Cost per life year gained was the measure which summarised the results. This figure was Dfl 4,179, with FDC. It is worth noting, however, that if treatment with FDC were restricted to patients at high risk of NSAID-induced ulcers, savings would occur instead of additional costs.
The main sensitivity results were showed that changes in percentages of complicated ulcers influenced the results more than changes in the percentage of endoscopic ulcers (because just 10% of those patients seek treatment). With respect to costs, the most sensitive parameters were the costs of diclofenac and FDC. Also, changing the treatment costs per complicated ulcer by +/- 20% changed the cost-effectiveness outcomes by 45%. The break-evenanalysis demonstrated that the relevant point for 'percentage of endoscopic ulcers treated ambulantly' was 25%, and, for the price difference between diclofenac and FDC it was Dfl 6.50 per month at the standard daily dosage.
The multi-way sensitivity analysis showed that for the present model 29% of the simulation outcomes were deemed acceptable (saving combined with lives gained) and 1% unacceptable (additional costs combined with lives lost), while the remaining 70% of outcomes had an acceptability rate dependent on the limit chosen by society on additional costs per additional survivor.