Aim: To test the hypothesis that general practitioners (GPs) with high prescribing levels of certain drugs will adopt new drugs belonging to the same therapeutic group faster than those with low prescribing levels. Methods: The adoption of four new drugs: esomeprazol, selective cyclo-oxygenase-2 inhibitors, new triptans, and angiotensin-II receptor blockers were analysed using population-based prescription data. We used the preference proportion (prescriptions for new rather than older alternatives for the same indication) to measure GPs' adoption rate. Annual prescribing volume and prevalence were used to measure previous prescribing of older drug alternatives. We modelled the preference proportion using multiple linear regression analysis and the prescribing of older drugs as independent variables. We controlled for the GPs' general prescribing level and weighted for practice size. In the first three analyses, we dichotomized data using the median, lower and upper quartile as cut-off point. Next, we grouped data into quartiles and finally, we used continuous data. Results: For esomeprazol and new triptans there was a higher preference for new drugs among 'high prescribers', but only when this term was defined as the upper quarter and the upper half of previous prescribing levels, respectively (mean difference in preference proportion: 10.2% (99% confidence interval = 1.3%, 19.1%) and 8.2% (0.2%, 16.2%)). For the remaining two drug classes the associations were weak and almost all statistically nonsignificant. Conclusion: There is no consistent association between GPs' level of drug prescribing and their adoption of new drugs of the same therapeutic group.