populate a Markov model. Within the trial, patients were followed for up to 5 years during which data was collected on clinical outcomes (recurrences and deaths), utility values (from EQ5D) and patient and healthcare system costs: long term data were modelled based on extrapolation from trial data and literature review. The model estimated costs and quality adjusted life expectancy (QALE) by follow-up strategy over a twenty year period. Future costs and utilities were discounted at a rate of 5%. Monte Carlo simulation estimated the uncertainty surrounding the incremental cost effectiveness ratio (ICER). A Bayesian value of information analysis was conducted to estimate the expected value of perfect information both overall and for each of the sets of inputs to the model.

Results: The total cost of CC follow-up was CAN$16,280; whereas the cost of FP follow up was CAN$15,457. The quality adjusted life expectancy was 9.2 for both CC and FP follow up. For a value of $50 000 which is considered a common threshold for economic evaluations in Canada, the probability that family physician follow up is the most cost effective was 51% compared to 49% for CC. Based on the VOI analysis, the inputs to the model which contribute most to the uncertainty were trial utility values ($5310 per patient) and trial transition probabilities ($2098).

Conclusion: The randomized controlled trial found no significant differences between groups in terms of clinical outcomes and quality of life. This study demonstrated that follow up of patients who have completed adjuvant therapy for breast cancer by family physicians may be more cost effective than follow up by CC. However, given the great deal of uncertainty surrounding the ICER estimate, giving patient a choice between who should follow them up appears optimal.