Multilevel modelling is a relatively new technique developed in the area of educational research. To illustrate the use of this technique in health economics, this paper estimates a multilevel logit model to examine the effect of competition on the behaviour of Australian general practitioners. The main hypothesis tested is that GPs in areas of high competition are more likely to recommend a follow up consultation compared to GPs in areas of low competition. The results suggest that competition influences the decision to recommend a follow up visit for one out of the five medical conditions analysed. The use of multilevel analysis represents a methodological improvement on previous models of GP behaviour. However, before multilevel analysis is more widely adopted it is argued that it should be more formally assessed against more standard and equivalent methods already used by economists, such as random effects panel data models.