Limited resources for health care means that techniques are required to aid the process of priority setting. This paper represents one of the first attempts to use discrete choice modelling (DCM) within the area of priority setting. It is shown how the technique can be used to estimate cost per unit of benefit ratios for competing clinical service developments. Integer programming is proposed as a method to be used, alongside DCM, to help policy makers select the optimal combination of clinical service developments within a given budget. The technique is also shown to be internally valid and internally consistent. It is argued that DCM is a potentially useful technique to be used within the area of priority setting more generally. However, further work is required to address methodological issues around the technique.